Binding specificity of the recombinant cytoplasmic domain of Cordyceps militaris β-1,3-glucan synthase catalytic subunit

The cytoplasmic domain of the medicinal mushroom Cordyceps militaris β-1,3-glucan synthase catalytic subunit Fks1 was expressed as a fusion protein with an N-terminal hexahistidine tag and glutathione S-transferase in an Escherichia coli cell-free translation system, and was assayed for binding specificity. The recombinant cytoplasmic domain bound specifically to UDP-agarose and lichenan (β-glucan), but not to ADP-agarose, GDP-agarose, or other carbohydrates.     

Expression of β-1,4-galactosyltransferase and suppression of β-N-acetylglucosaminidase to aid synthesis of complex N-glycans in insect Drosophila S2 cells

Previously, we have shown that simple paucimannosidic N-glycan structures in insect Drosophila S2 cells arise mainly because of β-N-acetylglucosaminidase (GlcNAcase) action. Thus, in an earlier report, we suppressed GlcNAcase activity and clearly demonstrated that more complex N-glycans with two terminal N-acetylglucosamine (GlcNAc) residues were then synthesized. In the present work, we investigated the synergistic effects of β-1,4-galactosyltransferase (GalT) expression and GlcNAcase suppression on N-glycan patterns. We found that the N-glycan pattern of human erythropoietin secreted by engineered S2 cells expressing GalT but not GlcNAcase was complete, even in small portion, except for sialylation; the N-glycan structures had two terminal galactose (Gal) residues. When GalT was expressed but GlcNAcase was not inhibited, N-glycan with GlcNAc and Gal at only one branch end was synthesized. Therefore, it will be possible to express a complete functional human glycoprotein in engineered Drosophila S2 cells by suppressing GlcNAcase and co-expressing additional glycosyltransferases of N-glycosylation pathway.     

Mice with a homozygous deletion of the Mgat2 gene encoding UDP-N-acetylglucosamine:α-6-d-mannoside β1,2-N-acetylglucosaminyltransferase II: a model for congenital disorder of glycosylation type IIa

Mice homozygous for a deletion of the Mgat2 gene encoding UDP-N-acetylglucosamine:α-6-d-mannoside β1,2-N-acetylglucosaminyltransferase II (GlcNAcT-II, EC 2.4.1.143) have been reported. GlcNAcT-II is essential for the synthesis of complex N-glycans. The Mgat2-null mice were studied in a comparison with the symptoms of congenital disorder of glycosylation type IIa (CDG-IIa) in humans. Mutant mouse tissues were shown to be deficient in GlcNAcT-II enzyme activity and complex N-glycan synthesis, resulting in severe gastrointestinal, hematologic and osteogenic abnormalities. All mutant mice died in early post-natal development. However, crossing the Mgat2 mutation into a distinct genetic background resulted in a low frequency of survivors exhibiting additional and novel disease signs of CDG-IIa. Analysis of N-glycan structures in the kidneys of Mgat2-null mice showed a novel bisected hybrid N-glycan structure in which the bisecting GlcNAc residue was substituted with a β1,4-linked galactose or the Lewis^x structure. These studies suggest that some of the functions of complex N-glycan branches are conserved in mammals and that human disease due to aberrant protein N-glycosylation may be modeled in the mouse, with the expectation in this case of gaining insights into CDG-IIa disease pathogenesis. Further analyses of the Mgat2-deficient phenotype in the mouse have been accomplished involving cells in which the Mgat2 gene is dispensable, as well as other cell lineages in which a severe defect is present. Pre-natal defects appear in a significant number of embryos, and likely reflect a limited window of time in which a future therapeutic approach might effectively operate.     

Control of glycoprotein synthesis: substrate specificity of rat liver UDP-GlcNAc:Manα3R β2-N-acetylglucosaminyl-transferase I using synthetic substrate analogues

UDP-GlcNAc: Man3R 2-N-acetylglucosaminyltransferase I (GlcNAc-T I; EC 2.4.1.101) is the key enzyme in the synthesis of complex and hybrid N-glycans. Rat liver GlcNAc-T I has been purified more than 25,000-fold (M _r 42,000). TheV _max for the pure enzyme with [Man6(Man3)Man6](Man3)Man4GlcNAc4GlcNAc-Asn as substrate was 4.6 µmol min^–1 mg^–1. Structural analysis of the enzyme product by proton nuclear magnetic resonance spectroscopy proved that the enzyme adds anN-acetylglucosamine (GlcNAc) residue in 1–2 linkage to the Man3Man-terminus of the substrate. Several derivatives of Man6(Man3)Man-R, a substrate for the enzyme, were synthesized and tested as substrates and inhibitors. An unsubstituted equatorial 4-hydroxyl and an axial 2-hydroxyl on the -linked mannose of Man6(Man3)Man-R are essential for GlcNAc-T I activity. Elimination of the 4-hydroxyl of the 3-linked mannose (Man) of the substrate increases theK _M 20-fold. Modifications on the 6-linked mannose or on the core structure affect mainly theK _M and to a lesser degree theV _max, e.g., substitutions of the Man6 residue at the 2-position by GlcNAc or at the 3- and 6-positions by mannose lower theK _M, whereas various other substitutions at the 3-position increase theK _M slightly. Man6(Man3)4-O-methyl-Man4GlcNAc was found to be a weak inhibitor of GlcNAc-T I.Abbreviations BSA Bovine serum albumin - Bn benzyl - Fuc, F l-fucose - Gal, G d-galactose - GalNAc, GA N-acetyl-d-galactosamine - Glc d-glucose - GlcNAc, Gn N-acetyl-d-glucosamine - HPLC high performance liquid chromatography - Man, M d-mannose - mco 8-methoxycarbonyl-octyl, (CH₂)₈ COOOCH₃ - Me methyl - MES 2-(N-morpholino)ethanesulfonate - NMR nuclear magnetic resonance - PMSF phenylmethylsulfonylfluoride - pnp p-nitrophenyl - SDS sodium dodecyl sulfate - T transferase - Tal d-talose - Xyl d-xylose; - {0, 2 + F} Man6 (GlcNAc2Man3) Man4GlcNAc4 (Fuc6) GlcNAc - {2, 2} GlcNAc2Man6 (GlcNAc2Man3) Man4GlcNAc4GlcNAc; M₅-glycopeptide, Man6 (Man3) Man6 (Man3) Man4 GlcNAc4GlcNAc-Asn Enzymes: GlcNAc-transferase I, EC 2.4.1.101; GlcNAc-transferase II, EC 2.4.1.143; GlcNAc-transferase III, EC 2.4.1.144; GlcNAc-transferase IV, EC 2.4.1.145; GlcNAc-transferase V, UDP-GlcNAc: GlcNAc2 Man6-R (GlcNAc to Man) 6-GlcNAc-transferase; GlcNAc-transferase VI, UDP-GlcNAc: GlcNAc6(GlcNAc2) Man6-R (GlcNAc to Man) 4-GlcNAc-transferase; Core 1 3-Gal-transferase, EC 2.4.1.122; 4-Gal-transferase, EC 2.4.1.38; 3-Gal-transferase, UDP-Gal: GlcNAc-R 3-Gal-transferase; blood group i 3-GlcNAc-transferase, EC 2.4.1.149; blood group I 6-GlcNAc-transferase, UDP-GlcNAc: GlcNAc3Gal-R (GlcNAc to Gal) 6-GlcNAc-transferase.     

In Vivo Expression of UDP-N-Acetylglucosamine: α-3-D-Mannoside β-1,2-N-Acetylglucosaminyltransferase I (GnT-1) in Aspergillus oryzae and Effects on the Sugar Chain of α-Amylase

UDP-N-Acetylglucosamine: α-3-D-mannoside β-1,2-N-acetylglucosaminyltransferase I (GnT-I) is an essential enzyme in the conversion of high mannose type oligosaccharide to the hybrid or complex type. The full length of the rat GnT-I gene was expressed in the filamentous fungus Aspergillus oryzae. A microsomal preparation from a recombinant fungus (strain NG) showed GnT-I activity that transferred N-acetylglucosamine residue to acceptor heptaose, Man₅GlcNAc₂. The N-linked sugar chain of α-amylase secreted by the strain showed a peak of novel retention on high performance liquid chromatography that was same as a reaction product of in vitro GnT-1 assay. The peak of oligosaccharide disappeared on HPLC after β-N-acetylglucosaminidase treatment. Mass analysis supported the presence of GlcNAcMan₅GlcNAc₂ as a sugar chain of α-amylase from strain NG. Chimera of GnT-I with green fluorescent protein (GFP) showed a dotted pattern of fluorescence in the mycelia, suggesting localization at Golgi vesicles. We concluded that GnT-1 was functionally expressed in A. oryzae cells and that N-acetylglucosamine residue was transferred to N-glycan of α-amylase in vivo. A. oryzae is expected to be a potential host for the production of glycoprotein with a genetically altered sugar chain.     

Synthetic substrate analogues for UDP-GlcNAc: Manα1-6R β(1-2)-N-acetylglucosaminyltransferase II. Substrate specificity and inhibitors for the enzyme

UDP-GlcNAc: Man1-6R (1-2)-N-acetylglucosaminyltransferase II (GlcNAc-T II; EC 2.4.1.143) is a key enzyme in the synthesis of complexN-glycans. We have tested a series of synthetic analogues of the substrate Man1-6(GlcNAc1-2Man1-3)Man-O-octyl as substrates and inhibitors for rat liver GlcNAc-T II. The enzyme attachesN-acetylglucosamine in 1-2 linkage to the 2-OH of the Man1-6 residue. The 2-deoxy analogue is a competitive inhibitor (K _i=0.13mm). The 2-O-methyl compound does not bind to the enzyme presumably due to steric hindrance. The 3-, 4- and 6-OH groups are not essential for binding or catalysis since the 3-, 4- and 6-deoxy and -O-methyl derivatives are all good substrates. Increasing the size of the substituent at the 3-position to pentyl and substituted pentyl groups causes competitive inhibition (K _i=1.0–2.5mm). We have taken advantage of this effect to synthesize two potentially irreversible GlcNAc-T II inhibitors containing a photolabile 3-O-(4,4-azo)pentyl group and a 3-O-(5-iodoacetamido)pentyl group respectively. The data indicate that none of the hydroxyls of the Man1-6 residue are essential for binding although the 2- and 3-OH face the catalytic site of the enzyme. The 4-OH group of the Man-O-octyl residue is not essential for binding or catalysis since the 4-deoxy derivative is a good substrate; the 4-O-methyl derivative does not bind. This contrasts with GlcNAc-T I which cannot bind to the 4-deoxy-Man- substrate analogue. The data are compatible with our previous observations that a bisectingN-acetylglucosamine at the 4-OH position prevents both GlcNAc-T I and GlcNAc-T II catalysis. However, in the case of GlcNAc-T II, the bisectingN-acetylglucosamine prevents binding due to steric hindrance rather than to removal of an essential OH group. The 3-OH of the Man1-3 is an essential group for GlcNAc-T II since the 3-deoxy derivative does not bind to the enzyme. The trisaccharide GlcNAc1-2Man1-3Man-O-octyl is a good inhibitor (K _i=0.9mm). The above data together with previous studies indicate that binding of the GlcNAc1-2Man1-3Man- arm of the branched substrate to the enzyme is essential for catalysis. Abbreviations: GlcNAc-T I, UDP-GlcNAc:Man1-3R (1-2)-N-acetylglucosaminyltransferase I (EC 2.4.1.101); GlcNAc-T II, UDP-GlcNAc:Man1-6R (1-2)-N-acetylglucosaminyltransferase II (EC 2.4.1.143); MES, 2-(N-morpholino)ethane sulfonic acid monohydrate.     

ProcessingO-glycan core 1, Galβ1-3GalNAcα-R. Specificities of core 2, UDP-GlcNAc: Galβ1-3GalNAc-R(GlcNAc to GalNAc) β6-N-acetylglucosaminyltransferase and CMP-sialic acid:Galβ1-3GalNAc-R α3-sialyltransferase

To elucidate control mechanisms ofO-glycan biosynthesis in leukemia and to develop biosynthetic inhibitors we have characterized core 2 UDP-GlcNAc:Gal1-3GalNAc-R(GlcNAc to GalNAc) 6-N-acetylglucosaminyl-transferase (EC 2.4.1.102; core 2 6-GlcNAc-T) and CMP-sialic acid: Gal1-3GalNAc-R 3-sialyltransferase (EC 2.4.99.4; 3-SA-T), two enzymes that are significantly increased in patients with chronic myelogenous leukemia (CML) and acute myeloid leukemia (AML). We observed distinct tissue-specific kinetic differences for the core 2 6-GlcNAc-T activity; core 2 6-GlcNAc-T from mucin secreting tissue (named core 2 6-GlcNAc-T M) is accompanied by activities that synthesize core 4 [GlcNAc1-6(GlcNAc1-3)GalNAc-R] and blood group I [GlcNAc1-6(GlcNAc1-3)Gal-R] branches; core 2 6-GlcNAc-T in leukemic cells (named core 2 -GlcNAc-T L) is not accompanied by these two activities and has a more restricted specificity. Core 2 6-GlcNAc-T M and L both have an absolute requirement for the 4- and 6-hydroxyls ofN-acetylgalactosamine and the 6-hydroxyl of galactose of the Gal1-3GalNAc-benzyl substrate but the recognition of other substituents of the sugar rings varies, depending on the tissue. 3-sialytransferase from human placenta and from AML cells also showed distinct specificity differences, although the enzymes from both tissues have an absolute requirement for the 3-hydroxyl of the galactose residue of Gal1-3GalNAc-Bn. Gal1-3(6-deoxy)GalNAc-Bn and 3-deoxy-Gal1-3GalNAc-Bn competitively inhibited core 2 6-GlcNAc-T and 3-sialyltransferase activities, respectively.Abbreviations AFGP antifreeze glycoprotein - AML acute myeloid leukemia - Bn benzyl - CML chronic myelogenous leukemia - Fuc l-fucose - Gal, G d-galactose - GalNAc, GA N-acetyl-d-galactosamine - GlcNAc, Gn N-acetyl-d-glucosamine - HC human colonic homogenate - HO hen oviduct microsomes - HPLC high performance liquid chromatography - mco 8-methoxycarbonyl-octy - Me methyl - MES 2-(N-morpholino)ethanesulfonate - MK mouse kidney homogenate - onp o-nitrophenyl - PG pig gastric mucosal microsomes - pnp p-nitrophenyl - RC rat colonic mucosal microsomes - SA sialic acid - T transferase Enzymes: UDP-GlcNAc:Gal1-3GalNAc-R (GlcNAc to GalNAc) 6-N-acetylglucosaminyltransferase,O-glycan core 2 6-GlcNAc-transferase, EC 2.4.1.102; CMP-sialic acid: Gal1-3GalNAc-R 3-sialyltransferase,O-glycan 3-sialic acid-transferase, EC 2.4.99.4.     

Crystal structures of the laminarinase catalytic domain from Thermotoga maritima MSB8 in complex with inhibitors: essential residues for β-1,3- and β-1,4-glucan selection

Laminarinases hydrolyzing the β-1,3-linkage of glucans play essential roles in microbial saccharide degradation. Here we report the crystal structures at 1.65-1.82 ? resolution of the catalytic domain of laminarinase from the thermophile Thermotoga maritima with various space groups in the ligand-free form or in the presence of inhibitors gluconolactone and cetyltrimethylammonium. Ligands were bound at the cleft of the active site near an enclosure formed by Trp-232 and a flexible GASIG loop. A closed configuration at the active site cleft was observed in some molecules. The loop flexibility in the enzyme may contribute to the regulation of endo- or exo-activity of the enzyme and a preference to release laminaritrioses in long chain carbohydrate hydrolysis. Glu-137 and Glu-132 are proposed to serve as the proton donor and nucleophile, respectively, in the retaining catalysis of hydrolyzation. Calcium ions in the crystallization media are found to accelerate crystal growth. Comparison of laminarinase and endoglucanase structures revealed the subtle difference of key residues in the active site for the selection of β-1,3-glucan and β-1,4-glucan substrates, respectively. Arg-85 may be pivotal to β-1,3-glucan substrate selection. The similarity of the structures between the laminarinase catalytic domain and its carbohydrate-binding modules may have evolutionary relevance because of the similarities in their folds.     

Expression of β-1,4-galactosyltransferase-I affects cellular adhesion in human peripheral blood CD4 T cells

β-1,4-galactosyltransferase-I (β-1,4-GalT-I) has two isoforms that differ only in the length of their cytoplasmic domains. In this study, we found that both the long and short isoforms of β-1,4-GalT-I were expressed in human CD4⁺ T lymphocytes, and localized in the cytoplasm and on the plasma membrane. The expression level of β-1,4-GalT-I was increased in CD4⁺ T cells after stimulation with interleukin (IL)-2, and was further increased after stimulation with IL-2 + IL-12, but decreased after stimulation with IL-2 + IL-4 when compared to stimulation with IL-2 alone. We also demonstrated that the cellular adhesion of CD4⁺ T cells was significantly increased upon cytokine stimulation, and was inhibited by α-lactalbumin, indicating that the increase in adhesion was positively correlated with the expression and activity of long β-1,4-GalT-I. Collectively, the data suggest that β-1,4-GalT-I plays a role in the cellular adhesion of CD4⁺ T cells.     

Function and conformation analyses of an aspartate substitution of the invariant glycine in the integrin βI domain α1-α1′ helix

We showed that the αLβ2 integrin with the non-functional mutation G150D cannot be induced with Mg/EGTA to express the mAb KIM127 epitope, which reports the leg-extended conformation. We extended the study to the αIIbβ3, an integrin without an αI domain. The equivalent mutation, i.e. G161D, also resulted in an expressible, but non-adhesive αIIbβ3 integrin. An NMR study of synthetic peptides spanning the α1-α1′ helix of the β3 I domain shows that both wild-type and mutant peptides are α-helical. However, whereas in the wild-type peptide this helix is continuous, the mutant presents a discontinuity, or kink, precisely at the site of mutation G161D. Our results suggest that the mutation may lock integrin heterodimers in a bent conformation that prevents integrin activation via conformational extension.     

Galectin-3 binds to MUC1-N-terminal domain and triggers recruitment of β-catenin in MUC1-expressing mouse 3T3 cells

Background

Galectin-3 is expressed in a variety of tumors and its expression level is related with tumor progression. Aberrant expression of MUC1 in various tumors is also associated with a poor prognosis. It has been reported that MUC1 is a natural ligand of galectin-3.

Methods

A stable MUC1 transfectant was produced by introducing MUC1 cDNA into mouse 3T3 fibroblasts (MUC1/3T3 cells). MUC1 was prepared from MUC1/3T3 cells; MUC1-N-terminal domain (MUC1-ND) and -C-terminal domain (MUC1-CD) were separated by CsCl ultracentrifugation, and then the galectin-3-binding domain was determined by co-immuniprecipitation assay. After ligation of galectin-3 to 3T3/MUC1 cells, MUC1-CD was immunoprecipitated from the cell lysate. The immunoprecipitate was subjected to SDS-PAGE and Western blotting, followed by detection of co-immunoprecipitated β-catenin.

Results

Galectin-3 binds to the N-terminal domain of MUC1 but not to the C-terminal one. Galectin-3 present on the cell surface increased with the expression of MUC1 and is colocalized with MUC1. It should be noted that β-catenin was detected in the immunoprecipitate with anti-MUC1-CD Ab from a lysate of galectin-3-treated 3T3/MUC1 cells.

Conclusions

Galectin-3 binds to MUC1-ND and triggers MUC1-mediated signaling in 3T3/MUC1 cells, leading to recruitment of β-catenin to MUC1-CD.

General significance

This signaling may be another MUC1-mediated pathway and function in parallel with a growth factor-dependent MUC1-mediated pathway.     

Three-dimensional structure of the catalytic domain of the yeast β-(1,3)-glucan transferase Gas1: a molecular modeling investigation

The three-dimensional (3D) structure of the catalytic domain of Gas1p, a protein belonging to the only family of β-(1,3)-glucan transferases so far identified in yeasts and some pathogenic fungi (family GH-72), has been predicted by combining results derived from threading methods, multiple sequence alignments and secondary-structure predictions. The 3D model has allowed the identification of several residues that are predicted to play a crucial role in structural integrity, substrate recognition and catalysis. In particular, the model of the catalytic domain can be useful for designing site-directed mutagenesis experiments and for developing inhibitors of Gas1p enzymatic activity. Figure Three-dimensional models of the Gas1p catalytic domain as predicted using as template 7A3H (PDB code) protein Electronic Supplementary Material Supplementary material is available for this article at     

Arachidonic acid induces an increase of β-1,4-galactosyltransferase I expression in MDA-MB-231 breast cancer cells

Arachidonic acid (AA) is a common dietary n‐6 cis polyunsaturated fatty acid that under physiological conditions is present in an esterified form in cell membrane phospholipids, and it might be present in the extracellular microenvironment. AA and its metabolites are implicated in FAK activation and cell migration in MDA‐MB‐231 breast cancer cells, and an epithelial‐to‐mesenchymal‐like transition process in mammary non‐tumorigenic epithelial cells MCF10A. During malignant transformation is present an altered expression of glycosiltransferases, which promote changes on the glycosilation of cell‐surface proteins. The β‐1,4‐galactosyltransferase I (GalT I) is an enzyme that participates in a variety of biological functions including cell growth, migration, and spreading. However, the participation of AA in the regulation of GalT I expression and the role of this enzyme in the cell adhesion process in breast cancer cells remains to be investigated. In the present study, we demonstrate that AA induces an increase of GalT I expression through a PLA2α, Src, ERK1/2, and LOXs activities‐dependent pathway in MDA‐MB‐231 breast cancer cells. Moreover, MDA‐MB‐231 cells adhere to laminin via GalT I expression and pretreatment of cells with AA induces an increase of cell adhesion to laminin. In conclusion, our findings demonstrate, for the first time, that AA promotes an increase of GalT I expression through an AA metabolism, Src and ERK1/2 activities‐dependent pathway, and that GalT I plays a pivotal role in cell adhesion to laminin in MDA‐MB‐231 breast cancer cells. J. Cell. Biochem. 113: 3330–3341, 2012. © 2012 Wiley Periodicals, Inc.     

Expression of biologically active human clotting factor IX in Drosophila S2 cells: γ-carboxylation of a human vitamin K-dependent protein by the insect enzyme

The Drosophila γ-glutamyl carboxylase (dγC) has substrate recognition properties similar to that of the vertebrate γ-carboxylase (γC), and its carboxylated product yield, in vitro, was shown to be more than that obtained with the human enzyme. However, whether the Drosophila enzyme is able to γ-carboxylate the human vitamin K-dependent (VKD) proteins, such as the human coagulation factor IX (hFIX), as synthesized in cultured Drosophila cells was not known. To examine this possibility, the Drosophila Schnider (S2) cell line was transfected with a metallothionein promoter-regulated hFIX-expressing plasmid. After induction with copper ion, expression efficiency of the active hFIX was analyzed by performing enzyme-linked immunosorbent assey (ELISA) and coagulation test on the culture supernatant of the transfected S2 cells during 72 h of postinduction. In comparison with Chinese hamster ovary cell line, S2 cells showed higher (≈ 12-fold) expression level of the hFIX. The γ-carboxylation of the Drosophila-derived hFIX was confirmed by evaluation of the expressed protein, after being precipitated with barium citrate. The biological activity of the S2 cell-derived hFIX indicated the capability of S2 cells to fulfill the required γ-carboxylation of the expressed hFIX. Coexpression of the human γ-glutamyl carboxylases (hγC) was also shown to improve both expression and γ-carboxylation of the hFIX. This is the first in vivo data to describe the ability of the dγC to recognize the human-based propeptide as substrate, which is an essential step for production of biologically active γ-carboxylated VKD proteins.     

Spectrophotometric Methods for Monitoring the Microbial Transformation of Steroids: I. Determination of 9α-Fluorohydrocortisone and 9α-Fluoro-16α-Hydroxyhydrocortisone in Fermentation Broths

A method has been devised for the determination of 9alpha-fluorohydrocortisone and 9alpha-fluoro-16alpha-hydroxyhydrocortisone in fermentation broths. The method involves extraction of the two steroids from the broth with ethyl acetate, separation through the formation of the water-soluble borate complex of the vicinally hydroxylated steroid, and estimation of each steroid spectrophotometrically.     

Decreased UDP-GlcNAc:Glycopeptideβ-2-N-Acetylglucosaminyltransferase II activity in a ricin-resistant mutant of baby hamster kidney (BHK) cells

The ricin-resistant mutant baby hamster kidney (BHK) cell line RIC^R21 is unable to make the sialylated bi- or triantennary complexN-glycans found in wild type cells and accumulates instead non-bisected hybrid structures containing three Man residues and one or two sialylated antennae (Hugheset al 1983, Carbohydr Res 120215-34). Specific assays forN-acetylglucosaminyltransferases I, II, III and IV were applied to Triton X-100 extracts of wild type BHK, RIC^R14 and RIC^R21 cells. It was shown that RIC^R21 cell extracts had a decreasedN-acetylglucosaminyltransferase II specific activity (17 to 27% of wild type values). It is suggested that in wild type cellsN-acetylglucosaminyltransferase II action proceeds quickly, leading to complexN-glycan synthesis, while in RIC^R21 cells potential substrates forN-acetylglucosaminyltransferase II move into the trans-Golgi compartment before the transferase can act, thereby leading to hybrid structures.