Mice lacking alpha1,3-fucosyltransferase IX demonstrate disappearance of Lewis x structure in brain and increased anxiety-like behaviors

The 3-fucosyl-N-acetyllactosamine [Lewis x (Le(x)), CD15, SSEA-1] carbohydrate structure is expressed on several glycolipids, glycoproteins, and proteoglycans of the nervous system and has been implicated in cell-cell recognition, neurite outgrowth, and neuronal migration during development. To characterize the functional role of Le(x) carbohydrate structure in vivo, we have generated mutant mice that lack alpha1,3-fucosyltransferase IX (Fut9(-/-)). Fut9(-/-) mice were unable to synthesize the Le(x) structure carried on glycoproteins and glycolipids in embryonic and adult brain. However, no obvious pathological differences between wild-type and Fut9(-/-) mice were found in brain. In behavioral tests, Fut9(-/-) mice exhibited increased anxiety-like responses in dark-light preference and in elevated plus maze tests. Immunohistochemical analysis showed that the number of calbindin-positive neurons was decreased in the basolateral amygdala in Fut9(-/-) mice. These observations indicated that the carbohydrates synthesized by Fut9 play critical roles in functional regulations of interneurons in the amygdalar subdivisions and suggested a role for the Le(x) structure in some aspects of emotional behavior in mice.     

Characterization of the rat alpha(1,3)galactosyltransferase: evidence for two independent genes encoding glycosyltransferases that synthesize Galalpha(1,3)Gal by two separate glycosylation pathways

The important xenoepitope Galalpha(1,3)Gal was thought to be exclusively synthesized by a single alpha(1,3)galactosyltransferase. However, the cloning of the distant family member rat iGb3 synthase, which is also capable of synthesizing Galalpha(1,3)Gal as the glycolipid structure iGb3, challenges the notion that alpha(1,3)galactosyltransferase is the sole Galalpha(1,3)Gal-synthesizing enzyme. We describe the cloning of the rat homolog of alpha(1,3)galactosyltransferase, showing that indeed the rat expresses two distinct alpha(1,3)galactosyltransferases, alpha(1,3)GT and iGb3 synthase. Rat alpha(1,3)galactosyltransferase shows a high amino acid sequence identity with the alpha(1,3)galactosyltransferase of mouse (90%), pig (76%), and ox (75%), in contrast to the low amino acid sequence identity (42%) with iGb3 synthase. The rat alpha(1,3)galactosyltransferase is expressed in heart, brain, spleen, kidney, and liver and has a similar intron/exon structure to the mouse alpha(1,3)galactosyltransferase. Transfection studies show that in contrast to the iGb3 synthase, rat alpha(1,3)galactosyltransferase can synthesize Galalpha(1,3)Gal on glycoproteins but cannot synthesize the glycolipid iGb3, defining two separate glycosylation pathways for the synthesis of Galalpha(1,3)Gal. Furthermore iGb3 synthase was found to be distinct from alpha(1,3)GT with its ability to synthesize poly-alpha-Gal glycolipid structures.     

Recent intron gain in elongation factor-1alpha of colletid bees (Hymenoptera: Colletidae)

We discovered the presence of a unique spliceosomal intron in the F1 copy of elongation factor-1alpha (EF-1alpha) restricted to the bee family Colletidae (Hymenoptera: Apoidae). The intron ranges in size from 101 to 1044 bp and shows no positional sliding. Our data also demonstrate the complete absence of this intron from exemplars representing all other bee families, as well as from close hymenopteran relatives. A review of the literature finds that this intron is likewise absent from all other arthropods for which data are available. This provides unambiguous evidence for a relatively recent intron insertion event in the colletid common ancestor and, at least in this specific instance, lends support to the introns-late hypothesis. The comparative distribution of this novel intron also supports the monophyly of Colletidae and the exclusion of the Stenotritidae from this family, providing an example of the potential of some introns to act as robust markers of shared descent.     

Glycoengineering of alphaGal xenoantigen on recombinant peptide bearing the J28 pancreatic oncofetal glycotope

In human pancreatic adenocarcinoma, alterations of glycosylation processes leads to the expression of tumor-associated carbohydrate antigens, representing potential targets for cancer immunotherapy. Among these pancreatic tumor-associated carbohydrate antigens, the J28 glycotope located within the O-glycosylated mucin-like C-terminal domain of the fetoacinar pancreatic protein (FAPP) and expressed at the surface of human tumoral tissues, can be a good target for anticancer therapeutic vaccines. However, the oncodevelopmental self character of the J28 glycotope associated with the low immunogenicity of tumor-associated carbohydrate antigens may be a major obstacle to effective anti-tumor vaccine therapy. In this study, we have investigated a method to increase the immunogenicity of the recombinant pancreatic oncofetal J28 glycotope by glycoengineering Galalpha1,3Galss1,4GlcNAc-R (alphaGal epitope) which may be recognized by natural anti-alphaGal antibody present in humans. For this purpose, we have developed a stable Chinese hamster ovary cell clone expressing the alphaGal epitope by transfecting the cDNA encoding the alpha1,3galactosyltransferase. These cells have been previously equipped to produce the recombinant O-glycosylated C-terminal domain of FAPP carrying the J28 glycotope. As a consequence, the C-terminal domain of FAPP produced by these cells carries the alphaGal epitope on oligosaccharide structures associated with the J28 glycotope. Furthermore, we show that this recombinant "alpha1,3galactosyl and J28 glycotope" may not only be targeted by human natural anti-alphaGal antibodies but also by the mAbJ28, suggesting that the J28 glycotope remains accessible to the immune system as vaccinating agent. This approach may be used for many identified tumor-associated carbohydrate antigens which can be glycoengineered to carry a alphaGal epitope to increase their immunogenicity and to develop therapeutic vaccines.     

Beta-galactofuranose-containing O-linked oligosaccharides present in the cell wall peptidogalactomannan of Aspergillus fumigatus contain immunodominant epitopes

O-linked oligosaccharide groups ranging from di- to hexasaccharide were beta-eliminated by mild alkaline treatment under reducting conditions from the peptidogalactomannan of Aspergillus fumigatus mycelial cell wall. The resulting reduced oligosaccharides, which were the minor components of the peptidogalactomannan fraction, were fractionated to homogeneity by successive gel filtration and high-performance liquid chromatography. Their primary structures were determined based on a combination of techniques including gas chromatography, ESI-QTOF-MS, 1H COSY and TOCSY, and 1H-13C HMQC NMR spectroscopy and methylation analysis, to be: alpha-Glcp-(1 --> 6)-Man-ol, beta-Galf-(1 --> 6)-alpha-Manp-(1 --> 6)-Man-ol, beta-Galf-(1 --> 5)-beta-Galf-(1 --> 6)-alpha-Manp-(1 --> 6)-Man-ol and beta-Galf-(1 --> 5)-[beta-Galf-(1 --> 5]3-beta-Galf-(1 --> 6)-Man-ol. The beta-Galf containing oligosaccharides have not been previously described as fungal O-linked oligosaccharides. The peptidogalactomannan is antigenic and was recognized by human sera of patients with aspergillosis when probed by ELISA, but de-O-glycosylation rendered a 50% decrease in its reactivity. Furthermore, when tested in a hapten inhibition test, the isolated oligosaccharide alditols were able to block, on a dose-response basis, recognition between human sera and the intact peptidogalactomannan. The immunodominant epitopes were present in the tetra- and hexasaccharide, which contain a beta-Galf-(1 --> 5)-beta-Galf terminal group. These results suggest that the O-glycosidically linked oligosaccharide chains, despite being the less abundant carbohydrate component of the A. fumigatus peptidogalactomannan, may account for a significant part of its antigenicity, other than the known activity associated with the galactomannan component.     

Reduced alpha4beta1 integrin/VCAM-1 interactions lead to impaired pre-B cell repopulation in alpha 1,6-fucosyltransferase deficient mice

Mice with a targeted gene disruption of Fut8 (Fut8(-/-)) showed an abnormality in the transition from pro-B cell to pre-B cell, reduced peripheral B cells, and a decreased immunoglobulin production. Alpha 1,6-fucosyltransferase (FUT8) is responsible for the alpha 1,6 core fucosylation of N-glycans, which could modify the functions of glycoproteins. The loss of a core fucose in both very late antigen 4 (VLA-4, alpha4beta1 integrin) and vascular cell adhesion molecule 1 (VCAM-1) led to a decreased binding between pre-B cells and stromal cells, which impaired pre-B cells generation in Fut8(-/-) mice. Moreover, the B lineage genes, such as CD79a, CD79b, Ebf1, and Tcfe2a, were downregulated in Fut8(-/-) pre-B cells. Indeed, the frequency of preBCR(+)CD79b(low) cells in bone marrow pre-B cells in Fut8(-/-) was much lower than that in Fut8(+/+) cells. These results reveal a new role of core fucosylated N-glycans in mediating early B cell development and functions.     

Hydrolysis of Man9GlcNAc2 and Man8GlcNAc2 oligosaccharides by a purified alpha-mannosidase from Candida albicans

A soluble alpha-mannosidase from Candida albicans was purified to homogeneity by sequential size exclusion, ion exchange, and affinity chromatographies in columns of Sepharose CL6B, DEAE Bio-Gel A, and Concanavalin A Sepharose 4B, respectively. Analytical electrophoresis of the purified preparation in 10% SDS-polyacrylamide gels stained with Coomassie blue revealed a single polypeptide of 43 kDa that was responsible for enzyme activity. The purified enzyme primarily trimmed Man(9)GlcNAc(2) to produce Man(8)GlcNAc(2) isomer B and mannose as a function of time of incubation up to 12 h at 37 degrees C. Prolonged incubation with the enzyme resulted in the accumulation after 24 h of other oligosaccharides corresponding to Man(7)GlcNAc(2) and probably Man(6)GlcNAc(2). These two products were also observed when Man(8)GlcNAc(2) isomer B instead of Man(9)GlcNAc(2) was used as substrate. Other oligosaccharides, such as Man(6)GlcNAc(2)-Asn, Man(5)GlcNAc(2)-Asn, and the alpha1,3- and alpha1,6-linked mannobiosides, were not hydrolyzed at all. These properties are consistent with an alpha1,2-mannosidase that may represent a new member of the glycosylhydrolase family 47.     

Sialylation and sulfation of lactosylceramide distinctly regulate anchorage-independent growth,apoptosis, and gene expression in 3LL Lewis lung carcinoma cells

To investigate the significance of sialylation and sulfation of lactosylceramide in transformed cells, we established ganglioside GM3- and lactosylsulfatide (SM3)-reconstituted cells by transfecting cDNAs of GM3 synthase and cerebroside sulfotransferase into the J5 subclone of 3LL Lewis lung carcinoma cells. The J5 clone was selected for the transfection of these genes because it lacks GM3 and SM3 but accumulates lactosylceramide. The anchorage-dependent growth of both GM3- and SM3-reconstituted cells was similar. However, anchorage-independent growth (as measured by colony-forming ability in soft agar) of the SM3- reconstituted cells was almost completely lost, which supports our previous observation showing the suppression of tumorigenic potential in vivo and beta1 integrin gene expression induced by the introduction of cerebroside sulfotransferase gene (Kabayama et al. [2001] J. Biol. Chem., 276, 26777-26783). The GM3-reconstituted cells formed a significantly higher number of colonies in soft agar compared to mock-transfected cells and began to proliferate and become resistant to apoptosis when serum was depleted, indicating that endogenous GM3 is essential for maintaining these fundamental properties of malignant cells. We also found that serum-induced ERK1/2 activation was suppressed in the GM3-reconstituted cells, suggesting that anchorage-independent cell cycle initiation by endogenous GM3 is elicited through pathway(s) independent of ERK1/2 activation. The selective down-regulation of platelet-derived growth factor (PDGF)-dependent ERK1/2 activation in the GM3-reconstituted cells was due to the substantial decreases of PDGF alpha receptor mRNA and protein, but in the SM3-reconstituted cells PDGF alpha receptor expression was similar to mock cells. Thus, endogenously produced GM3 and SM3 differentially and distinctly regulate tumor-progression ability, that is, GM3 leads the transformed phenotype of J5 cells to promotion and SM3 to abrogation.     

Down-regulation of trypsinogen expression is associated with growth retardation in alpha1,6-fucosyltransferase-deficient mice: attenuation of proteinase-activated receptor 2 activity

Alpha1,6-fucosyltransferase (Fut8) plays important roles inphysiological and pathological conditions. Fut8-deficient (Fut8^–/–)mice exhibit growth retardation, earlier postnatal death, andemphysema-like phenotype. To investigate the underlying molecularmechanism by which growth retardation occurs, we examined themRNA expression levels of Fut8^–/– embryos (18.5days postcoitum [dpc]) using a cDNA microarray. The DNA microarrayand real-time polymerase chain reaction (PCR) analysis showedthat a group of genes, including trypsinogens 4, 7, 8, 11, 16,and 20, were down-regulated in Fut8^–/– embryos.Consistently, the expression of trypsinogen proteins was foundto be lower in Fut8^–/– mice in the duodenum, smallintestine, and pancreas. Trypsin, an active form of trypsinogen,regulates cell growth through a G-protein-coupled receptor,the proteinase-activated receptor 2 (PAR-2). In a cell culturesystem, a Fut8 knockdown mouse pancreatic acinar cell carcinoma,TGP49-Fut8-KDs, showed decreased growth rate, similar to thatseen in Fut8^–/– mice, and the decreased growth ratewas rescued by the application of the PAR-2-activating peptide(SLIGRL-NH₂). Moreover, epidermal growth factor (EGF)-inducedreceptor phosphorylation was attenuated in TGP49-Fut8-KDs, whichwas highly associated with a reduction of trypsinogens mRNAlevels. The addition of exogenous EGF recovered c-fos, c-jun,and trypsinogen mRNA expression in TGP49-Fut8-KDs. Again, theEGF-induced up-regulation of c-fos and c-jun mRNA expressionwas significantly blocked by the protein kinase C (PKC) inhibitor.Our findings clearly demonstrate a relationship between Fut8and the regulation of EGF receptor (EGFR)-trypsin-PAR-2 pathwayin controlling cell growth and that the EGFR-trypsin-PAR-2 pathwayis suppressed in TGP49-Fut8-KDs as well as in Fut8^–/–mice.     

A specific detection of GlcNAcbeta1-6Manalpha1 branches in N-linked glycoproteins based on the specificity of N-acetylglucosaminyltransferase VI

Malignant transformation is often accompanied by an aberrant glycosylation profile of the cell surface-in particular, the production of GlcNAcbeta1-6Manalpha1 branches in N-linked glycoproteins. To identify the target glycoproteins, we show a method using recombinant chicken N-acetylglucosaminyltransferase VI (GnT VI) and radiolabeled uridine (5'-)diphosphate-GlcNAc. The assay exploits the fact that GnT VI has a strict requirement for the GlcNAcbeta1-6Manalpha1 structure for activity, when a pyridylaminated free N-glycan is used as the acceptor substrate. Human asialo-agalacto alpha1-acid glycoprotein (AGP), which is known to contain GlcNAcbeta1-6Manalpha1 branches in its N-linked glycan chains, was radiolabeled when reacted with GnT VI, whereas human asialo-agalacto transferrin and bovine fetuin, neither of which contains a GlcNAcbeta1-6Manalpha1 structure were not, thus corroborating the specificity of the assay. Several proteins from human serum after pretreatment with sialidase and beta-galactosidase could be detected using the assay. One was identified as AGP from its mobility on SDS-PAGE, demonstrating the potential of this assay even with crude materials. Furthermore, this method could detect a protein that was also positively stained with leukoagglutinating phytohemagglutinin (L(4)-PHA) using glycoproteins prepared from WiDr human colon cancer cells. This method should provide a useful complement to the current method, which relies on the specificity of L(4)-PHA.     

Anti-alpha-galactosyl antibodies recognizing epitopes terminating with alpha1,4-linked galactose: human natural and mouse monoclonal anti-NOR and anti-P1 antibodies

The rare NOR erythrocytes, which are agglutinated by most human sera, contain unique glycosphingolipids (globoside elongation products) terminating with the sequence Galalpha1-4GalNAcbeta1-3Gal- recognized by common natural human antibodies. Anti-NOR antibodies were isolated from several human sera by affinity procedures, and their specificity was tested by inhibition of antibody binding to NOR-tri-polyacrylamide (PAA) conjugate (ELISA) by the synthetic oligosaccharides, Galalpha1-4GalNAcbeta1-3Gal (NOR-tri), Galalpha1-4GalNAc (NOR-di), Galalpha1-4Galbeta1-3Galbeta1-4Glc ((Gal)3Glc), and Galalpha1-4Gal (P1-di). Two major types of subspecificity of anti-NOR antibodies were found. Type 1 antibodies were found to react strongly with (Gal)3Glc and NOR-tri and weakly with P1-di and NOR-di, which indicated specificity for the trisaccharide epitope Galalpha1-4Gal/GalNAcbeta1-3Gal. Type 2 antibodies were specific to Galalpha1-4GalNAc, because they were inhibited most strongly by NOR-tri and NOR-di and were not (or very weakly) inhibited by (Gal)3Glc and P1-di. Monoclonal anti-NOR antibodies were obtained by immunizing mice with NOR-tri-human serum albumin (HSA) conjugate and were found to have type 2 specificity. All anti-NOR antibodies reacted specifically with NOR glycolipids on thin-layer plates. The cross-reactivity of type 1 anti-NOR antibodies with Galalpha1-4Gal drew attention to a possible antigenic relationship between NOR and blood group P system glycolipids. The latter glycolipids include Pk (Galalpha1-4Galbeta1-4Glc-Cer) present in all normal erythrocytes and P1 (Galalpha1-4Galbeta1-4GlcNAcbeta1-3Galbeta1-4Glc-Cer) present only in P1 erythrocytes. Sera of some P2 (P1-negative) persons contain natural anti-P1 antibodies. This prompted us to test the specificity of anti-P1 antibodies. Natural human anti-P1 isolated from serum of P2 individual and mouse monoclonal anti-P1 were best inhibited by Galalpha1-4Galbeta1-4GlcNAc (P1-tri) and did not react with NOR-tri and NOR-di. Monoclonal anti-P1 bound to Pk and P1 glycolipids and not to NOR glycolipids. These results indicated an entirely different specificity of anti-NOR and anti-P1 antibodies. Human serum samples differed in the content of anti-alpha-galactosyl antibodies, including both types of anti-NOR. In the sera of some individuals, type 1 or type 2 anti-NOR antibodies dominated, and other samples contained mixtures of both types of anti-NOR. The biological significance of these new abundant anti-alpha-galactosyl antibodies still awaits elucidation.     

Evidence of positively selected sites in mammalian alpha-defensins

Alpha-defensins are a family of mammalian antimicrobial peptides that exhibit variable activity against a panel of microbes, including bacteria, fungi, and enveloped viruses. We have employed a maximum-likelihood approach to detect evidence of positive selection (adaptive evolution) in the evolution of these important molecules of the innate immune response. We have identified 14 amino acid sites that are predicted to be subject to positive selection. Furthermore, we show that all these sites are located in the mature antimicrobial peptide and not in the prepropeptide region of the molecule, implying that they are of functional importance. These results suggest that mammalian alpha-defensins have been under selective pressure to evolve in response to potentially infectious challenges by fast-evolving microbes.     

GlycoPEGylation of recombinant therapeutic proteins produced in Escherichia coli

Covalent attachment of polyethylene glycol, PEGylation, has been shown to prolong the half-life and enhance the pharmacodynamics of therapeutic proteins. Current methods for PEGylation, which rely on chemical conjugation through reactive groups on amino acids, often generate isoforms in which PEG is attached at sites that interfere with bioactivity. Here, we present a novel strategy for site-directed PEGylation using glycosyltransferases to attach PEG to O-glycans. The process involves enzymatic GalNAc glycosylation at specific serine and threonine residues in proteins expressed without glycosylation in Escherichia coli, followed by enzymatic transfer of sialic acid conjugated with PEG to the introduced GalNAc residues. The strategy was applied to three therapeutic polypeptides, granulocyte colony stimulating factor (G-CSF), interferon-alpha2b (IFN-alpha2b), and granulocyte/macrophage colony stimulating factor (GM-CSF), which are currently in clinical use.     

Core fucosylation of N-linked glycans in leukocyte adhesion deficiency/congenital disorder of glycosylation IIc fibroblasts

Leukocyte adhesion deficiency/congenital disorder of glycosylation IIc (LAD II/CDG IIc) is a genetic disease characterized by a decreased expression of fucose in glycoconjugates, resulting in leukocyte adhesion deficiency and severe morphological and neurological abnormalities. The biochemical defect is a reduced transport of guanosine diphosphate-L-fucose (GDP-L-fucose) from cytosol into the Golgi compartment, which reduces its availability as substrate for fucosyltransferases. The aim of this study was to determine the effects of a limited supply of GDP-L-fucose inside the Golgi on core fucosylation (alpha1,6-fucose linked to core N-acetylglucosamine [GlcNAc]) of N-linked glycans in LAD II fibroblasts. The results showed that, although [3H]fucose incorporation was generally reduced in LAD II cells, core fucosylation was affected to a greater extent compared with other types of fucosylation of N-linked oligosaccharides. In particular, core fucosylation was found to be nearly absent in biantennary negatively charged oligosaccharides, whereas other types of structures, in particular triantennary neutral species, were less affected by the reduction. Expression and activity of alpha1,6-fucosyltransferase (FUT8) in control and LAD II fibroblasts were comparable, thus excluding the possibility of a decreased activity of the transferase. The data obtained confirm that the concentration of GDP-L-fucose inside the Golgi can differentially affect the various types of fucosylation in vivo and also indicate that core fucosylation is not dependent only on the availability of GDP-L-fucose, but it is significantly influenced by the type of oligosaccharide structure. The relevant reduction in core fucosylation observed in some species of oligosaccharides could also provide clues for the identification of glycans involved in the severe developmental abnormalities observed in LAD II.     

The binding of human glial cell line-derived neurotrophic factor to heparin and heparan sulfate: importance of 2-O-sulfate groups and effect on its interaction with its receptor,GFRalpha1

We report ELISA studies of the glycosaminoglycan binding properties of recombinant human glial cell line-derived neurotrophic factor (GDNF). We demonstrate relatively high affinity binding as soluble heparin competes with an IC50 of 0.1 micro g/ml. The binding of GDNF to heparin is particularly dependent on the presence of 2-O-sulfate groups. Highly sulfated heparan sulfate is also an effective competitor for GDNF binding. We also show that heparin at low concentrations protects GDNF from proteolytic modification by an endoprotease and also promotes the binding of GDNF to its receptor polypeptide, GFRalpha1. In both of these actions, 2-O-desulfated heparin is less effective. Considered overall, these findings provide strong support for a hypothesis that the bioactivity of GDNF during prenatal development is essentially dependent on the binding of this growth factor to 2-O-sulfate-rich heparin-related glycosaminoglycan.     

N-glycans of recombinant human acid alpha-glucosidase expressed in the milk of transgenic rabbits

Pompe disease is a lysosomal glycogen storage disorder characterized by acid alpha-glucosidase (GAA) deficiency. More than 110 different pathogenic mutations in the gene encoding GAA have been observed. Patients with this disease are being treated by intravenous injection of recombinant forms of the enzyme. Focusing on recombinant approaches to produce the enzyme means that specific attention has to be paid to the generated glycosylation patterns. Here, human GAA was expressed in the mammary gland of transgenic rabbits. The N-linked glycans of recombinant human GAA (rhAGLU), isolated from the rabbit milk, were released by peptide-N(4)-(N-acetyl-beta-glucosaminyl)asparagine amidase F. The N-glycan pool was fractionated and purified into individual components by a combination of anion-exchange, normal-phase, and Sambucus nigra agglutinin-affinity chromatography. The structures of the components were analyzed by 500 MHz one-dimensional and 600 MHz cryo two-dimensional (total correlation spectroscopy [TOCSY] nuclear Overhauser enhancement spectroscopy) (1)H nuclear magnetic resonance spectroscopy, combined with two-dimensional (31)P-filtered (1)H-(1)H TOCSY spectroscopy, matrix-assisted laser desorption ionization time-of-flight mass spectrometry, and high-performance liquid chromatography (HPLC)-profiling of 2-aminobenzamide-labeled glycans combined with exoglycosidase digestions. The recombinant rabbit glycoprotein contained a broad array of different N-glycans, comprising oligomannose-, hybrid-, and complex-type structures. Part of the oligomannose-type glycans showed the presence of phospho-diester-bridged N-acetylglucosamine. For the complex-type glycans (partially) (alpha2-6)-sialylated (nearly only N-acetylneuraminic acid) diantennary structures were found; part of the structures were (alpha1-6)-core-fucosylated or (alpha1-3)-fucosylated in the upper antenna (Lewis x). Using HPLC-mass spectrometry of glycopeptides, information was generated with respect to the site-specific location of the various glycans.