Exploring the glycan repertoire of genetically modified mice by isolation and profiling of the major glycan classes and nano-NMR analysis of glycan mixtures.

The production of mice with genetic alterations in glycosyltransferases has highlighted the need to isolate and study complex mixtures of the major classes of oligosaccharides (glycans) from intact tissues. We have found that nano-NMR spectroscopy of whole mixtures of N- and O-glycans can complement HPLC profiling methods for elucidating structural details. Working toward obtaining such glycan mixtures from mouse tissues, we decided to develop an approach to isolate not only N- and O-glycans, but also to separate out glycosphingolipids, glycosaminoglycans and glycosylphosphatidylinositol anchors. We describe here a comprehensive Glycan Isolation Protocol that is based primarily upon the physicochemical characteristics of the molecules, and requires only commonly available reagents and equipment. Using radiolabeled internal tracers, we show that recovery of each major class of glycans is as good or better than with conventional approaches for isolating individual classes, and that cross-contamination is minimal. The recovered glycans are of sufficient purity to provide a "glycoprofile" of a cell type or tissue. We applied this approach to compare the N- and O-glycans from wild type mouse tissues with those from mice genetically deficient in glycosyltransferases. N- and O-glycan mixtures from organs of mice deficient in ST6Gal-I (CMP-Sia:Galbeta1-4GlcNAc alpha2-6 sialyltransferase) were studied by the nano-NMR spectroscopy approach, showing no detectable alpha2-6-linked sialic acids. Thus, ST6Gal-I is likely responsible for generating most or all of these residues in normal mice. Similar studies indicate that this linkage is very rare in ganglioside glycans, even in wild-type tissues. In mice deficient in GalNAcT-8 (UDP-GalNAc:polypeptide O-Ser/Thr GalNAc transferase 8), HPLC profiling indicates that O-glycans persist in the thymus in large amounts, without a major change in overall profile, suggesting that other enzymes can synthesize the GalNAc-O-Ser/Thr linkage in this tissue. These results demonstrate the applicability of nano-NMR spectroscopy to complex glycan mixtures, as well as the versatility of the Glycan Isolation Protocol, which makes possible the concurrent examination of multiple glycan classes from intact vertebrate tissues.

Trans-sialidase from Trypanosoma cruzi catalyzes sialoside hydrolysis with retention of configuration.

The trans -sialidase from Trypanosoma cruzi is a member of the sialidase superfamily that functions as a sialidase in the absence of a carbohydrate acceptor. We have used(1)H nuclear magnetic resonance (NMR) spectroscopy to investigate the stereospecificity of the hydrolysis of two substrates, namely, 4-methyl-umbelliferyl- N -acetylneur-aminic acid and alpha(2-3)-sialyllactose, catalyzed by a recombinant T.cruzi trans -sialidase. We demonstrate that, in aqueous solution, the thermodynamically less stable alpha-form of N -acetylneuraminic acid is the initial product of the hydrolysis; subsequent mutarotation leads eventually to an equilibrium mixture of the alpha and beta forms, in molar ratio 8:92. In a mixed water/methanol solution, the hydrolysis reaction produces also the alpha-methyl sialoside but not its beta-methyl counterpart. We also show that 4-methyl-umbelliferyl- N -acetylneuraminic acid is a significantly better substrate for the sialidase than alpha(2-3)-sialyllactose. Prolonged incubation of alpha(2-3)-sialyllactose with an excess of trans -sialidase produced a trace of 2-deoxy-2,3-didehydro- N -acetylneuraminic acid, as identified by NMR spectroscopy and by gas liquid chromatography/mass spectro-metry. In conclusion, this study shows that the stereo-selectivity of the sialidase activity of T.cruzi trans -sialidase is identical to that of bacterial, viral, and mammalian sialidases, suggesting a similar active-site architecture.

Molecular characterization and immunohistochemical localization of IV(4)GalNAcGgOse(4)Cer: a naturally occurring novel neutral glycosphingolipid in bovine brain.

A pair of novel neutral glycosphingolipids (Ngsls) has been identified in bovine brain. Their mobilities on thin layer chromatography were slightly different from a standard pentaglycosylceramide (nLcOse(5)Cer from bovine erythrocytes). The compounds were purified to homogeneity by column chromatography. Their fatty acid and base compositions, their monosaccharide compositions and sugar linkage positions were determined by gas-liquid chromato-graphy/mass spectrometry. Carbohydrate sequence analy-sis by(1)H NMR spectroscopy and stepwise exoglyco-sidase digestion indicated the following pentaglycosyl structure for the oligosaccharide moiety of both Ngsls: GalNAcbeta1-4Galbeta1-3GalNAcbeta1-4Galbeta1-4Gl c. The two Ngsls (abbreviated as IV(4)GalNAcGgOse(4)Cer or GalNAc-GA1), differ in their ceramide compositions, having d18:0 and d18:1 sphingosine as their long chain bases. A monospecific polyclonal anti-GalNAc-GA1 antibody, prepared in rabbit and purified by affinity chromatography, stained the neurons of cerebral cortex and cerebellum including Purkinje cells in adult rat brain, indicating that the novel GalNAc-GA1 is associated with cerebellar and other neurons in vertebrate central nervous system.

Accurate and precise measurement of heteronuclear long-range couplings by a gradient-enhanced two-dimensional multiple-bond correlation experiment.

We propose a phase-sensitive gradient-enhanced two-dimensional heteronuclear multiple-bond correlation (psge-2D HMBC) experiment for speedy, accurate, and precise measurement of 2JCH and 3JCH. The experiment does not suppress one-bond correlations. Rather, the value of a desired long-range JCH is obtained from the pertinent cross-peak pattern in the HMBC spectrum, using the corresponding 1JCH correlation pattern as reference. The application of the proposed experiment is illustrated for the trisaccharide raffinose.

A 1H NMR spectroscopic approach to the unambiguous determination of glycosyl linkage positions in oligosaccharides.

A sensitive (< 1 mg) and nondestructive method was devised for the unambiguous determination of the glycosyl linkage positions in oligosaccharides using 1H NMR spectroscopy. The technique is based on the absence of a hydroxyl group on the carbon atom that participates in the glycosidic linkage. The "missing" hydroxyl group is identified by recording 1H NMR spectra of the oligosaccharide in H2O and D2O. The method is illustrated by determining the glycosyl linkage positions of maltose, Glc alpha 1-4Glc, and a branched mannopentaose, Man alpha 1-6(Man alpha 1-3)Man alpha 1-6(Man alpha 1-3)Man.

Structure of the capsular polysaccharide of Clostridium perfringens Hobbs 5 as determined by NMR spectroscopy.

The complete primary structure of the capsular polysaccharide of Clostridium perfringens Hobbs 5, an anaerobic bacterium implicated in food poisoning, was determined. The polysaccharide was isolated from C. perfringens Hobbs 5 cells, after deproteination, by ethanol precipitation and by ion-exchange chromatography. The polysaccharide was comprised of glucose, galactose, mannose, N-acetylglucosamine, N-acetylgalactosamine, and glucuronic acid, in equimolar ratios. Sequence and linkage assignments of the glycosyl residues were obtained by NMR spectroscopy, specifically by the combination of two-dimensional homonuclear TOCSY and NOESY experiments and heteronuclear (1H, 13C) multiple-quantum coherence (HMQC, HMQC-COSY, HMQC-TOCSY and HMBC) experiments. Thus, the envelope polysaccharide of C. perfringens Hobbs 5 was found to be a polymer composed of a hexasaccharide repeating unit with the following structure: [formula: see text] This structure is novel among bacterial cell-surface polysaccharides, and it is the first of many serotypically distinct capsular polysaccharides of C. perfringens to be described.

Molecular motions of a glycopeptide from human serum transferrin studied by 13C nuclear magnetic resonance.

The molecular motions of a 21-amino-acid glycopeptide (Gp21) containing multiple glycoforms of an N-linked diantennary oligosaccharide were studied by two-dimensional 1H-detected 13C relaxation measurements at natural abundance. Gp21 was derived from human serum transferrin, its amino acid sequence is QQQHLFGSNVTDCSGNFCLFR, and its N-glycan structure is [Formula: see text] The measured longitudinal and transverse relaxation rate constants and the nuclear Overhauser enhancements for the methine carbons of Gp21 were analyzed by using the model-free approach to obtain information about the internal motions in the molecule. The calculated order parameters S2 of the alpha carbons in both NH2- and COOH-terminal segments of the peptide are smaller than those in the interior segment of the Gp21 peptide moiety, implying that the internal motions in the terminal segments are less restricted than in the interior segment. The average S2 value is 0.72-0.91 for the glycosyl residues in the pentasaccharide core of Gp21, 0.58-0.59 for the interior GlcNAc-5,5' residues in the two branches, and 0.35-0.51 for the terminal GlcNAc-5, Gal-6,6', and NeuAcN,N' residues in the two branches, indicating that the internal motions in the glycan core are more restricted than in the two branches.

Complete 1H and 13C resonance assignments of a 21-amino acid glycopeptide prepared from human serum transferrin.

A 21-amino acid glycopeptide (Gp21) was isolated and purified in multi-milligram yields from commercially available human serum transferrin (HSTF) by a combination of tryptic digestion, Con A affinity chromatography, and reverse phase HPLC. The peptide chain of Gp21 contains a single N-glycosylation site to which a diantennary oligosaccharide is attached. The amino acid sequence and the glycan primary structure of Gp21 have been verified by peptide sequencing, electrospray mass spectrometry, and one-dimensional 1H NMR spectroscopy. Different glycoforms were found for the glycan of Gp21 derived from two different batches of commercial HSTF. These glycoforms differ from one another in the number of NeuAc residues (ranging from 0 to 2) and/or the number of Gal residues (ranging from 1 to 2). As for the monogalacto species, in the two-dimensional nuclear Overhauser effect (NOE) spectrum of Gp21, interglycosidic NOEs were observed between Man4 in the alpha (1-->3) branch and the terminal GlcNAc beta (1-->2) residue. No interglycosidic NOE was observed between Man4' in the alpha (1-->6) branch and the terminal GlcNAc residue. These observations indicate that the terminal GlcNAc residue in the minor glycoforms of Gp21 is exclusively located in the alpha (1-->3) branch of the Gp21 glycan. The occurrence of such a carbohydrate structure in HSTF has not been reported before. The 1H and 13C NMR spectra of Gp21 have been completely assigned by two-dimensional homonuclear and heteronuclear spectroscopy. The close similarity of the 1H and 13C chemical shift values for the Gp21 glycan with the respective values for the peptide-free diantennary oligosaccharide (Wieruszeski et al., Glycoconjugate J., 6 (1989) 183-194) indicates that the 1H and 13C chemical shifts of the diantennary oligosaccharide are not perturbed by the presence of the Gp21 peptide fragment. The complete 1H and 13C resonance assignments and the full characterization of the primary structure of Gp21 will permit us to study the conformation and dynamics of the N-linked diantennary oligosaccharides while covalently attached to a polypeptide fragment.

Molecular cloning of a developmentally regulated N-acetylgalactosamine alpha2,6-sialyltransferase specific for sialylated glycoconjugates.

A cDNA encoding a novel sialyltransferase has been isolated employing the polymerase chain reaction using degenerate primers to conserved regions of the sialylmotif that is present in all eukaryotic members of the sialyltransferase gene family examined to date. The cDNA sequence revealed an open reading frame coding for 305 amino acids, making it the shortest sialyltransferase cloned to date. This open reading frame predicts all the characteristic structural features of other sialyltransferases including a type II membrane protein topology and both sialylmotifs, one centrally located and the second in the carboxyl-terminal portion of the cDNA. When compared with all other sialyltransferase cDNAs, the predicted amino acid sequence displays the lowest homology in the sialyltransferase gene family. Northern analysis shows this sialyltransferase to be developmentally regulated in brain with expression persisting through adulthood in spleen, kidney, and lung. Stable transfection of the full-length cDNA in the human kidney carcinoma cell line 293 produced an active sialyltransferase with marked specificity for the sialoside, Neu5Ac-alpha2,3Gal-beta1,3GalNAc and glycoconjugates carrying the same sequence such as G(M1b) and fetuin. The disialylated tetrasaccharide formed by reacting the sialyltransferase with the aforementioned sialoside was analyzed by one- and two-dimensional 1H and 13C NMR spectroscopy and was shown to be the Neu5Ac-alpha2,3Gal-beta1,3(Neu5Ac-alpha2,6)GalNAc sialoside. This indicates that the enzyme is a GalNAc alpha-2,6-sialyltransferase. Since two other ST6GalNAc sialyltransferase cDNAs have been isolated, this sialyltransferase has been designated ST6GalNAc III. Of these three, ST6GalNAc III displays the most restricted acceptor specificity and is the only sialyltransferase cloned to date capable of forming the developmentally regulated ganglioside G(D1alpha) from G(M1b).

Chemical and immunological characterization of galactosyl-beta 1-3-globoside in bovine, human, and rat brain.

Our studies of bovine brain neutral glycosphingolipids (Ngsls) have revealed the presence of several short-chain (containing -CHO 1-4) and previously uncharacterized long-chain (-CHO > 4-5) Ngsls. We reported the structural characterization of brain GgOse4Cer (GA1) and have now purified another brain Ngsl to homogeneity. The purified Ngsl migrated close to standard GgOse4Cer and nLcOse5Cer on a TLC plate employing two different solvent systems. The carbohydrate molar composition indicated the presence of Gal/Glc/GalNAc in a ratio of 2.8:1.0:0.9. Five permethylated alditol acetate peaks were characterized as 2,3,4,6-OMe4Gal, 2,4,6-OMe3Gal, 2,3,6-OMe3Gal, 2,3,6-OMe3Glc, and 4,6-OMe2GalNAcMe by gas chromatography-mass spectrometry. The anomeric carbohydrate sequence has been determined by specific exoglycosidase digestion. Six-hundred megahertz 1H NMR spectroscopy of the oligosaccharide released by ceramide glycanase hydrolysis confirmed the structure of the Ngsl as Gal beta 1-3GalNAc beta 1-3Gal alpha 1-4Gal beta 1-4Glc beta 1-1Cer or IV3GalGbOse4Cer. Using the immunooverlay technique with anti-stage-specific embryonic antigen 3 antibody, it was found in bovine, rat, and normal adult human brain and bovine myelin, but not in human or rat myelin.

Evidence for a transient interresidue hydrogen bond in sucrose in aqueous solution obtained by rotating-frame exchange NMR spectroscopy under supercooled conditions.

We have used the hydroxyl protons of sucrose dissolved in supercooled water as NMR probes for the detection of intramolecular hydrogen bonds. Neither the measured OH temperature shift coefficients, 3JHCOH scalar couplings, isotope-induced 13C chemical shifts (delta delta COH/COD), nor OH exchange rates allowed us to single out any hydroxyl group in sucrose with characteristics indicative of involvement in strong hydrogen bonding. However, two-dimensional rotating-frame exchange (ROESY) spectroscopy revealed a direct exchange between the glucosyl OH2 and fructosyl OH1 protons. We conclude that an O2g::H::O1f interresidue hydrogen bond transiently exists in sucrose in aqueous solution. ROESY of sugars in supercooled water is proposed as a novel method to detect this type of weak hydrogen bonding.

Molecular mass and carbohydrate structure of prostate specific antigen: studies for establishment of an international PSA standard.

Despite the widely accepted use of prostate specific antigen (PSA) as a marker of prostate cancer, this molecule has not yet been completely characterized. Past studies have well established, however, using both amino acid and cDNA sequencing techniques, that PSA contains 237 amino acids, with a molecular mass of 26,079 Da for the peptide moiety of the molecule. The present study reports analysis of this protein by ion spray mass spectrometry (ISMS) and analysis of its carbohydrate moiety by NMR spectroscopy. The predominant PSA molecular species detected by ISMS was at relative molecular mass (M(r)) of 28,430, indicating that PSA contains a carbohydrate residue of M(r) 2,351, for a total percentage of carbohydrate of 8.3%. Analysis of PSA by SDS-PAGE, however, showed a M(r) of 32,000 to 33,000, suggesting an overestimation of the molecular weight by the latter technique. The complete primary structure of the PSA carbohydrate chain was determined by NMR spectroscopy in combination with carbohydrate composition analysis. The experimentally determined carbohydrate content of PSA confirms that only one N-glycosylation site is occupied in the protein. The proposed carbohydrate structure is a diantennary N-linked oligosaccharide of the N-acetyllactosamine type, with a sialic acid group at the end of each of the two branches. In addition, our data indicate that approximately 70% of the PSA molecules contain a fucose group in the core chitobiose moiety. The calculated molecular weight of this carbohydrate structure (M(r) 2,351.8) is in excellent agreement with the predicted molecular weight of the carbohydrate group, based on the M(r) 28,430 for PSA measured by ion spray mass spectrometry and M(r) 26,079 calculated from the consensus sequence for the peptide portion of the molecule. ISMS of PSA is thus proposed as a convenient and reliable method of quality control, an indispensible step towards international standardization of this very important tumor marker for detection and monitoring of prostatic diseases, especially prostate cancer.

Isolation and structural characterization of adhesin polysaccharide receptors.

The procedure for the purification of the adhesin polysaccharide receptor and its hexasaccharide repeating unit from whole S. oralis ATCC 55229 by chemical, enzymatic, and chromatographic techniques has been described. Chemical, chromatographic, and mass spectrometric procedures allow preliminary structural characterization of the hexasaccharide repeating unit and polysaccharide. The structural characterizations of the hexasaccharide and polysaccharide are completed using several 1D and 2D NMR techniques. Identification of the anomeric 1H and 13C signals of the glycosyl residues permits, by virtue of their chemical shifts and coupling constants (3JHH and 1JCH), the determination of the configurations of the glycosidic linkages. The HMBC connectivities permit the establishment of the hexasaccharide sequence as Rhap alpha(1-->2)Rhap alpha(1-->3)Galp alpha(1-->3)Galp beta(1-->4)Glcp beta(1-->3)Gal. The 1H NMR chemical shifts of the polysaccharide, as determined by the combination of COSY and TOCSY experiments, and the observed interglycosidic NOESY cross-peaks reveal the structure of the polysaccharide to be [formula: see text] where the position of the glycerol (Gro) phosphate moiety has been determined by [1H, 31P] NMR spectroscopy.

A revised structure for the disialosyl globo-series gangliosides of human erythrocytes and chicken skeletal muscle.

Disialosyl globo-series gangliosides have previously been isolated from chicken skeletal muscle (E. L. Hogan, R. D. Happel, and J.-L. Chien (1982) Adv. Exp. Med. Biol. 152, 273-278; S. Dasgupta, J.-L. Chien, E. L. Hogan, and H. van Halbeek (1991) J. Lipid Res. 32, 499-506) and human erythrocytes (S. K. Kundu, B. E. Samuelsson, I. Pascher, and D. Marcus (1983) J. Biol. Chem. 258, 13857-13866). In both cases, the structure of this ganglioside was proposed to be NeuAc alpha 2-->3(NeuAc alpha 2-->6)Gal beta 1-->3GalNAc beta 1-->3Gal alpha 1-->Gal alpha 1-->4Gal beta 1-->1Cer (V3NeuAcV6NeuAcGb5Cer). We have reinvestigated the human erythrocyte antigen and now propose an alternative structure differing in the location of the NeuAc alpha 2-->6 residue: NeuAc alpha 2-->3Gal beta 1-->3 (NeuAc alpha 2-->6)GalNAc beta 1-->3Gal alpha 1-->4Gal beta 1-->4Glc beta 1-->1 Cer (V3NeuAcIV6NeuAcGb5Cer). This novel structure is supported by results of 1H-NMR spectroscopy, negative ion fast atom bombardment mass spectrometry, and methylation linkage analysis with capillary gas chromatography--mass spectrometry in both electron impact and chemical ionization modes. Furthermore, based on new results from negative ion fast atom bombardment mass spectrometry and linkage analysis, we propose that the chicken skeletal muscle antigen also has this revised structure, differing only in ceramide composition. The terminal tetrasaccharide of these gangliosides is identical to that of GD1 alpha, NeuAc alpha 2-->3Gal beta 1-->3(NeuAc alpha 2-->6)GalNAc beta 1-->4Gal beta 1-->4Glc beta 1-->1 Cer(IV3NeuAcIII6NeuAcGg4Cer), previously identified in a rat ascites hepatoma cell line (T. Taki, Y. Hirabayashi, H. Ishikawa, S. Ando, K. Kon, Y. Tanaka, and M. Matsumoto (1986) J. Biol. Chem. 261, 3075-3078) and a murine lymphoma cell line with low metastatic potential (K. Murayama, S. B. Levery, V. Schirrmacher, and S. Hakomori (1986) Cancer Res. 46, 1395-1402), although they appear to be immunologically distinct.

Sulfate detection in glycoprotein-derived oligosaccharides by artificial neural network analysis of Fourier-transform infrared spectra.

We report the use of an artificial neural network to analyze the fingerprint region of Fourier-transform infrared (ir) spectra of oligosaccharides for the presence of sulfate groups. This assay can rapidly and nondestructively detect the presence of sulfate in as little as 1 nmol (approximately 2 micrograms) of a glycoprotein-derived monosulfated decasaccharide. The neural network was trained to recognize the presence of sulfate groups by presenting it with 45 ir spectra of sulfated and nonsulfated mono- and oligosaccharides. No prior knowledge of the characteristic ir spectral features of a sulfate group was needed as input. The training process required between 3 and 10 h, while analysis of a spectrum with the trained neural network requires only 0.1 s.

Carbohydrate dynamics at a micellar surface: GD1a headgroup transformations revealed by NMR spectroscopy.

The conformational dynamics of the carbohydrate headgroup of ganglioside GD1a, NeuAc alpha 2-->3Gal beta 1-->3GalNAc beta 1-->4[NeuAc alpha 2-->3]Gal beta 1-->4Glc beta 1-->1Cer, anchored in a perdeuterated dodecylphosphocholine micelle in aqueous solution, were probed by high resolution NMR spectroscopy. The observed 1H/1H NOE interactions revealed conformational averaging of the terminal NeuAc alpha 2-->3Gal and Gal beta 1-->3GalNAc glycosidic linkages. The pronounced flexibility of this trisaccharide moiety was substantiated further by two-dimensional proton-detected 13C T1, T1 rho and 1H/13C NOE measurements. The anchoring effect of the micelle allowed the detection of conformational fluctuations of the headgroup on the time scale of a few hundred picoseconds. NMR experiments performed on the GD1a/DPC micelles in H2O at low temperatures permitted the observation of hydroxyl proton resonances, contributing valuable conformational information.

Branched monosialo gangliosides of the lacto-series isolated from bovine erythrocytes: characterization of a novel ganglioside, NeuGc-isooctaosylceramide.

We report the isolation, purification, and structural characterization of three monosialo gangliosides (G-1, G-2, and G-3) from bovine erythrocytes. Each of the purified compounds migrates as a single band on thin-layer chromatography in three solvent systems. All three gangliosides contain ceramide (Cer) as the lipid protein, with d18:1 sphingosine as the predominant long-chain base and with C18:0, C18:1, C20:0, C20:1, and C22:1 fatty acids, as determined by gas chromatography. The structural characterization of the carbohydrate moieties of G-1, G-2, and G-3 involved glycosyl composition analysis, methylation studies, sequential exoglycosidase hydrolysis, and one-dimensional 1H NMR spectroscopy of the native gangliosides. Furthermore, the oligosaccharides were released from the sphingolipids by endoglycoceramidase and fully sequenced by one- and two-dimensional 1H NMR spectroscopy in conjunction with fast-atom-bombardment mass spectrometry and exoglycosidase treatment. The structures are as follows: [formula: see text] Gangliosides such as G-1, G-2, and G-3, with branched oligosaccharide chains comprising a number of N-acetyllactosamine (Gal-GlcNAc) moieties, are abundant in erythrocytes from various mammalian species. The simultaneous occurrence of sialic acid and alpha-galactose as terminal sugars in these gangliosides, however, is relatively rare. Specifically, G-1 represents a ganglioside with a novel structure.

Structures of monosialyl oligosaccharides isolated from the respiratory mucins of a non-secretor (O, Lea+b-) patient suffering from chronic bronchitis. Characterization of a novel type of mucin carbohydrate core structure.

Mucin glycopeptides were prepared from the respiratory mucus of a non-secretor, chronic bronchitis patient with blood group O, Le(a)+b-. Oligosaccharides were released by alkaline borohydride treatment and purified by anion-exchange chromatography, size-exclusion chromatography and high-performance liquid chromatography on a silica-bonded alkylamine column. Structural studies employed 500-MHz 1H-NMR spectroscopy and fast atom bombardment-mass spectrometry. Twenty-six monosialyl oligosaccharides, ranging in size from di- to octasaccharide, were fully characterized in this study. The sialic acid occurs either alpha(2-->3)- or alpha(2-->6)-linked to a galactosyl residue, or alpha(2-->6)-linked to GalNAc-ol. In keeping with the non-secretor status of the patient, no structures with an alpha(-->2)-linked fucose residue were found. Nine of the structures had fucose present in alpha(-->3) linkage, either in the Le(x) determinant or in the sialyl-Le(x) determinant, or both (structure 11). Only one oligosaccharide (structure 3b) was seen with fucose alpha (1-->4)-linked in the Le(a) determinant. Eight structures isolated from the mucins of the non-secretor patient had not been found previously in the respiratory mucins of secretor individuals; they are listed below. Among these, structure 4 alpha represents a novel type of mucin carbohydrate core structure.

Structure determination of the intact major sialylated oligosaccharide chains of recombinant human erythropoietin expressed in Chinese hamster ovary cells.

Recombinant human erythropoietin (rHuEPO) is used abundantly in the clinic to stimulate red blood cell growth in anaemic patients. The efficacy of the drug depends strongly on the extent of sialylation of its carbohydrate moiety. Prompted by conflicting literature reports on the issue, we reinvestigated the structures of the intact sialylated carbohydrate chains of rHuEPO expressed in Chinese hamster ovary (CHO) cells. The asparagine-linked oligosaccharides were released from rHuEPO with N-glycanase and fractionated by anion-exchange chromatography. The O-linked oligosaccharides were released under alkaline borohydride conditions. The primary structures of the major sialylated N- and O-type oligosaccharides were identified by 500-MHz 1H-NMR spectroscopy, supported by data from composition analysis, methylation analysis, low- and high-pH anion-exchange chromatography, and fast atom bombardment-mass spectrometry. The most abundant N-linked oligosaccharides in CHO cell-derived rHuEPO were found to be di-antennary, 2,4-branched tri-antennary, 2,6-branched tri-antennary and tetra-antennary chains (in the ratio of 7:6:5:82), with the latter containing between zero and three repeating N-acetyllactosamine units, in well-defined branches. The major (> 95%) di-, tri- and tetra-antennary structures are fully sialylated, i.e. they have two, three and four sialic acid residues, respectively, linked exclusively alpha(2-->3) to galactose residues. The majority (> 95%) of N-linked structures contain alpha(1-->6)-linked fucose at the proximal GlcNAc residue. The O-type mono- and disialyl oligosaccharides were characterized as a linear tri- and a branched tetra-saccharide, respectively.