Synthesis of tetrasaccharide analogues of the N-glycan substrate of beta-(1-->2)-N-acetylglucosaminyltransferase II using trisaccharide precursors and recombinant beta-(1-->2)-N-acetylglucosaminyltransferase I

Use of synthetic oligosaccharide substrate analogs to map the active sites of N-acetylglucosaminyltransferases I and II

Tables III and IV summarize substrate analog data presented in Tables I and II, respectively. Data for GlcNAc-T I shown in Tables I and III correlate very well with the crystal structure for GlcNAc-T I. This indicates that substitution of the various hydroxyl groups by hydrogen, [table: see text] O-methyl, and maybe even larger O-alkyl groups does not cause appreciable changes to either the overall conformation of the oligosaccharide or the binding mode, thus supporting this approach of chemical modification of oligosaccharide substrates for mapping of the binding site. There is as yet no crystal structure for GlcNAc-T II. These studies indicate both advantages and disadvantages of this approach for elucidating the catalytic and binding sites of an enzyme. Substrate analog data indicate which chemical groups in the substrate are essential for catalysis and binding and suggest the type of linkage involved (hydrogen bond donor or acceptor). However, no information has been obtained on the protein groups involved in these interactions. If a crystal structure is available, the substrate analog conclusions are primarily confirmatory. However, whether or not a crystal structure is available, this approach can be very helpful in the design of specific inhibitors.

Synthesis of paucimannose N-glycans by Caenorhabditis elegans requires prior actions of UDP-N-acetyl-D-glucosamine:alpha-3-D-mannoside beta1,2-N-acetylglucosaminyltransferase I, alpha3,6-mannosidase II and a specific membrane-bound beta-N-acetylglucosaminidase

We have previously reported three Caenorhabditis elegans genes ( gly-12, gly-13 and gly-14 ) encoding UDP- N -acetyl-D-glucosamine:alpha-3-D-mannoside beta1,2- N -acetylglucosaminyltransferase I (GnT I), an enzyme essential for hybrid and complex N-glycan synthesis. GLY-13 was shown to be the major GnT I in worms and to be the only GnT I cloned to date which can act on [Manalpha1,6(Manalpha1,3)Manalpha1,6](Manalpha1,3)Manbeta1, 4GlcNAcbeta1,4GlcNAc-R, but not on Manalpha1,6(Manalpha1,3)Manbeta1- O -R substrates. We now report the kinetic constants, bivalent-metal-ion requirements, and optimal pH, temperature and Mn(2+) concentration for this unusual enzyme. C. elegans glycoproteins are rich in oligomannose (Man(6-9)GlcNAc(2)) and 'paucimannose' Man(3-5)GlcNAc(2)(+/-Fuc) N-glycans, but contain only small amounts of complex and hybrid N-glycans. We show that the synthesis of paucimannose Man(3)GlcNAc(2) requires the prior actions of GnT I, alpha3,6-mannosidase II and a membrane-bound beta- N -acetylglucosaminidase similar to an enzyme previously reported in insects. The beta- N -acetylglucosaminidase removes terminal N -acetyl-D-glucosamine from the GlcNAcbeta1, 2Manalpha1,3Manbeta- arm of Manalpha1,6(GlcNAcbeta1,2Manalpha1,3) Manbeta1,4GlcNAcbeta1,4GlcNAc-R to produce paucimannose Man(3)GlcNAc(2) N-glycan. N -acetyl-D-glucosamine removal was inhibited by two N -acetylglucosaminidase inhibitors. Terminal GlcNAc was not released from [Manalpha1,6(Manalpha1,3)Manalpha 1,6] (GlcNAcbeta1,2Manalpha1,3)Manbeta1,4GlcNAcbeta1,4GlcNAc-R nor from the GlcNAcbeta1,2Manalpha1,6Manbeta- arm. These findings indicate that GLY-13 plays an important role in the synthesis of N-glycans by C. elegans and that therefore the worm should prove to be a suitable model for the study of the role of GnT I in nematode development.

Cloning and expression of a novel UDP-GlcNAc:alpha-D-mannoside beta1,2-N-acetylglucosaminyltransferase homologous to UDP-GlcNAc:alpha-3-D-mannoside beta1,2-N-acetylglucosaminyltransferase I

A TBLASTN search with human UDP-GlcNAc:alpha-3-d-mannoside beta-1,2-N-acetylglucosaminyltransferase I (GnT I; EC 2.4.1.101) as a probe identified human and mouse Unigenes encoding a protein similar to human GnT I (34% identity over 340 amino acids). The recombinant protein converted Man(alpha1-6)[Man(alpha1-3)]Man(beta1-)O-octyl to Man(alpha1-6)[GlcNAc(beta1-2)Man(alpha1-3)]Man(beta1-)O-octyl, the reaction catalysed by GnT I. The enzyme also added GlcNAc to Man(alpha1-6)[GlcNAc(beta1-2)Man(alpha1-3)]Man(beta1-)O-octyl (the substrate for beta-1,2-N-acetylglucosaminyltransferase II), Man(alpha1-)O-benzyl [with K(m) values of approximately 0.3 and >30 mM for UDP-GlcNAc and Man(alpha1-)O-benzyl respectively] and the glycopeptide CYA[Man(alpha1-)O-T]AV (K(m) approximately 12 mM). The product formed with Man(alpha1-)O-benzyl was identified as GlcNAc(beta1-2)Man(alpha1-)O-benzyl by proton NMR spectroscopy. The enzyme was named UDP-GlcNAc:alpha-d-mannoside beta-1,2-N-acetylglucosaminyltransferase I.2 (GnT I.2). The human gene mapped to chromosome 1. Northern-blot analysis showed a 3.3 kb message with a wide tissue distribution. The cDNA has a 1980 bp open reading frame encoding a 660 amino acid protein with a type-2 domain structure typical of glycosyltransferases. Man(beta1-)O-octyl, Man(beta1-)O-p-nitrophenyl and GlcNAc(beta1-2)Man(alpha1-6)[GlcNAc(beta1-2)Man(alpha1-3)]Man(beta1-4)GlcNAc(beta1-4)GlcNAc(beta1-)O-Asn were not acceptors, indicating that GnT I.2 is specific for alpha-linked terminal Man and does not have N-acetylglucosaminyltransferase III, IV, V, VII or VIII activities. CYA[Man(alpha1-)O-T]AV was between three and seven times more effective as an acceptor than the other substrates, suggesting that GnT I.2 may be responsible for the synthesis of the GlcNAc(beta1-2)Man(alpha1-)O-Ser/Thr moiety on alpha-dystroglycan and other O-mannosylated proteins.

Cloning and expression of Drosophila melanogaster UDP-GlcNAc:alpha-3-D-mannoside beta1,2-N-acetylglucosaminyltransferase I

A TBLASTN search of the Drosophila melanogaster expressed sequence tag (EST) database with the amino acid sequence of human UDP-N-acetylglucosamine:alpha-3-D-mannoside beta-1,2-N-acetylglucosaminyltransferase I (GnT I, EC 2.4.1.101) as probe yielded a clone (GM01211) with 56% identity over 36 carboxy-terminal amino acids. A 550 base pair (bp) probe derived from the EST clone was used to screen a Drosophila cDNA library in lambda-ZAP II and two cDNAs lacking a start ATG codon were obtained. 5'-Rapid amplification of cDNA ends (5'-RACE) yielded a 2828 bp cDNA containing a full-length 1368 bp open reading frame encoding a 456 amino acid protein with putative N-terminal cytoplasmic (5 residues) and hydrophobic transmembrane (20 residues) domains. The protein showed 52% amino acid sequence identity to human GnT I. This cDNA, truncated to remove the N-terminal hydrophobic domain, was expressed in the baculovirus/Sf9 system as a secreted protein containing an N-terminal (His)6 tag. Protein purified by adsorption to and elution from nickel beads converted Man alpha1-6(Man alpha1-3)Man beta-octyl (M3-octyl) to Man alpha1-6(GlcNAc beta1-2Man alpha1-3)Man beta-octyl. The Km values (0.7 and 0.03 mM for M3-octyl and UDP-GlcNAc respectively), temperature optimum (37 degrees C), pH optimum (pH 5 to 6) and divalent cation requirements (Mn > Fe, Mg, Ni > Ba, Ca, Cd, Cu) were similar to mammalian GnT I. TBLASTN searches of the Berkeley Drosophila Genome Project database with the Drosophila GnT I cDNA sequence as probe allowed localization of the gene to chromosomal region 2R; 57A9. Comparison of the cDNA and genomic DNA sequences allowed the assignment of seven exons and six introns; all introns showed GT-AG splice site consensus sequences. This is the first insect GnT I gene to be cloned and expressed.

The joys of HexNAc. The synthesis and function of N- and O-glycan branches

This review covers discoveries made over the past 30-35 years that were important to our understanding of the synthetic pathway required for initiation of the antennae or branches on complex N-glycans and O-glycans. The review deals primarily with the author's contributions but the relevant work of other laboratories is also discussed. The focus of the review is almost entirely on the glycosyltransferases involved in the process. The following topics are discussed. (1) The localization of the synthesis of complex N-glycan antennae to the Golgi apparatus. (2) The "evolutionary boundary" at the stage in N-glycan processing where there is a change from oligomannose to complex N-glycans; this switch correlates with the appearance of multicellular organisms. (3) The discovery of the three enzymes which play a key role in this switch, N-acetylglucosaminyltransferases I and II and mannosidase II. (4) The "yellow brick road" which leads from oligomannose to highly branched complex N-glycans with emphasis on the enzymes involved in the process and the factors which control the routes of synthesis. (5) A short discussion of the characteristics of the enzymes involved and of the genes that encode them. (6) The role of complex N-glycans in mammalian and Caenorhabditis elegans development. (7) The crystal structure of N-acetylglucosaminyltransferase I. (8) The discovery of the enzymes which synthesize O-glycan cores 1, 2, 3 and 4 and their elongation.

Expression of three Caenorhabditis elegans N-acetylglucosaminyltransferase I genes during development

UDP-N-acetylglucosamine:alpha-3-D-mannoside beta-1, 2-N-acetylglucosaminyltransferase I (GnT I) is a key enzyme in the synthesis of Asn-linked complex and hybrid glycans. Studies on mice with a null mutation in the GnT I gene have indicated that N-glycans play critical roles in mammalian morphogenesis. This paper presents studies on N-glycans during the development of the nematode Caenorhabditis elegans. We have cloned cDNAs for three predicted C. elegans genes homologous to mammalian GnT I (designated gly-12, gly-13, and gly-14). All three cDNAs encode proteins (467, 449, and 437 amino acids, respectively) with the domain structure typical of previously cloned Golgi-type glycosyltransferases. Expression in both insect cells and transgenic worms showed that gly-12 and gly-14, but not gly-13, encode active GnT I. All three genes were expressed throughout worm development (embryo, larval stages L1-L4, and adult worms). The gly-12 and gly-13 promoters were expressed from embryogenesis to adulthood in many tissues. The gly-14 promoter was expressed only in gut cells from L1 to adult developmental stages. Transgenic worms that overexpress any one of the three genes show no obvious phenotypic defects. The data indicate that C. elegans is a suitable model for further study of the role of complex N-glycans in development.

Activity of UDP-GlcNAc:GlcNAc beta 1-->6(GlcNAc beta 1-->2) Man alpha 1-->R[GlcNAc to Man] beta 1-->4N-acetylglucosaminyltransferase VI (GnT VI) from the ovaries of Oryzias latipes (Medaka fish)

UDP-GlcNAc:GlcNAc beta 1-->(GlcNAc beta 1-->2)Man alpha 1-R[GlcNAc to Man] beta 1-->4N-acetylglucosaminyltransferase VI (GnT VI) activity was shown to be present in crude homogenates of Medaka fish (Oryzias latipes) ovaries using UDP-[14C]GlcNAc and synthetic GlcNAc beta 1-->6 (GlcNAc beta 1-->2)Man alpha 1-->6Glc beta 1-->octyl as substrates. Characterization of this activity showed a pH optimum at about pH 7.0 and an absolute requirement for divalent cations. The optimum concentration of Mn2+ was at about 25 mM. This finding is the first report on GnT VI activity in fish; the enzyme has previously been described only in avian tissues.

Synthetic substrate analogues for UDP-GlcNAc: Man alpha 1-3R beta 1-2-N-acetylglucosaminyltransferase I. Substrate specificity and inhibitors for the enzyme

UDP-GlcNAc:Man alpha 1-3R beta 1-2-N-acetylglucosaminyltransferase I (GlcNAc-T I; EC 2.4.1.101) catalyses the conversion of [Man alpha 1-6(Man alpha 1-3)Man alpha 1-6][Man alpha 1-3]Man beta-O-R to [Man alpha 1-6(Man alpha 1-3)Man alpha 1-6] [GlcNAc beta 1-2Man alpha 1-3]Man beta-O-R (R = 1-4GlcNAc beta 1-4GlcNAc- Asn-X) and thereby controls the conversion of oligomannose to complex and hybrid asparagine-linked glycans (N-glycans). GlcNAc-T I also catalyses the conversion of Man alpha 1-6(Man alpha 1-3)Man beta-O-octyl to Man alpha 1-6(GlcNAc beta 1-2Man alpha 1-3)Man beta-O-octyl. We have therefore tested a series of synthetic analogues of Man"alpha 1-6(Man'alpha 1-3)Man beta-O-octyl as substrates and inhibitors for rat liver GlcNAc-T I. The 2"-deoxy and the 3"-, 4"- and 6"-O-methyl derivatives are all good substrates confirming previous observations that the hydroxyl groups of the Man"alpha 1-6 residue do not play major roles in the binding of substrate to enzyme. In contrasts, all four hydroxyl groups on the Man'alpha 1-3 residue are essential since the corresponding deoxy derivatives either do not bind (2'- and 3'-deoxy) or bind very poorly (4'- and 6'-deoxy) to the enzyme. The 2'- and 3'-O-methyl derivatives also do not bind to the enzyme. However, the 4'-O-methyl derivative is a substrate (KM = 2.6 mM) and the 6'-O-methyl compound is a competitive inhibitor (Ki = 0.76 mM). We have therefore synthesized various 4'- and 6'-O-alkyl derivatives, some with reactive groups attached to an O-pentyl spacer, and tested these compounds as reversible and irreversible inhibitors of GlcNAc-T I. The 6'-O-(5-iodoacetamido-pentyl) compound is a specific time dependent inhibitor of the enzyme. Four other 6'-O-alkyl compounds showed competitive inhibition while the remaining compounds showed little or no binding indicating that the electronic properties of the attached O-pentyl groups influence binding.

Synthesis of uridine-5-propylamine derivatives and their use in affinity chromatography of N-acetylglucosaminyltransferases I and II

The C-5 substituted uridine derivatives UDP-5-propylamine (7) and UDP-GlcNAc-5-propylamine (8) were synthesized in good yields by Heck alkylation of the 5-mercuriuridines, followed by hydrogenation. The products were characterized by 1H and 13C NMR spectroscopy, electrospray mass spectrometry and UV spectrophotometry. The amines are of interest for the preparation of affinity probes for glycosyltransferases. The benzoylbenzamides of 7 and 8 show strong competitive inhibition of N-acetylglucosaminyltransferase I and II with Ki values ranging from 30 to 100 microM (without irradiation) and may be useful as active site-directed photoaffinity labels. A conjugate of 8 and Sepharose was used for affinity chromatographic purification of N-acetylglucosaminyltransferases I and II. The results indicate that this affinity gel is a stable alternative to the commonly used but unstable UDP-GlcNAc-5-Hg-thiopropyl conjugate.

Synthesis of pentasaccharide analogues of the N-glycan substrates of N-acetylglucosaminyltransferases III, IV and V using tetrasaccharide precursors and recombinant beta-(1-->2)-N-acetylglucosaminyltransferase II

Recombinant human UDP-GlcNAc: alpha-Man-(1-->6)R beta-(1-->2)-N-acetylglucosaminyltransferase II (EC 2.4.1.143, GlcNAc-T II) was produced in the Sf9 insect cell/baculovirus expression system as a fusion protein with a (His)6 tag and partially purified by affinity chromatography on a metal chelating column. The partially purified enzyme was used to catalyze the transfer of GlcNAc from UDP-GlcNAc to R-alpha-Man(1-->6)(beta-GlcNAc(1-->2)alpha-Man(1-->3))beta-Man-O-octyl to form beta-GlcNAc(1-->2)R-alpha-Man(1-->6)(beta-GlcNAc(1-->2)alpha- Man(1-->3))beta-Man-O-octyl where there is either no modification of the alpha-Man(1-->6) residue (7), or where R is 3-deoxy (8), 4-deoxy (9) or 6-deoxy (10). The yields ranged from 64-80%. Products were characterized by 1H and 13C nuclear magnetic resonance spectroscopy and fast atom bombardment mass spectrometry. Compounds 7-10 are pentasaccharide analogues of the biantennary N-glycan substrates of N-acetylglucosaminyltransferases III, IV and V.

The human UDP-N-acetylglucosamine: alpha-6-D-mannoside-beta-1,2- N-acetylglucosaminyltransferase II gene (MGAT2). Cloning of genomic DNA, localization to chromosome 14q21, expression in insect cells and purification of the recombinant protein

UDP-GlcNAc:alpha-6-D-mannoside [GlcNAc to Man alpha 1-6] beta-1,2-N-acetylglucosaminyltransferase II (GlcNAc-T II, EC 2.4.1.143) is a Golgi enzyme catalyzing an essential step in the conversion of oligomannose to complex N-glycans. A 1.2-kb probe from a rat liver cDNA encoding GlcNAc-T II was used to screen a human genomic DNA library in lambda EMBL3. Southern analysis of restriction endonuclease digests of positive phage clones identified two hybridizing fragments (3.0 and 3.5 kb) which were subcloned into pBlueScript. The inserts of the resulting plasmids (pHG30 and pHG36) are over-lapping clones containing 5.5 kb of genomic DNA. The pHG30 insert (3.0 kb) contains a 1341-bp open reading frame encoding a 447-amino-acid protein, 250 bp of G + C-rich 5'-upstream sequence and 1.4 kb of 3'-downstream sequence. The pHG36 insert (3.5 kb) contains 2.75 kb of 5'-upstream sequence and 750 bp of the 5'-end of the open reading frame. The protein sequence showed the domain structure typical of all previously cloned glycosyltransferases, i.e. a short 9-residue putative cytoplasmic N-terminal domain, a 20-residue hydrophobic non-cleavable putative signal-anchor domain and a 418-residue C-terminal catalytic domain. Northern analysis of human tissues showed a major message at 3 kb and minor signals at 2 and 4.5 kb. There is no sequence similarity to any previously cloned glycosyltransferases including human UDP-GlcNAc:alpha-3-D-mannoside [GlcNAc to Man alpha 1-3] beta-1,2-N-acetylglucosaminyltransferase I (GlcNAc-T I) which has 445 amino acids with a 418-residue C-terminal catalytic domain. The human GlcNAc-T I and II genes (MGAT1 and MGAT2) map to chromosome bands 5q35 and 14q21, respectively, by fluorescence in situ hybridization. The entire coding regions of human GlcNAc-T I and II are each on a single exon. There is 92% identity between the amino acid sequences of the catalytic domains of human and rat GlcNAc-T II. Southern analysis of restriction enzyme digests of human genomic DNA indicates that there is only a single copy of the MGAT2 gene. The full-length coding region of GlcNAc-T II has been expressed in the baculovirus/Sf9 insect cell system, the recombinant enzyme has been purified to near homogeneity with a specific activity of about 20 mumol.min-1.mg-1 and the product synthesized by the recombinant enzyme has been identified by high-resolution 1H-NMR spectroscopy and mass spectrometry.

Carbohydrate-deficient glycoprotein syndrome type II. An autosomal recessive N-acetylglucosaminyltransferase II deficiency different from typical hereditary erythroblastic multinuclearity, with a positive acidified-serum lysis test (HEMPAS)

Carbohydrate-deficient glycoprotein syndromes (CDGS) are a family of multisystemic congenital diseases resulting in underglycosylated glycoproteins, suggesting defective N-glycan assembly. Fibroblast extracts from two patients with a recently described variant of this disease (CDGS type II) have previously been shown to have over 98% reduced activity of UDP-GlcNAc:alpha-6-D-mannoside beta-1,2-N-acetylglucosaminyltransferase II [GlcNAc-TII; Jaeken, J., Schachter, H., Carchon, H., De Cock, P., Coddeville, B. & Spik, G. (1994) Arch. Dis. Childhood 71, 123-127]. We show in this paper that mononuclear cell extracts from one of these CDGS type-II patients have no detectable GlcNAc-TII activity and that similar extracts from 12 blood relatives of the patient, including his father, mother and brother, have GlcNAc-TII levels 32-67% that of normal levels (average 50.1% +/- 10.7% SD), consistent with an autosomal recessive disease. The poly(N-acetyllactosamine) content of erythrocyte membrane glycoproteins bands 3 and 4.5 of this CDGS patient were estimated, by tomato lectin blotting, to be reduced by 50% relative to samples obtained from blood relatives and normal controls. Similar to patients with hereditary erythroblastic multinuclearity with a positive acidified-serum lysis test (HEMPAS), erythrocyte membrane glycoproteins in the CDGS patient have increased reactivities with concanavalin A, demonstrating the presence of hybrid or oligomannose carbohydrate structures. However, bands 3 and 4.5 in HEMPAS erythrocytes have almost complete lack of poly(N-acetyllactosamine). Furthermore, CDGS type-II patients have a totally different clinical presentation and their erythrocytes do not show the serology typical of HEMPAS, suggesting that the genetic lesions responsible for these two diseases are possibly different.