[Frontiers in Bioscience 6, d1235-1244, October 1, 2001]


Gouyer V.1, Leteurtre E. 1, Zanetta J.P.2, Lesuffleur T. 3, Delannoy P. 2, and Huet G. 1

1 Unité INSERM 377, Place de Verdun, 59045 Lille cedex, 2 Unité de Glycobiologie structurale et fonctionnelle, UMR CNRS n�8576, Laboratoire de chimie biologique, Université des sciences et technologies de Lille, F-59655 Villeneuve d'Ascq, 3Unité INSERM 505, 15 rue de l'école de Médecine, 75006 Paris, France


1. Abstract
2. Introduction
3. Intracellular metabolisation of GalNAca-O-bn
4. Inhibition of several glycosyltransferases in HT-29 cells
5. Morphological changes
6. Effect of GalNAca-O-bn on secreted mucins
6.1. Glycosylation of mucins
6.2. Secretion of mucins
7. Effect of GalNAca-O-bn on membrane glycoproteins
7.1. Glycosylation of membrane glycoproteins
7.2. Intracellular trafficking of membrane glycoproteins
8. Effect of GalNAca-O-bn on lysosomal enzyme processing
9. Perspectives
10. References


To address the function of carbohydrates in mucins, GalNAca-O-bn has been used in in vivo experiments on several human mucosal cultured cells as a potential competitor of the glycosylation of N-acetylgalactosamine residues. GalNAca-O-bn is metabolized by glycosyltransferases expressed in the cell, and give rise to different internal derivatives starting in particular from the formation of the disaccharide Galb1-3GalNAca-O-bn. In this line, GalNAca-O-bn exposure inhibits peripheral glycosylation according a cell-type specific manner. The metabolic alterations are very important in HT-29 cell line, leading to a massive accumulation of GalNAca-O-bn oligosaccharide derivatives and to a strong inhibition of the terminal elongation of O-glycans by a2,3 sialyltransferase ST3Gal I. GalNAca-O-bn treatment also induced alterations at the cellular level, exhibiting a large scale in HT-29 cells, i.e. 1) an inhibition of mucin secretion, 2) a blockade in the targeting of some membrane glycoproteins (brush border glycoproteins such as dipeptidylpeptidase IV, carcinoembryonic antigen and the mucin-like glycoprotein MUC1, and the basolateral cell adhesion molecule CD44), 3) an inhibition in the processing of lysosomal enzymes. Morphological abnormalities have been evidenced in GalNAca-O-bn treated cells, in particular the accumulation of numerous intracellular vesicles in HT-29 cells. Taken together, these data suggest that O-glycosylation might be involved in the regulation of the targeting of O-glycosylproteins through carrier vesicles.


To address the function of carbohydrates, inhibitors of glycosylation are used in in vivo experiments on cultured cells. Different substances are available for inhibiting the processing of N-glycans, such as tunicamycin, deoxynojirimycin, castanospermine, desoxymannojirimycin, swaisonine, etc (for review, 1-5). Similar tools do not exist for the biosynthesis of O-glycans. However, the inhibition of O-glycosylation processes has been obtained through the use of chemically synthesized sugar analogues. Aryl and alkyl O-b-D-xylosides have been developed as competitive substrates for inhibiting the elongation of O-linked xylose on proteoglycans (6). Similarly, aryl-N-acetyl-a-galactosaminides have been synthesized as potential competitors of the elongation of O-linked GalNAc residues on mucins (7, 8). Recently, the synthesis of C-glycoside analogue of benzyl-N-acetyl-a-galactosaminide has been reported (9).


Benzyl-N-acetyl-a-galactosaminide (GalNAca-O-bn) was particularly used on human mucosal cells which can be grown in long-term culture (10), in the aim of studying the function of the O-glycosylation of mucins. Investigations in different cell-types (LS174T, Kato III, Caco-2 and HT-29 cell lines) showed that this derivative comes through the membrane and undergoes intracellular metabolisation (7, 8, 11, 12). The disaccharide Galb1-3GalNAca-O-bn accounts for a main GalNAca-O-bn derivatives synthesized intracellularly, but many other different derivatives were also identified in the different cell types by metabolic labeling. Complex benzyl-oligosaccharides are synthesized inside the cells (up to 6-7 sugars in Caco-2 cells, 8-9 sugars in Kato III cells (11), and 6 sugars in HT-29 cells (13)).

The identification of synthesized GalNAca-O-bn derivatives was investigated in HT-29 cells. In the first experiments by reverse phase HPLC, two GalNAca-O-bn derived oligosaccharides were identified after a time exposure of 24 hours: the disaccharide Galb1-3GalNAca-O-bn and the trisaccharide NeuNeu5Aca2-3Galb1-3GalNAca-O-bn (12). Then, after a permanent exposure, using MALDI-TOF, GC/MS and 1H-NMR analyses, eleven different GalNAca-O-bn derivatives were identified and are shown under a synthesis pattern in Figure 1 (13). They could be divided into three families of compounds, according to the addition of further residues on GalNAc: Galb1-3 and / or GlcNAcb1-6 or NeuNeu5Aca2-6. The simplest benzyl-oligosaccharide was Galb1-3GalNAca-O-bn (compound # 1). As shown in Figure 1, the Galb1-3 branch could be substituted either by another Galb1-3 residue (compound # 2) or by a Neu5Aca2-3 residue (compounds # 9-a, 10,

11). In addition, GalNAc residue of Galb1-3GalNAca-O-bn motif could also be substituted by a GlcNAcb1-6 (compound # 4). This GlcNAc residue constituted the starting point of a series of compounds, firstly always comprising an additional Galb1-4 (compound # 6) that could be sialylated by Neu5Aca a2-3 (compounds # 9-b, 10 and 11). GlcNAc residue could be itself substituted by Fuca1-3 (compounds # 9-b and 11), forming sialyl-Lex motif. Compound # 3 was unambiguously identified as Galb1-4GlcNAcb1-6GalNAca-O-benzyl, a member of the core-6 series of O-glycans.

The structure of the oligosaccharide side chains of the mucins synthesized by mucin-secreting HT-29 cells (HT-29 MTX subpopulation, 14) have been determined and the major structure found was the trisaccharide, sialyl T, i.e. Neu5Aca2-3Galb1-3GalNAc. It may be stressed that this structure was also found on GalNAca-O-bn, i.e. compound # 5, and furthermore, this compound accounted for a derivative synthesized in a relatively high amount. Consequently, GalNAca-O-bn carries oligosaccharide structures similar as in mucins. However, complex fucosylated GalNAca-O-bn derivatives are also synthesized in differentiated HT-29 cells (of enterocytic phenotype, HT-29 G- cells, (13) and also of mucin-secreting phenotype, HT-29 MTX cells, unpublished data) despite the fact that no fucosylated oligosaccharide structure was detected in the mucins of HT-29 MTX cells (15). Furthermore, among all the GalNAca-O-bn derivatives, only the oligosaccharide structures of the compounds # 5, 8, 9a and 10 were previously found in HT-29 MTX mucins. This observation finally shows that the elongation of GalNAca-O-bn does not strictly reflects the elongation of GalNAc residues linked to apomucins, and suggests that the specificity of glycosylation of mucins is also relevant to the peptide backbone of mucins.

The amount of GalNAca-O-bn derivatives accumulated in HT-29 cells is very high: 1 mg hexose / mg protein. In contrast, no accumulation of either GalNAca-O-bn derivative occurred in Caco-2 cells (13). This observation indicated a lower rate of biosynthesis and/or a higher rate of degradation of GalNAca-O-bn derivatives in Caco-2 cells than in HT-29 cells. Obviously, the fate of GalNAca-O-bn can be very different according to the cell type.


Figure 1 shows the possible pathway for the biosynthesis of the different GalNAca-O-bn oligosaccharide derivatives by the known glycosyltransferases. It was interesting to note that the biosynthesis of GalNAca-O-bn oligosaccharides was perfectly in agreement with the commonly described O-glycan biosynthesis pathways and with the substrate specificity determined for the different enzymes in in vitro assays (16). GalNAca-O-bn treatment could be therefore a convenient tool to determine the O-glycosylation pattern of cultured cells. All the glycosyltransferases involved in the synthesis of these derivatives (Table 1) might thus be potentially inhibited for the substitution of endogenous substrates.

As previously mentioned, GalNAca -O-bn constitutes a very good substrate for UDP-Gal : GalNAc b1,3 galactosyltransferase (core-1 b 3-Gal-T) in different types of cultured cells, including HT-29 cells (7, 8, 12). A cDNA sequence encoding a human core-1 b 3-Gal-T recently appeared in the databases (Ju, unpublished data) but the enzyme has not been precisely characterized to date. The formed Galb 1-3GalNAca -O-bn also behaves as an acceptor substrate for several other glycosyltransferases: Galb 1-3GalNAc a 2,3-sialyltransferases : ST3Gal I (17) and ST3Gal II (18, 19) and core-2 b 1,6-N-acetyl-glucosaminyltransferase I (20) and II (21). The oligosaccharide structure Neu5Aca 2-3Galb 1-3-GalNAc (compound # 5), found in a high amount, is synthesized through the catalytic action of ST3Gal I which is highly expressed in HT-29 cells (15). Furthermore, some of these firstly processed compounds are further elongated. For instance, Neu5Aca 2-3Galb 1-3GalNAca -O-bn can be substituted by Neu5Aca 2-3Galb 1-3GalNAc (sialyl to GalNAc) a 2,6-sialyltransferases ST6GalNAc III (22) and IV (23), leading to the synthesis of the disialylated form of T-antigen (compound # 9-a). In this connection, neither Neu5Aca 2-6GalNAca-O-bn (sialyl-Tn) nor Galb 1-3[Neu5Aca 2-6]GalNAca-O-bn were observed. This observation fits well with our knowledge of the substrate specificity of ST6GalNAc I (24) and ST6GalNAc II (25), the sialyltransferases involved in the biosynthesis of these structures, since both require a peptide aglycon to be active. GlcNAcb1-6[Galb1-3]GalNAca-O-bn (compound # 4) can be further elongated onto GlcNAcb 1,6-branch by the sequential addition of a b 1,4- linked Gal residue (compound # 6) and of a a 2,3-linked Neu5Ac residue (compound # 8) transferred by one of the human b 1,4-galactosyltransferases (26) and a Galb 1-4GlcNAc a 2,3-sialyltransferase, respectively. Two different Galb 1-4GlcNAc a 2,3-sialyltransferases: ST3Gal IV (27, 28) and VI (29) have been characterized and we previously showed that ST3Gal IV is highly expressed in HT-29 cells (15). Compound # 8 is further the precursor of compounds # 9-b (the sialyl-LewisX epitope) and # 10. These two compounds reflect the catalytic action of either a Neu5Aca 2-3Galb 1-4GlcNAc a 1,3-fucosyltransferase, probably Fuc-TVII which specifically synthesizes the sialyl-Lewisx epitope (30, 31), or ST3Gal I, respectively. It is clear that in HT-29 cells the biosynthesis of sialyl-LewisX determinant requires the a 2,3-sialylation of Galb 1-4GlcNAc terminal sequence prior the a 1,3-fucosylation (32). Finally, the more complex GalNAca -O-bn derivative (compound # 11) which contains both a sialyl-Lewisx epitope onto the b 1,6-branch and a a 2,3-linked Neu5Ac residue to the terminal Gal of core 1, reflects the action of both ST3Gal I and Fuc-TVII. In parallel with that main biosynthetic pathway of GalNAca -O-bn derivatives, several other minor compounds were observed. As an example, compound # 6 may be used as an acceptor substrate by an a 3-fucosyltransferase different from Fuc-TVII leading to compound # 7-a which contains a Lewisx epitope onto the b 1,6-branch. Several a 3-fucosyltransferases such as Fuc-TIV, Fuc-TV, Fuc-TVI (33), or the newly cloned Fuc-TIX (34) could be responsible for the biosynthesis of this compound.

For some compounds (# 2, 3, and 7b), the biosynthesis pattern could not be clearly defined with the known human glycosyltransferases. Compounds # 1 and 6 apparently serve as acceptor substrates for a putative b1,3-galactosyltransferase to generate the compounds # 2 and 7-b, respectively. However the Galb 1-3Galb 1- terminal sequence has never been observed in oligosaccharides from mammalian mucins and no UDP-Gal : Galb 1-3GalNAc b 1,3galactosyltransferase has been detected in mammalian tissues or cells till now (16). But it can be hypothesized that compound #2 is formed by the action of the b 1,3-galactosyltransferase acting on glycosaminoglycans. Surprisingly, even if most compounds described here derive from Galb 1-3GalNAca -O-bn, we could identify the structure of Galb1-4GlcNAcb1-6GalNAca1-O-bn (compound # 3) at a significant level. This compound could correspond to a degradation of compound # 6 by the action of a specific galactosidase, all the more than the disaccharide GlcNAcb1-6GalNAca-O-bn could not be detected.

In conclusion, GalNAca-O-bn treatment of HT-29 cells can potentially inhibit many glycosyltransferases involved in the biosynthesis of O- and / or N-glycans (Table 1).


Morphological changes were observed after a long time exposure to GalNAca-O-bn. In LS174T cells, the formation of intercellular cysts was detected from day 3 of exposure, increased thereafter and reached a steady number after one week of exposure (7). The appearance of pleomorphic vacuoles and an increase in lysosomes was also observed. Strong morphological changes were found after permanent exposure of HT-29 cells to

GalNAca-O-bn: a dramatic swelling of the cells, and an accumulation of numerous intracytoplasmic vesicles (Figure 2 a, b) (35). In contrast, such changes were not observed for Caco-2 cells. Taken together, these observations evidenced the cell-type specificity of the cell reaction to GalNAca-O-bn. Differences in the structure and / or level of accumulated GalNAca-O-bn metabolites may account for the specific behavior regarding cell morphology. In HT-29 cells, several data support the idea that the GalNAca-O-bn oligosaccharides accumulate inside the numerous intracytoplasmic vesicles: i) these derivatives are recovered in the glycolipid phase of Folch extraction (13), ii) the presence of some glycan epitopes of GalNAca-O-bn derivatives could be shown in intracytoplamic vesicles (unpublished data).


6.1. Glycosylation of mucins

First experiments on mucin producing cancer cell lines were carried out by Kuan et al. (7). Mucin labeling with [3H] ]-glucosamine revealed that a 24 hour exposure to benzyl-, phenyl-, and p-nitrophenyl-N-acetyl-a-galactosaminide inhibited the glycosylation of mucins in LS174T cells (7). GalNAca-O-bn was further shown to inhibit mucin glycosylation in other cell lines (Kato III, Caco-2 and HT-29) after 24 to 48 hours incubation, and studies were undertaken in order to bring information about the alterations induced in these different cell lines, using chemical analysis and / or immunoassays with lectins and glycan specific antibodies (8, 11, 36, 15). In all cell lines, GalNAca-O-bn treatment induced an increase in the expression of Tn antigen (GalNAc-Thr / Ser), but also, an increase in the expression of T antigen (Galb1-3GalNAc). The latter observation was surprising because GalNAca-O-bn was initially used as a competitive inhibitor of the activity of core-1 b3-Gal-T through the formation of the disaccharide Galb1-3GalNAca-O-bn. In fact, GalNAca-O-bn does not efficiently inhibited core-1 b3-Gal-T in vivo on the different cultured cells because the activity of this enzyme is very high in these cells and can transfer a Gal residue on both the endogenous (mucins) and exogenous (GalNAca-O-bn) substrates. In contrast, the high amount of the formed disaccharide Galb1-3GalNAca-O-bn acts as an efficient inhibitor of the elongation of the T antigen structures of mucins (8). In parallel to the increased expression of core region carbohydrate antigens Tn and T, GalNAca-O-bn was found to decrease the expression of peripheral carbohydrate antigens, this time according a cell line-specific manner: sialyl Lex and sulfomucin in LS174T cells (8), sialyl Lex in Kato III cells (11), terminal fucose (e.g. H-antigen) in Caco-2 cells (11), and sialyl T antigen in HT-29 MTX cells (36). This specific effect of GalNAca-O-bn upon peripheral glycosylation is likely dependent upon the glycosyltransferase expression pattern in each cell type.

A chemical analysis of mucins synthesized in presence of GalNAca-O-bn was investigated in mucin-secreting HT-29 MTX cells. This analysis was carried out after a short time exposure (24h) on differentiated cells at a late post-confluence state (day 21) (35), and also after a permanent exposure of cells from their undifferentiated state (day 2 after seeding) up to late confluence (day 21) (15). Carbohydrate compositions are shown as molar ratios to GalNAc (Table 2) in comparison to the carbohydrate composition of mucins of control cells. The short time exposure to GalNAca-O-bn particularly induced a dramatic decrease (by 13-fold) in the relative amount of sialic acid. No change was found in the relative amount of Gal residues. This result showed that the sialylation of mucins was primarily affected in this condition. After the permanent exposure, the decrease in the sialic acid content was not so marked (by 2.6-fold). In addition, a decrease in the content of Gal residues became apparent. Consequently, in the permanent exposure, the inhibition of sialylation was lower, and an inhibition of galactosylation occurred.

Structural investigations have determined that oligasaccharides of HT-29 MTX mucins consist of short sialylated structures (two to seven residues), mainly of core types 1 (57.5 %) and 2 (20.5 %), and that incorporation of sialic acid occurred primarily via an a2,3-linkage to a terminal Gal residue. The major structure was the monosialylated trisaccharide of core type 1, i.e. Neu5Aca2-3Galb1-3GalNAc (41 %) (15). Consequently, the inhibition of the incorporation of Gal and/or Neu5Ac residues in HT-29 MTX mucins by GalNAca-O-bn might be related to an inhibition of core-1 b3-Gal-T and / or ST3Gal I glycosyltransferases.

6.2. Secretion of mucins

An inhibition of mucin secretion was found in mucin-secreting HT-29 MTX cells, but not in other cell lines (7, 8, 11, 15, 36). Metabolic labeling experiments with [3H] threonine on differentiated HT-29 MTX cells over 24 hours in presence of GalNAca-O-bn showed that the secretion of mucins was inhibited by 51 % (36). Analysis of mucin secretion after permanent exposure to GalNAca-O-bn was investigated in HT-29 MTX cells by metabolic labeling from day 3 to day 21. From day 7 onwards, control HT-29 MTX cells began to secrete high amounts of mucins. Maximal secretion was observed at days 9-11 and then the level of secreted mucins decreased progressively from day 13 to day 21. In contrast, the mucin secretion of treated cells remained at a very low level at day 7 and even gradually decreased up to day 21.

Regarding the strong inhibition of the constitutive secretion of mucins in HT-29 MTX cells, the effect of GalNAca-O-bn upon the secretory response was then examined using the combined action of Ca2+ ionophore and PMA, or of Ca2+ ionophore and forskolin. Whatever the combination of secretagogues used, the secretory response of GalNAca-O-bn treated cells was dramatically reduced in comparison to control cells. In line with the knowledge that MUC5AC is the major gene expressed in HT-29 MTX cells, the inhibition of both constitutive and secretagogue-induced MUC5AC secretion was specified by Western blot and ELISA experiments using an anti-apomucin (MUC5AC) antibody (15).

These results showed that, although the glycosylation of mucins was altered in the different cell-types, their secretion appeared only altered in HT-29 cells. This observation could be relevant to the cell-type specific effect of GalNAca-O-bn upon peripheral glycosylation of mucins and / or the particular accumulation of GalNAca-O-bn derivatives in the HT-29 cell-type.


7.1. Glycosylation of membrane glycoproteins

In addition to secreted mucins, the effect of GalNAca-O-bn exposure upon the glycosylation of other glycoproteins was also analyzed particularly for brush-border associated glycoproteins and cell adhesion molecules in HT-29 cells. Different brush-border associated glycoproteins, i.e. dipeptidylpeptidase-IV (DPP-IV), the mucin-like glycoprotein MUC1, and carcinoembryonic antigen (CEA), and basolateral glycoproteins i.e. integrins, gp525, and CD44 were examined for their glycosylation (35, 37, 38). All these glycoproteins were found contain a terminal sialylation by a2,3-linked sialic acid. After permanent GalNAca-O-bn treatment, this terminal a2-,3-sialylation was found to be inhibited for some glycoproteins (DPP-IV, MUC1, CEA, CD44) but not for others (integrin b4 and a6,

gp525). Among the glycoproteins with inhibited sialylation, the appearance of T antigen expression could be shown for three glycoproteins (DPP-IV, MUC1, and CD44). This antigen is a structure specifically O-linked to threonine or serine residues, and its substitution by a2,3-linked sialic acid is only obtained through the enzymatic activity of ST3Gal I. These observations evidenced that GalNAca-O-bn treatment primarily inhibited the elongation of O-glycans by ST3Gal I in HT-29 cells. This conclusion is further supported by the observation that ST3Gal I is also highly expressed in enterocytic HT-29 cells as in mucus-secreting HT-29 cells. In contrast, N-glycosylated glycoproteins appeared to show variable sensitivity to GalNAca-O-bn treatment. Indeed, we previously mentioned that sialytransferases involved in the biosynthesis of N-glycans could be also competitively inhibited by metabolites of GalNAca-O-bn, and ST3Gal IV is also a sialyltransferase highly expressed in HT-29 cells.

The effect of GalNAca-O-bn was also studied in other cell types using short-time exposure. Alfalah et al. (39) reported an inhibition of O-glycosylation of the brush border glycoprotein sucrase-isomaltase in Caco-2 cells, Nakano et al (40) an inhibition of the sialylation of CD44 in B16BL6 melanoma cells, and Ait Slimane et al. (41) an inhibition of the sialylation of a soluble mouse DPP-IV form in Caco-2 and MDCK cells. However as for secreted mucins, the changes in the glycosylation of membrane glycoproteins are also dependant upon the glycosyltransferase expression pattern of the cell (38).

7.2. Intracellular trafficking of membrane glycoproteins

In addition to the inhibition of mucin secretion in HT-29 MTX cells, we showed that GalNAca-O-bn also blocked the apical targeting of several membrane brush border glycoproteins: a transmembrane glycoprotein (DPP-IV) (Figure 2 c), a glycoprotein anchored through glycosylphosphatidylinositol moiety (CEA), and a mucin-like membrane glycoprotein (MUC1) in differentiated HT-29 cells of mucin-secreting or enterocytic phenotype (35, 38). These glycoproteins no more reached their membrane apical localization, but remained stored in the cytoplasm. This finding raised the hypothesis of a role of Neu5Aca2-,3- glycosylation in the apical targeting in differentiated HT-29 cells.

In the line of such hypothesis, no change was found for the basolateral distribution of gp120 (35) and gp525 in GalNAca-O-bn treated HT-29 cells (38). In contrast, we recently reported an alteration of the basolateral distribution of CD44. This different behavior for CD44 might be related to the fact that CD44 is both N- and O- glycosylated, whereas gp120 and gp525 are only N-glycosylated. Indeed, as mentioned in the previous subsection, GalNAca-O-bn was found to inhibit the sialylation of CD44 but not that of gp525 and of integrins (37). These data might suggest a role of O-glycans in intracellular trafficking of O-glycosylproteins. In this line, O-glycosylated stalk region of sucrase-isomaltase was recently reported to be involved in prerequisite determinants for apical targeting via sphingolipid-cholesterol microdomains (42).

In Caco-2 cells, GalNAca-O-bn was also reported to induce a selective alteration in the apical targeting of the brush border glycoproteins sucrase-isomaltase (39) and DPP-IV (41). However, we recently showed that long-term GalNAca-O-bn treatment of Caco-2 cells did not induce such a massive alteration in the intracellular trafficking as in HT-29 cells. In particular, no intracellular accumulation of glycoproteins occurred, but only a reduced expression at the apical cell surface (38). This observation can be related to the differences induced by GalNAca-O-bn treatment in the glycosylation of membrane glycoproteins between the two cell-types and / or the specific accumulation of GalNAca-O-bn derivatives in HT-29 cells.


Recent work showed that the processing of lysosomal enzymes was altered in GalNAca-O-bn treated HT-29 MTX cells (37). In particular, a-glucosidase was shown to accumulate as an immature molecular species in a post-trans-Golgi network (TGN) compartment. Incomplete maturation of the enzyme procathepsin D was also found. All these data suggested an alteration in the intracellular trafficking via endosomal/lysosomal compartments. Such hypothesis agrees with the inhibition of the degradation of GalNAca-O-bn derivatives in these cells (13).


Altogether, these data show that GalNAca-O-bn treatment differently affects the glycosylation of secreted mucins and membrane glycoproteins according to the cell type. Finally, this derivative does not behave primarily as an inhibitor of the elongation of O-linked GalNAc residues as previously expected, but as an inhibitor of peripheral glycosylation, particularly concerning the elongation of T antigen Galb1-3GalNAc. The changes in peripheral glycosylation are related to the cell expression pattern of glycosyltransferases and are mediated through the intracellular synthesis of GalNAca-O-bn oligosaccharides by glycosyltransferases. The HT-29 cell-type is peculiar in this latter field, being able to synthesize and accumulate a series of complex GalNAca-O-bn oligosaccharide derivatives.

At the cellular level, the changes induced by GalNAca-O-bn also deeply differ according to the cell type. Generally, alterations in the intracellular trafficking occurred, particularly concerning the apical targeting, suggesting that some steps could be regulated by O-glycosylation. But it must be stressed that the HT-29 cell reaction is spectacular. The strong alterations in the intracellular trafficking are very likely in close relation to the presence of numerous vesicles in the cytoplasm. A first possible mechanism to explain the specific phenotype acquired by HT-29 cells would be a blockade in the advance of carrier vesicles coming from the trans-Golgi network. This would lead to an intracellular accumulation of mistargeted glycoproteins in cellular compartments probably in the route of late endosomes/lysosomes. As another possible mechanism, a blockade in the recycling of endocytosed glycoproteins was suggested (37). Finally, the characterization of accumulated vesicles will help to elucidate the mechanisms by which GalNAca-O-bn induces the expression of such phenotype in HT-29 cells.


1. R.F. Green, H.K. Meiss & E. Rodriguez-Boulan: Glycosylation does not determine segregation of viral envelope proteins in the plasma membrane of epithelial cells. J. Cell Biol. 89, 230-239 (1981)

2. A. D. Elbein, Y. T. Pan, R. Solf & K. Vosbeck: Effect of swainsonine, an inhibitor of glycoprotein processing, on cultured mammalian cells. J. Cell. Physiol. 115, 265-275 (1983)

3. A. D. Elbein, J. E. Tropea, M. Mitchell & G. P. Kaushal: Kifunensine, a potent inhibitor of the glycoprotein processing mannosidase I. J. Biol. Chem. 265, 15599-15605 (1990)

4. R. Kornfeld & S. Kornfeld: Assembly of asparagine-linked oligosaccharides. Annu. Rev. Biochem. 54, 631-664 (1985)

5. V. Duronio, S. Jacobs, P.A. Romero & A. Herscovics: Effects of inhibitors of N-linked oligosaccharide processing on the biosynthesis and function of insulin and insulin-like growth factor-I receptors. J. Biol. Chem. 263, 5436-5445 (1988)

6. M. Sobue, H. Habuchi, K. Ito, H. Yonekura, K. Oguri, K. Sakurai, S. Kamohara, Y. Ueno, R. Noyori & S. Suzuki: b beta-D-xylosides and their analogues as artificial initiators of glycosaminoglycan chain synthesis. Aglycone-related variation in their effectiveness in vitro and in ovovivo. Biochem. J, 241, 591-601 (1987)

7. S.F. Kuan, J.C. Byrd, C. Basbaum & Y.S. Kim: Inhibition of mucin glycosylation by aryl-N-acetyl-a -galactosaminides in human colon cancer cells. J. Biol. Chem. 264, 19271-19277 (1989)

8. J. Huang, J.C. Byrd, W.H. Yoor & Y.S. Kim: Effect of benzyl-a -GalNAc, an inhibitor of mucin glycosylation, on cancer-associated antigens in human colon cancer cells. Oncol. Res. 4, 507-515 (1992)

9. R. San Martin, B. Tavassoli, K.E. Walsh, D.S. Walter, & T. Gallagher: Radical-mediated synthesis of a -C-glycosides based on N-acylgalactosamine. Org. lett. 2, 4051-4054 (2000)

10. A. Zweibaum, M. Laburthe, E. Grasset & D. Louvard: Use of cultured cell lines in studies of intestinal cell differentiation and function, in " Intestinal Absorption and Secretion." Handbook of physiology. Vol. 4 (Field, M. and Frizzel, R.A., eds.), American Physiological Society, Bethesda, MD, 223-255 (1991)

11. J.C. Byrd, R. Dahiya, J. Huang and Y.S. Kim: Inhibition of mucin synthesis by benzyl-alpha-GalNAc in KATO III gastric cancer and Caco-2 colon cancer cells. Eur. J. Cancer 9, 1498-1505 (1995)

12. P. Delannoy, I. Kim, N. Emery, C. De Bolos, A Verbert, P. Degand & G. Huet: Benzyl-N-acetyl-a -D-galactosaminide inhibits the sialylation and the secretion of mucins by a mucin secreting HT-29 cell subpopulation. Glycoconj. J. 13, 717-726 (1996)

13. J.P. Zanetta, V. Gouyer, E. Maes, A. Pons, B. Hemon, A. Zweibaum, P. Delannoy & G. Huet: Massive in vitro synthesis of tagged oligosaccahrides in 1-benzyl-2-acetamido-2-deoxy-a-D-galactopyranoside treated HT-29 cells. Glycobiology 10, 565-575 (2000)

14. T. Lesuffleur, A. Barbat, E. Dussaulx & A. Zweibaum: Growth adaptation to methotrexate of HT-29 human colon carcinoma cells is associated with their ability to differentiate into columnar absorptive and mucus-secreting cells. Cancer Res. 50, 6334-6343 (1990)

15. S. Hennebicq-Reig, T. Lesuffleur, C. Capon, C. de Bolos, I. Kim, O. Moreau, C. Richet, B. Hémon, M-A. Recchi, E. Maës, J.P. Aubert, F. X. Real, A. Zweibam, P. Delannoy, P. Degand, & G. Huet: Permanent exposure of mucin-secreting HT-29 cells to benzyl-N-acetyl-a -D-galactosaminide induces abnormal O-glycosylation of mucins and inhibits constitutive and stimulated MUC5AC secretion. Biochem. J. 334, 283-295 (1998)

16. I. Brockhausen. In "Glycoproteins", J. Montreuil, J.F.G. Vliegenthart and H. Scachter, eds, New comprehensive Biochemistry Elsevier Science B.V., Amsterdam 29a, 201-259 (1995)

17. H. Kitagawa & J.C. Paulson: Differential expression of five sialyltransferase genes in human tissues. J. Biol. Chem. 269, 17872-17878 (1994)

18. V. Giordanengo, S. Bannwarth, C. Laffont, V. Van Miegem, A. Harduin-Lepers, P. Delannoy & J.C. Lefebvre: Cloning and Expression of cDNA for the human Galb 1-3GalNAc a 2,3-sialyltransferase type II (ST3Gal II). Eur. J. Biochem. 247, 558-566 (1997)

19. Y.J. Kim, K.S. Kim, S.H. Kim, C.H. Kim, J.H. Ko, I.S. Choe, S. Tsuji & Y.C. Lee: Molecular cloning and expression of human Galb 1,3GalNAc a 2,3-sialytransferase (hST3Gal II). Biochem. Biophys. Res. Commun. 228, 324-327 (1996)

20. M.F.A. Bierhuizen & M. Fukuda: Expression cloning of a cDNA encoding UDP-GlcNAc:Galb 1-3GalNAc-R (GlcNAc to GalNAc) b 1-6GlcNAc transferase by gene transfer into CHO cells expressing polyoma large tumor antigen. Proc. Natl. Acad. Sci. U.S.A., 89, 9326-9330 (1992)

21. J.C. Yeh, E. Ong & M. Fukuda: Molecular cloning and expression of a novel b 1,6-N-acetylglucosaminyltransferase that forms core 2, core 4, and I branches. J. Biol. Chem. 274, 3215-3221 (1999)

22. E.R. Sjoberg, H. Kitagawa, J. Glushka, H. van Halbeek & J.C. Paulson: Molecular cloning of a developmentally regulated N-acetylgalactosamine a 2,6-sialyltransferase specific for sialylated glycoconjugates. J. Biol. Chem. 271, 7450-7459 (1996)

23. A. Harduin-Lepers, D.C. Stokes, W.F.A. Steelant, B. Samyn-Petit, M.A. Krzewinski-Recchi, V. Vallejo-Ruiz, J.P. Zanetta, C. Augé & P. Delannoy: Cloning, Expression and Gene Organisation of a human Neu5Aca 2-3Galb 1-3GalNAc a 2,6-sialyltransferase : hST6GalNAc IV. Biochem. J. 352, 37-48 (2000)

24. Y. Ikehara, N. Kojima, N. Kurosawa, T. Kudo, M. Kono, S. Nishihara, S. Issiki, K. Morozumi, S. Itzkowitz, T. Tsuda, S.I. Nishimura, S. Tsuji & H. Narimatsu: Cloning and expression of a human gene encoding an N-acetylgalactosamine a 2,6-sialyltransferase (ST6GalNAc I): a candidate for synthesis of cancer-associated sialyl-Tn antigens. Glycobiology 9, 1213-1224 (1999)

25. B. Samyn-Petit, M.A. Krzewinski-Recchi, W.F.A. Steelant, P. Delannoy & A. Harduin-Lepers: Molecular cloning and Functional expression of Human ST6GalNAc II. Biochim. Biophys. Acta 1474, 201-211 (2000)

26. N.W. Lo, J.H. Shaper, J. Pevsner & N.L. Shaper: The expanding b 4-galactosyltransferase gene family: messages from the databanks. Glycobiology 8, 517-526 (1998)

27. H. Kitagawa & J.C. Paulson: Cloning of a novel a 2,3-sialyltransferase that sialylates glycoprotein and glycolipid carbohydrate groups. J. Biol. Chem. 269, 1394-1401 (1994)

28. K. Sasaki, E. Watanabe, K. Kawashima, S. Sekine, T. Dohi, M. Oshima, N. Hanai, T. Nishi & M. Hasegawa: Expression cloning of a novel Galb 1-3/1-4GlcNAc a 2,3-sialyltransferase using lectin resistance selection. J Biol Chem. 268, 22782-22787 (1993)

29. T. Okajima, S. Fukumoto, H. Miyazaki, H. Ishida, M. Kiso, K. Furukawa, T Urano & K. Furukawa: Molecular cloning of a novel a 2,3-sialyltransferase (ST3Gal VI) that sialylates type II lactosamine structures on glycoproteins and glycolipids. J. Biol. Chem. 274, 11479-11486 (1999)

30. K. Sasaki, K. Kurata, K. Funayama, M. Nagata, E. Watanabe, S. Ohta, N. Hanai & T. Nishi: Expression cloning of a novel a 1,3-fucosyltransferase that is involved in biosynthesis of the sialyl Lewis x carbohydrate determinants in leukocytes. J Biol Chem. 269, 14730-14737 (1994)

31. S. Natsuka, K.M. Gersten, K. Zenita, R. Kannagi & J.B. Lowe: Molecular cloning of a cDNA encoding a novel human leukocyte a -1,3-fucosyltransferase capable of synthesizing the sialyl Lewis x determinant. J. Biol. Chem. 269, 16789-16794 (1994)

32. A. Harduin-Lepers, M.A. Recchi & P. Delannoy: 1994: the year of sialyltransferases. Glycobiology 5, 741-758 (1995)

33. R. Mollicone, J.J. Candelier, B. Mennesson, P. Couillin, A.P. Venot & R. Oriol: Five specificity patterns of (1-3)-a -L-fucosyltransferase activity defined by use of synthetic oligosaccharide acceptors. Differential expression of the enzymes during human embryonic development and in adult tissues. Carbohydr Res. 228, 265-276 (1992)

34. M. Kaneko, T. Kudo, H. Iwasaki, Y. Ikehara, S. Nishihara, S. Nakagawa, K. Sasaki, T. Shiina, H. Inoko, N. Saitou & H. Narimatsu: a 1,3-fucosyltransferase IX (Fuc-TIX) is very highly conserved between human and mouse; molecular cloning, characterization and tissue distribution of human Fuc-TIX. FEBS Lett. 453, 237-242 (1999)

35. G. Huet, S. Hennebicq-Reig, C. de Bolos, F. Ulloa, T. Lesuffleur, A. Barbat, V. Carrière, I. Kim, F.X. Real, P. Delannoy & A. Zweibaum: GalNAc-a -O-benzyl inhibits NeuAca 2-3 glycosylation and blocks the intracellular transport of apical glycoproteins and mucus in differentiated HT-29 cells. J. Cell Biol. 15, 1311-1322 (1998)

36. G. Huet, I. Kim, C. de Bolos, J.M. Lo-Guidice, O. Moreau, B. Hemon, C. Richet, P. Delannoy, F.X. Real & P. Degand: Characterization of mucins and proteoglycans synthesized by a mucin-secreting HT-29 cell subpopulation. J. Cell Science 108, 1275-1285 (1995)

37. F. Ulloa, C. Franci & FX. Real: GalNAca -O-Benzyl inhibits sialylation of de novo synthesized apical, but not basolateral, sialoglycoproteins and blocks lysosomal enzyme processing in a post-TGN compartment. J. Biol. Chem. 275, 18785-18793 (2000).

38. V. Gouyer, E. Leteurtre, P. Delmotte, W. F.A. Steelant, M.A. Krzewinski-Recchi, J.P. Zanetta, T. Lesuffleur, G. Trugnan, P. Delannoy & G. Huet: Differential effect of GalNAca-O-bn on intracellular trafficking in enterocytic HT-29 and Caco-2 cells : relation with the glycosyltransferase expression pattern. J. Cell Science, in press (2001)

39. M. Alfalah, R. Jacob, u. preuss, k.P. zimmer, h. naim & h.Y. naim: O-linked glycans mediate apical sorting of human intestinal sucrase-isomaltase through association with lipid rafts. Current Biology 9, 593-596 (1999)

40. T. Nakano, T. Matsui & T. Ota: Benzyl-a -GalNAc inhibits sialylation of O-glycosidic sugar chains on CD44 and enhances experimental metastatic capacity in B16BL6 melanoma cells. Anticancer Res. 16, 3577-3584 (1996)

41. T. Ait-Slimane, C. Lenoir, C. Sapin, M. Maurice & G. Trugnan: Apical secretion and sialylation of soluble dipeptidyl peptidase IV are two related events. Exp. Cell. Res. 258, 184-194 (2000)

42. R. Jacob, M. Alfalah, J. Grünberg, M. Obendorf & H.Y. Naim: Structural determinants required for apical sorting of an intestinal brush-border membrane protein. J Biol Chem. 275, 6566-6572 (2000)

43. S. Tsuji, M. Datta & J.C. Paulson: A new systematic nomenclature for Sialyltransferases. Glycobiology 6, n�7, v-vii (1996)

44. R. Oriol, R. Mollicone, A. Cailleau, L. Balanzino & C. Breton: Divergent evolution of fucosyltransferase genes from vertebrates, invertebrates, and bacteria. Glycobiology 9, 323-334 (1999)

45. S.E. Goelz, C. Hession, D Goff, B. Griffiths, R. Tizard, B. Newman G. Chi-Rosso & R. Lobb: ELFT: a gene that directs the expression of an ELAM-1 ligand. Cell 63, 1349-1356 (1990)

46. B.W. Weston, R.P. Nair, R.D. Larsen & J.B. Lowe: Isolation of a novel human a (1,3)fucosyltransferase gene and molecular comparison to the human Lewis blood group a (1,3/1,4)fucosyltransferase gene. Syntenic, homologous, nonallelic genes encoding enzymes with distinct acceptor substrate specificities. J. Biol. Chem. 267, 4152-4160 (1992)

47. B.W. Weston, P.L. Smith, R.J. Kelly & J.B. Lowe: Molecular cloning of a fourth member of a human a (1,3)fucosyltransferase gene family. Multiple homologous sequences that determine expression of the Lewis x, sialyl Lewis x, and difucosyl sialyl Lewis x epitopes. J. Biol. Chem. 267, 24575-24584 (1992)

48. A. Harduin-Lepers, M.A. Krzewinski-Recchi, M. Hebbar, B. Samyn-Petit, V. Vallejo-Ruiz, S. Julien, J.P. Peyrat & P. Delannoy: Sialyltransferases and Breast Cancer. In �Recent Research Developments in Cancer� in press (2001)

Footnotes: The nomenclature of sialyltransferases is based on Tsuji et al. (43), the nomenclature of fucosyltransferases is based on Oriol et al. (44) and the nomenclature of the other enzymes involved in O-glycan biosynthesis is based on Brockhausen (16)

Key Words: Sugar, Carbohydrate, Mucin, Review

Send correspondence to: Dr G. Huet, Unité INSERM 377, Place de Verdun, 59045 Lille cedex, Tel: 33 3 20 29 88 60, Fax: 33 3 20 53 85 62, E-mail: huet@lille.inserm.fr