Glucokinase expression is regulated by glucose through O-GlcNAc glycosylation

Blood glucose fluctuates with the fasting-feeding cycle. One of the liver's functions is to maintain blood glucose concentrations within a physiological range. Glucokinase (GCK) or hexokinase IV, is the main enzyme that regulates the flux and the use of glucose in the liver leading to a compensation of hyperglycemia. In hepatocytes, GCK catalyzes the phosphorylation of glucose into glucose-6-phosphate. This critical enzymatic reaction is determinant for the metabolism of glucose in the liver which includes glycogen synthesis, glycolysis, lipogenesis and gluconeogenesis. In liver, simultaneous increase of glucose and insulin enhances GCK activity and gene expression, changes its subcellular location and interaction with regulatory proteins. The post-translational O-linked β-N-acetylglucosaminylation (O-GlcNAcylation) acts as a glucose-sensitive modification and is believed to take part in hepatic glucose sensing by modifying key regulatory proteins. Therefore, we aimed to determine whether GCK is modified by O-GlcNAcylation in the liver of mice and investigated the role that this modification plays in regulating GCK protein expression. We demonstrated that endogenous GCK expression correlated with O-GlcNAc levels in the pathophysiological model ob/ob mice. More specifically, in response to the pharmacological inhibition of O-GlcNAcase (OGA) contents of GCK increased. Using the GlcNAc specific lectin succinylated-WGA and click chemistry labeling approaches, we demonstrated that GCK is modified by O-GlcNAcylation. Further, we demonstrated that siRNA-mediated Ogt knock-down not only decreases O-GlcNAc content but also GCK protein level. Altogether, our in vivo and in vitro results demonstrate that GCK expression is regulated by nutrient-sensing O-GlcNAc cycling in liver.

Integrative view of α2,3-sialyltransferases (ST3Gal) molecular and functional evolution in deuterostomes: significance of lineage-specific losses

Sialyltransferases are responsible for the synthesis of a diverse range of sialoglycoconjugates predicted to be pivotal to deuterostomes’ evolution. In this work, we reconstructed the evolutionary history of the metazoan α2,3-sialyltransferases family (ST3Gal), a subset of sialyltransferases encompassing six subfamilies (ST3Gal I–ST3Gal VI) functionally characterized in mammals. Exploration of genomic and expressed sequence tag databases and search of conserved sialylmotifs led to the identification of a large data set of st3gal-related gene sequences. Molecular phylogeny and large scale sequence similarity network analysis identified four new vertebrate subfamilies called ST3Gal III-r, ST3Gal VII, ST3Gal VIII, and ST3Gal IX. To address the issue of the origin and evolutionary relationships of the st3gal-related genes, we performed comparative syntenic mapping of st3gal gene loci combined to ancestral genome reconstruction. The ten vertebrate ST3Gal subfamilies originated from genome duplication events at the base of vertebrates and are organized in three distinct and ancient groups of genes predating the early deuterostomes. Inferring st3gal gene family history identified also several lineage-specific gene losses, the significance of which was explored in a functional context. Toward this aim, spatiotemporal distribution of st3gal genes was analyzed in zebrafish and bovine tissues. In addition, molecular evolutionary analyses using specificity determining position and coevolved amino acid predictions led to the identification of amino acid residues with potential implication in functional divergence of vertebrate ST3Gal. We propose a detailed scenario of the evolutionary relationships of st3gal genes coupled to a conceptual framework of the evolution of ST3Gal functions.

O-GlcNAcylation stabilizes β-catenin through direct competition with phosphorylation at threonine 41

Dysfunctions in Wnt signaling increase β-catenin stability and are associated with cancers, including colorectal cancer. In addition, β-catenin degradation is decreased by nutrient-dependent O-GlcNAcylation. Human colon tumors and colons from mice fed high-carbohydrate diets exhibited higher amounts of β-catenin and O-GlcNAc relative to healthy tissues and mice fed a standard diet, respectively. Administration of the O-GlcNAcase inhibitor thiamet G to mice also increased colonic expression of β-catenin. By ETD-MS/MS, we identified 4 O-GlcNAcylation sites at the N terminus of β-catenin (S23/T40/T41/T112). Furthermore, mutation of serine and threonine residues within the D box of β-catenin reduced O-GlcNAcylation by 75%. Interestingly, elevating O-GlcNAcylation in human colon cell lines drastically reduced phosphorylation at T41, a key residue of the D box responsible for β-catenin stability. Analyses of β-catenin O-GlcNAcylation mutants reinforced T41 as the most crucial residue that controls the β-catenin degradation rate. Finally, inhibiting O-GlcNAcylation decreased the β-catenin/α-catenin interaction necessary for mucosa integrity, whereas O-GlcNAcase silencing improved this interaction. These results suggest that O-GlcNAcylation regulates not only the stability of β-catenin, but also affects its localization at the level of adherens junctions. Accordingly, we propose that O-GlcNAcylation of β-catenin is a missing link between the glucose metabolism deregulation observed in metabolic disorders and the development of cancer.

The hexosamine biosynthetic pathway and O-GlcNAcylation drive the expression of β-catenin and cell proliferation

The short half-life protooncogene β-catenin acquires a remarkable stability in a large subset of cancers, mainly from mutations affecting its proteasomal degradation. In this sense, colorectal cancers (CRC) form a group of pathologies in which early steps of development are characterized by an aberrant expression of β-catenin and an uncontrolled proliferation of epithelial cells. Diet has long been described as an influence in the emergence of CRC, but the molecular events that link metabolic disorders and CRC remain elusive. Part of the explanation may reside in hexosamine biosynthetic pathway (HBP) flux. We found that fasted mice being force-fed with glucose or glucosamine leads to an increase of β-catenin and O-GlcNAcylation levels in the colon. MCF7 cells possessing intact Wnt/β-catenin signaling heavily expressed β-catenin when cultured in high glucose; this was reversed by the HBP inhibitor azaserine. HBP inhibition also decreased the expression of β-catenin in HT29 and, to a lesser extent, HCT116 cells. The same observation was made with regard to the transcriptional activity of β-catenin in HEK293 cells. Inhibition of HBP also blocked the glucose-mediated proliferation capacity of MCF7 cells, demonstrating that glucose affects both β-catenin expression and cell proliferation through the HBP. The ultimate element conducting these events is the dynamic posttranslational modification O-GlcNAcylation, which is intimately linked to HBP; the modulation of its level affected the expression of β-catenin and cell proliferation. In accordance with our findings, we propose that metabolic disorders correlate to CRC via an upregulation of HBP that reverberates on high O-GlcNAcylation levels including modification of β-catenin.

Identification of phosphorylated oligosaccharides in cells of patients with a congenital disorders of glycosylation (CDG-I)

Protein N-glycosylation is initiated by the dolichol cycle in which the oligosaccharide precursor Glc(3)Man(9)GlcNAc(2)-PP-dolichol is assembled in the endoplasmic reticulum (ER). One critical step in the dolichol cycle concerns the availability of Dol-P at the cytosolic face of the ER membrane. In RFT1 cells, the lipid-linked oligosaccharide (LLO) intermediate Man(5)GlcNAc(2)-PP-Dol accumulates at the cytosolic face of the ER membrane. Since Dol-P is a rate-limiting intermediate during protein N-glycosylation, continuous accumulation of Man(5)GlcNAc(2)-PP-Dol would block the dolichol cycle. Hence, we investigated the molecular mechanisms by which accumulating Man(5)GlcNAc(2)-PP-Dol could be catabolized in RFT1 cells. On the basis of metabolic labeling experiments and in comparison to human control cells, we identified phosphorylated oligosaccharides (POS), not found in human control cells and present evidence that they originate from the accumulating LLO intermediates. In addition, POS were also detected in other CDG patients' cells accumulating specific LLO intermediates at different cellular locations. Moreover, the enzymatic activity that hydrolyses oligosaccharide-PP-Dol into POS was identified in human microsomal membranes and required Mn(2+) for optimal activity. In CDG patients' cells, we thus identified and characterized POS that could result from the catabolism of accumulating LLO intermediates.

Differential effects of lobe A and lobe B of the Conserved Oligomeric Golgi complex on the stability of {beta}1,4-galactosyltransferase 1 and {alpha}2,6-sialyltransferase 1

Initially described by Jaeken et al. in 1980, congenital disorders of glycosylation (CDG) is a rapidly expanding group of human multisystemic disorders. To date, many CDG patients have been identified with deficiencies in the conserved oligomeric Golgi (COG) complex which is a complex involved in the vesicular intra-Golgi retrograde trafficking. Composed of eight subunits that are organized in two lobes, COG subunit deficiencies have been associated with Golgi glycosylation abnormalities. Analysis of the total serum N-glycans of COG-deficient CDG patients demonstrated an overall decrease in terminal sialylation and galactosylation. According to the mutated COG subunits, differences in late Golgi glycosylation were observed and led us to address the question of an independent role and requirement for each of the two lobes of the COG complex in the stability and localization of late terminal Golgi glycosylation enzymes. For this, we used a small-interfering RNAs strategy in HeLa cells stably expressing green fluorescent protein (GFP)-tagged β1,4-galactosyltransferase 1 (B4GALT1) and α2,6-sialyltransferase 1 (ST6GAL1), two major Golgi glycosyltransferases involved in late Golgi N-glycosylation. Using fluorescent lectins and flow cytometry analysis, we clearly demonstrated that depletion of both lobes was associated with deficiencies in terminal Golgi N-glycosylation. Lobe A depletion resulted in dramatic changes in the Golgi structure, whereas lobe B depletion severely altered the stability of B4GALT1 and ST6GAL1. Only MG132 was able to rescue their steady-state levels, suggesting that B4GALT1- and ST6GAL1-induced degradation are likely the consequence of an accumulation in the endoplasmic reticulum (ER), followed by a retrotranslocation into the cytosol and proteasomal degradation. All together, our results suggest differential effects of lobe A and lobe B for the localization/stability of B4GALT1 and ST6GAL1. Lobe B would be crucial in preventing these two Golgi glycosyltransferases from inappropriate retrograde trafficking to the ER, whereas lobe A appears to be essential for maintaining the overall Golgi structure.

Overexpression of Man2C1 leads to protein underglycosylation and upregulation of endoplasmic reticulum-associated degradation pathway

Unfolded glycoproteins retained in the endoplasmic reticulum (ER) are degraded via the ER-associated degradation (ERAD) pathway. These proteins are subsequently transported to the cytosol and degraded by the proteasomal complex. Although the sequential events of ERAD are well described, its regulation remains poorly understood. The cytosolic mannosidase, Man2C1, plays an essential role in the catabolism of cytosolic free oligomannosides, which are released from the degraded proteins. We have investigated the impact of Man2C1 overexpression on protein glycosylation and the ERAD process. We demonstrated that overexpression of Man2C1 led to modifications of the cytosolic pool of free oligomannosides and resulted in accumulation of small Man(2-4)GlcNAc(1) glycans in the cytosol. We further correlated this accumulation with incomplete protein glycosylation and truncated lipid-linked glycosylation precursors, which yields an increase in N-glycoprotein en route to the ERAD. We propose a model in which high mannose levels in the cytosol interfere with glucose metabolism and compromise N-glycan synthesis in the ER. Our results show a clear link between the intracellular mannose-6-phosphate level and synthesis of the lipid-linked precursors for protein glycosylation. Disturbance in these pathways interferes with protein glycosylation and upregulated ERAD. Our findings support a new concept that regulation of Man2C1 expression is essential for maintaining efficient protein N-glycosylation.

Molecular phylogeny and functional genomics of beta-galactoside alpha2,6-sialyltransferases that explain ubiquitous expression of st6gal1 gene in amniotes

Sialyltransferases are key enzymes in the biosynthesis of sialoglycoconjugates that catalyze the transfer of sialic residue from its activated form to an oligosaccharidic acceptor. β-Galactoside α2,6-sialyltransferases ST6Gal I and ST6Gal II are the two unique members of the ST6Gal family described in higher vertebrates. The availability of genome sequences enabled the identification of more distantly related invertebrates' st6gal gene sequences and allowed us to propose a scenario of their evolution. Using a phylogenomic approach, we present further evidence of an accelerated evolution of the st6gal1 genes both in their genomic regulatory sequences and in their coding sequence in reptiles, birds, and mammals known as amniotes, whereas st6gal2 genes conserve an ancestral profile of expression throughout vertebrate evolution.

Arginine 469 is a pivotal residue for the Hsc70-GlcNAc-binding property

The members of the 70kDa-heat shock proteins (HSP70) family play numerous fundamental functions in the cell such as promoting the assembly of multimeric complexes or helping the correct folding of nascent proteins to take place. In numerous previous studies we demonstrated that Hsp70 and its constitutive isoform Hsc70 are endowed of a GlcNAc-binding activity. The molecular modeling of the substrate binding domain of Hsc70 and in silico docking experiments using Ser/Thr-O-GlcNAc motifs allowed to define the potential carbohydrate-recognition region and to point out the crucial position of Arg469 as an amino-acid directly interacting with the sugar moiety. We cloned a flagged Hsc70 in a pCMV.SPORT6 vector and we showed that the mutation R469A decreased the GlcNAc-binding property of the chaperone of around 70%. This is the first work reporting the localization of the GlcNAc-binding domain of a member of the HSP70 family.

Dysregulation of the nutrient/stress sensor O-GlcNAcylation is involved in the etiology of cardiovascular disorders, type-2 diabetes and Alzheimer's disease

O-GlcNAcylation is widespread within the cytosolic and nuclear compartments of cells. This post-translational modification is likely an indicator of good health since its intracellular level correlates with the availability of extracellular glucose. Apart from its status as a nutrient sensor, O-GlcNAcylation may also act as a stress sensor since it exerts its fundamental effects in response to stress. Several studies report that the cell quickly responds to an insult by elevating O-GlcNAcylation levels and by unmasking a newly described Hsp70-GlcNAc binding property. From a more practical point of view, it has been shown that O-GlcNAcylation impairments contribute to the etiology of cardiovascular diseases, type-2 diabetes and Alzheimer's disease (AD), three illnesses common in occidental societies. Many studies have demonstrated that O-GlcNAcylation operates as a powerful cardioprotector and that by raising O-GlcNAcylation levels, the organism more successfully resists trauma-hemorrhage and ischemia/reperfusion injury. Recent data have also shown that insulin resistance and, more broadly, type-2 diabetes can be controlled by O-GlcNAcylation of the insulin pathway and O-GlcNAcylation of the gluconeogenesis transcription factors FoxO1 and CRCT2. Lastly, the finding that AD may correspond to a type-3 diabetes offers new perspectives into the knowledge of the neuropathology and into the search for new therapeutic avenues.

Molecular cloning and characterization of the expression pattern of the zebrafish alpha2, 8-sialyltransferases (ST8Sia) in the developing nervous system

Sialyltransferases are Golgi type II transmembrane glycoproteins involved in the biosynthesis of sialylated glycolipids and glycoproteins. These sialylated compounds play fundamental roles in the development of a variety of tissues including the nervous system. In this study, we have molecularly cloned from zebrafish sources, the orthologues of the six human alpha2,8-sialyltransferases (ST8Sia), a family of sialyltransferases implicated in the alpha2-8-mono-, oligo-, and poly-sialylation of glycoproteins and gangliosides and we have analysed their expression pattern in the embryonic zebrafish nervous system, using in situ hybridization. Our results show that all six ST8Sia exhibit distinct and overlapping patterns of expression in the developing zebrafish central nervous system with spatial and temporal regulation of the expression of these genes, which suggests a role for the alpha2-8-sialylated compounds in the developing nervous system.

Protein ubiquitination is modulated by O-GlcNAc glycosylation

During the past two decades, O-GlcNAc modification of cytosolic and nuclear proteins has been intensively studied. Nevertheless, the function of this post-translational modification remains unclear. It has been recently speculated that O-GlcNAc could act as a protective signal against proteasomal degradation, both by modifying target substrates and/or by inhibiting the proteasome itself. In this work, we have investigated the putative relation between O-GlcNAc and the ubiquitin pathway. First, we showed that the level of both modifications increased rapidly after thermal stress but, unlike ubiquitinated proteins, O-GlcNAc-modified proteins failed to be stabilized by inhibiting proteasome function. Increasing O-GlcNAc levels, using glucosamine or PUGNAc, enhanced ubiquitination. Inversely, when O-GlcNAc levels were reduced, using forskolin or glucose deprivation, ubiquitination decreased. Targeted-RNA interference of O-GlcNAc transferase also reduced ubiquitination and moreover halved cell thermotolerance. Finally, we demonstrated that the ubiquitin-activating enzyme E1 was O-GlcNAc modified and that its glycosylation and its interaction with Hsp70 varied according to the conditions of cell culture. Altogether, these results show that O-GlcNAc and ubiquitin are not strictly antagonistic post-translational modifications, but rather that the former might regulate the latter, and also suggest that E1 could be one of the common links between the two pathways.

Hsp70-GlcNAc-binding activity is released by stress, proteasome inhibition, and protein misfolding

Numerous recent works strengthen the idea that the nuclear and cytosolic-specific O-GlcNAc glycosylation protects cells against injuries. We have first investigated O-GlcNAc level and Hsp70-GlcNAc-binding activity (HGBA) behaviour after exposure of HeLa and HepG(2) cells to a wide variety of stresses. O-GlcNAc and HGBA responses were different according to the stress and according to the cell. HGBA was released for almost all stresses, while O-GlcNAc level was modified either upwards or downwards, depending to the stress. Against all expectations, we demonstrated that energy charge did not significantly vary with stress whereas UDP-GlcNAc pools were more dramatically affected even if differences in UDP-GlcNAc contents were not correlated with O-GlcNAc variations suggesting that O-GlcNAc transferase is itself finely regulated during cell injury. Finally, HGBA could be triggered by proteasome inhibition and by L-azetidine-2-carboxylic acid (a proline analogue) incorporation demonstrating that protein misfolding is one of the key-activator of this Hsp70 property.

Discrimination between lumenal and cytosolic sites of deglycosylation in endoplasmic reticulum-associated degradation of glycoproteins by using benzyl mannose in CHO cell lines

Recent studies demonstrated that deglycosylation step is a prerequisite for endoplasmic reticulum (ER)-associated degradation of misfolded glycoproteins. Here, we report the advantages of using benzyl mannose during pulse-chase experiments to study the subcellular location of the deglycosylation step in Chinese hamster ovary (CHO) cell lines. Benzyl mannose inhibited both the ER-to-cytosol transport of oligomannosides and the trimming of cytosolic-labeled oligomannosides by the cytosolic mannosidase in vivo. We pointed out the occurrence of two subcellular sites of deglycosylation. The first one is located in the ER lumen, and led to the formation of Man8GlcNAc2 (isomer B) in wild-type CHO cell line and Man4GlcNAc2 in Man-P-Dol-deficient cell line. The second one was revealed in CHO mutant cell lines for which a high rate of glycoprotein degradation was required. It occurred in the cytosol and led to the liberation of oligosaccharides species with one GlcNAc residue and with a pattern similar to the one bound onto glycoproteins. The cytosolic deglycosylation site was not specific for CHO mutant cell lines, since we demonstrated the occurrence of cytosolic pathway when the formation of truncated glycans was induced in wild-type cells.

Endoplasmic reticulum-associated degradation of glycoproteins bearing Man5GlcNAc2 and Man9GlcNAc2 species in the MI8-5 CHO cell line

Endoplasmic reticulum-associated degradation of newly synthesized glycoproteins has been demonstrated previously using various mammalian cell lines. Depending on the cell type, glycoproteins bearing Man9 glycans and glycoproteins bearing Man5 glycans can be efficiently degraded. A wide variety of variables can lead to defective synthesis of lipid-linked oligosaccharides and, therefore, in mammalian cells, species derived from Man9GlcNAc2 or Man5GlcNAc2 are often recovered on newly synthesized glycoproteins. The degradation of glycoproteins bearing these two species has not been studied. We used a Chinese hamster ovary cell line lacking Glc-P-Dol-dependent glucosyltransferase I to generate various proportions of Man5GlcNAc2 and Man9GlcNAc2 on newly synthesized glycoproteins. By studying the structure of the soluble oligomannosides produced by degradation of these glycoproteins, we demonstrated the presence of a higher proportion of soluble oligomannosides originating from truncated glycans, showing that glycoproteins bearing Man5GlcNAc2 glycans are degraded preferentially.

Identification and functional expression of a second human beta-galactoside alpha2,6-sialyltransferase, ST6Gal II

BLAST analysis of the human and mouse genome sequence databases using the sequence of the human CMP-sialic acid:beta-galactoside alpha-2,6-sialyltransferase cDNA (hST6Gal I, EC2.4.99.1) as a probe allowed us to identify a putative sialyltransferase gene on chromosome 2. The sequence of the corresponding cDNA was also found as an expressed sequence tag of human brain. This gene contained a 1590 bp open reading frame divided in five exons and the deduced amino-acid sequence didn't correspond to any sialyltransferase already known in other species. Multiple sequence alignment and subsequent phylogenic analysis showed that this new enzyme belonged to the ST6Gal subfamily and shared 48% identity with hST6Gal-I. Consequently, we named this new sialyltransferase ST6Gal II. A construction in pFlag vector transfected in COS-7 cells gave raise to a soluble active form of ST6Gal II. Enzymatic assays indicate that the best acceptor substrate of ST6Gal II was the free disaccharide Galbeta1-4GlcNAc structure whereas ST6Gal I preferred Galbeta1-4GlcNAc-R disaccharide sequence linked to a protein. The alpha2,6-linkage was confirmed by the increase of Sambucus nigra agglutinin-lectin binding to the cell surface of CHO transfected with the cDNA encoding ST6Gal II and by specific sialidases treatment. In addition, the ST6Gal II gene showed a very tissue specific pattern of expression because it was found essentially in brain whereas ST6Gal I gene is ubiquitously expressed.

Delineation of the minimal catalytic domain of human Galbeta1-3GalNAc alpha2,3-sialyltransferase (hST3Gal I)

The CMP-Neu5Ac:Galbeta1-3GalNAc alpha2,3-sialyltransferase (ST3Gal I, EC 2.4.99.4) is a Golgi membrane-bound type II glycoprotein that catalyses the transfer of sialic acid residues to Galbeta1-3GalNAc disaccharide structures found on O-glycans and glycolipids. In order to gain further insight into the structure/function of this sialyltransferase, we studied protein expression, N-glycan processing and enzymatic activity upon transient expression in the COS-7 cell line of various constructs deleted in the N-terminal portion of the protein sequence. The expressed soluble polypeptides were detected within the cell and in the cell culture media using a specific hST3Gal I monoclonal antibody. The soluble forms of the protein consisting of amino acids 26-340 (hST3-Delta25) and 57-340 (hST3-Delta56) were efficiently secreted and active. In contrast, further deletion of the N-terminal region leading to hST3-Delta76 and hST3-Delta105 gave also rise to various polypeptides that were not active within the transfected cells and not secreted in the cell culture media. The kinetic parameters of the active secreted forms were determined and shown to be in close agreement with those of the recombinant enzyme already described (H. Kitagawa, J.C. Paulson, J. Biol. Chem. 269 (1994)). In addition, the present study demonstrates that the recombinant hST3Gal I polypeptides transiently expressed in COS-7 cells are glycosylated with complex and high mannose type glycans on each of the five potential N-glycosylation sites.

Expression of a membrane-bound form of Trypanosoma cruzi trans-sialidase in baculovirus-infected insect cells: a potential tool for sialylation of glycoproteins produced in the baculovirus-insect cells system

A chimeric protein containing the catalytic domain of Trypanosoma cruzi trans-sialidase, the transmembrane domain of the major envelope glycoprotein of the baculovirus (gp67), and the signal peptide of ecdysteroid glucosyltransferase of the baculovirus was expressed under the control of the very late promoter p10 in baculovirus-infected lepidopteran cells. The recombinant protein was found to be enzymatically active. Three days after infection, equal amounts of activity were found associated to the plasma membrane and in the infection medium, both forms having the same apparent molecular weight and being N-glycosylated. When exogenous galactosylated acceptors (lactose or asialo-alpha1-acid glycoprotein) were added in the culture medium of cells infected with the recombinant baculovirus in the presence of a sialylated donor, a sialylation could be observed. Therefore, we propose the use of trans-sialidase as a potential tool for sialylation of glycoconjugates in the baculovirus-insect cells system.

N-glycan structure of a short-lived variant of ribophorin I expressed in the MadIA214 glycosylation-defective cell line reveals the role of a mannosidase that is not ER mannosidase I in the process of glycoprotein degradation

A soluble form of ribophorin I (RI(332)) is rapidly degraded in Hela and Chinese hamster ovary (CHO) cells by a cytosolic proteasomal pathway, and the N-linked glycan present on the protein may play an important role in this process. Specifically, it has been suggested that endoplasmic reticulum (ER) mannosidase I could trigger the targeting of improperly folded glycoproteins to degradation. We used a CHO-derived glycosylation-defective cell line, MadIA214, for investigating the role of mannosidase(s) as a signal for glycoprotein degradation. Glycoproteins in MadIA214 cells carry truncated Glc(1)Man(5)GlcNAc(2) N-glycans. This oligomannoside structure interferes with protein maturation and folding, leading to an alteration of the ER morphology and the detection of high levels of soluble oligomannoside species caused by glycoprotein degradation. An HA-epitope-tagged soluble variant of ribophorin I (RI(332)-3HA) expressed in MadIA214 cells was rapidly degraded, comparable to control cells with the complete Glc(3)Man(9)GlcNAc(2) N-glycan. ER-associated degradation (ERAD) of RI(332)-3HA was also proteasome-mediated in MadIA214 cells, as demonstrated by inhibition of RI(332)-3HA degradation with agents specifically blocking proteasomal activities. Two inhibitors of alpha1,2-mannosidase activity also stabilized RI(332)-3HA in the glycosylation-defective cell line. This is striking, because the major mannosidase activity in the ER is the one of mannosidase I, specific for a mannose alpha1,2-linkage that is absent from the truncated Man(5) structure. Interestingly, though the Man(5) derivative was present in large amounts in the total protein pool, the two major species linked to RI(332)-3HA shortly after synthesis consisted of Glc(1)Man(5 )and Man(4), being replaced by Man(4 )and Man(3) when proteasomal degradation was inhibited. In contrast, the untrimmed intermediate of RI(332)-3HA was detected in mutant cells treated with mannosidase inhibitors. Our results unambiguously demonstrate that an alpha1,2-mannosidase that is not ER mannosidase I is involved in ERAD of RI(332-)3HA in the glycosylation-defective cell line, MadIA214.

Reciprocal relationship between alpha1,2 mannosidase processing and reglucosylation in the rough endoplasmic reticulum of Man-P-Dol deficient cells

The study of the glycosylation pathway of a mannosylphosphoryldolichol-deficient CHO mutant cell line (B3F7) reveals that truncated Glc(0-3)Man5GlcNAc2 oligosaccharides are transferred onto nascent proteins. Pulse-chase experiments indicate that these newly synthesized glycoproteins are retained in intracellular compartments and converted to Man4GlcNAc2 species. In this paper, we demonstrate that the alpha1,2 mannosidase, which is involved in the processing of Man5GlcNAc2 into Man4GlcNAc2, is located in the rough endoplasmic reticulum. The enzyme was shown to be inhibited by kifunensine and deoxymannojirimycin, indicating that it is a class I mannosidase. In addition, Man4GlcNAc2 species were produced at the expense of Glc1Man5GlcNAc2 species. Thus, the trimming of Man5GlcNAc2 to Man4GlcNAc2, which is catalyzed by this mannosidase, could be involved in the control of the glucose-dependent folding pathway.

Use of inhibitors to characterize intermediates in the processing of N-glycans synthesized by insect cells: a metabolic study with Sf9 cell line

The most frequent type of N-glycan synthesized by lepidopteran Sf9 cells appears to be fucosylated Man3GlcNAc2,and this has been a limitation for a large scale production and utilization of therapeutic glycoproteins in cultured insect cells. The current knowledge of the protein glycosylation pathway derived from structural studies on recombinant glyco-proteins expressed by using baculovirus vectors. In this work we provide more direct evidence for the sequential events occurring in the processing of endogenous N-glycoproteins of noninfected Sf9 cells. By metabolic labeling with radioactive mannose, we characterized the glycan structures which accumulated in the presence of processing inhibitors (castanospermine and swainsonine) and in the presence of an intracellular trafficking inhibitor (monensin). We thus demonstrated that from the glycan precursor Glc3Man9GlcNAc2 to GlcNAcMan5(Fuc)GlcNAc2 intermediate, the processing pathway in Sf9 cells paralleled the one demonstrated in mammalian cells. By using monensin, we demonstrated the formation of Man3(Fuc)GlcNAc2 from GlcNAcMan3(Fuc)GlcNAc2, a reaction which has not been described in mammalian cells. Our results support the idea that the hexosaminidase activity is of physiological relevance to the glycosylation pathway and is Golgi located.

Cytosolic deglycosylation process of newly synthesized glycoproteins generates oligomannosides possessing one GlcNAc residue at the reducing end

Recent studies on the mechanism of degradation of newly synthesized glycoproteins suggest the involvement of a retrotranslocation of the glycoprotein from the lumen of the rough endoplasmic reticulum into the cytosol, where a deglycosylation process takes place. In the studies reported here, we used a glycosylation mutant of Chinese hamster ovary cells that does not synthesize mannosylphosphoryldolichol and has an increased level of soluble oligomannosides originating from glycoprotein degradation. In the presence of anisomycin, an inhibitor of protein synthesis, we observed an accumulation of glucosylated oligosaccharide-lipid donors (Glc3Man5GlcNAc2-PP-Dol), which are the precursors of the soluble neutral oligosaccharide material. Inhibition of rough endoplasmic reticulum glucosidase(s) by castanospermine led to the formation of Glc3Man5GlcNAc2(OSGn2) (in which OSGn2 is an oligomannoside possessing two GlcNAc residues at its reducing end), which was then retained in the lumen of intracellular vesicles. Thus they were protected during an 8 h chase period from the action of cytosolic chitobiase, which is responsible for the conversion of OSGn2 to oligomannosides possessing one GlcNAc residue at the reducing end (OSGn1). In contrast, when protein synthesis was maintained in the presence of castanospermine, glucosylated oligomannosides (Glc1-3Man5GlcNAc1) were recovered in cytosol. Except for monoglucosylated Man5 species, which are potential substrates for luminal calnexin and calreticulin, the pattern of oligomannosides was similar to that observed on glycoproteins. The occurrence in the cytosol of glucosylated species with one GlcNAc residue at the reducing end implies that the deglycosylation process that generates glucosylated OSGn1 from glycoproteins occurs in the cytosol.

'Glyco-deglyco' processes during the synthesis of N-glycoproteins

For the past 15 years, it has appeared increasingly evident that the N-glycosylation process was accompanied by the release of oligomannoside type oligosaccharides. This material is constituted of oligosaccharide-phosphates and of neutral oligosaccharides possessing one GlcNAc (OS-Gn1) or two GlcNAc (OS-Gn2) at the reducing end. It has been demonstrated that oligosaccharide-phosphates originated from the cleavage by a specific pyrophosphatase, of non-glycosylated cytosolic faced oligosaccharide-PP-Dol and chiefly the Man5GlcNAc2-PP-Dol. The Man5GlcNAc2-P, as the main product, is recovered in the cytosolic compartment and is further degraded to Man5GlcNAc1 by as for yet not depicted enzymes. In contrast, OS-Gn2 produced from hydrolysis of oligosaccharide-PP-Dol (presumably as a transfer reaction onto water) when the amount of protein acceptor is limiting, are generated into the lumen of rough endoplasmic reticulum (ER). They are further submitted to processing alpha-glucosidases and rough ER mannosidase and are (mainly as Man8GlcNAc2) exported into the cytosolic compartment. This material is further degraded into a single component, the Man5GlcNAc1: Man alpha 1-2Man alpha 1-2Man alpha 1-3 (Man alpha 1-6)Man beta 1-4GlcNAc by the sequential action of a cytosolic neutral chitobiase followed by cytosolic mannosidase. Furthermore, OS-Gn1 could have a dual origin: on the one hand, they originate from OS-Gn2 by the cytosolic degradation pathway indicated above; on the other hand, we will discuss a possible origin from the degradation or remodeling of newly synthesized glycoproteins. Considered first as a minor phenomenon, these observations have lead to the concept of intracellular oligomannoside trafficking, a process which results from more fundamental phenomena such as the control of the dolichol cycle, and the so-called quality-control of glycoprotein. In this review, we would like to describe the evolution of ideas on the origin, intracellular trafficking and putative roles of these oligomannosides released during the N-glycosylation process. We propose that these early stage 'glyco-deglyco' processes represent a way of control of N-glycosylation and of the fate of N-glycoproteins.

Oligomannosides or oligosaccharide-lipids as potential substrates for rat liver cytosolic alpha-D-mannosidase

We have previously reported the substrate specificity of the cytosolic alpha-D-mannosidase purified from rat liver using Man9GlcNAc, i.e. Man alpha 1-2Man alpha 1-3(Man alpha 1-2Man alpha 1-6)Man alpha 1-6(Man alpha 1-2Man alpha 1-2Man alpha 1-3) Man beta 1-4G1cNAc, as substrate [Grard, Saint-Pol, Haeuw, Alonso, Wieruszeski, Strecker and Michalski (1994) Eur. J. Biochem. 223, 99-106]. Man9 G1cNAc is hydrolysed giving Man5GlcNAc, i.e. Man alpha 1-2 Man alpha 1-2Man alpha 1-3(Man alpha 1-6)Man beta 1-4GlcNAc, possessing the same structure as the oligosaccharide of the dolichol pathway formed in the cytosolic compartment during the biosynthesis of N-glycosylprotein glycans. We study here the activity of the purified cytosolic alpha-D-mannosidase towards the oligosaccharide-diphosphodolichol intermediates formed during the biosynthesis of N-glycans, and also towards soluble oligosaccharides released from the endoplasmic reticulum which are glucosylated or not and possessing at their reducing end either a single N-acetylglucosamine residue or a di-N-acetylchitobiose sequence. We demonstrate that (1) dolichol pyrophosphate oligosaccharide substrates are poorly hydrolysed by the cytosolic alpha-D-mannosidase; (2) oligosaccharides with a terminal reducing di-N-acetylchitobiose sequence are not hydrolysed at all; (3) soluble oligosaccharides bearing a single reducing N-acetylglucosamine are the real substrates for the enzyme. These results suggest a role for alpha-D-mannosidase in the catabolism of glycans released from the endoplasmic reticulum rather than in the regulation of the biosynthesis of asparagine-linked oligosaccharides.

Occurrence of a cytosolic neutral chitobiase activity involved in oligomannoside degradation: a study with Madin-Darby bovine kidney (MDBK) cells

Neutral oligomannosides possessing one GlcNAc (OS-Gn1) and two GlcNAc (Os-Gn2) at the reducing end have been reported to be released during the N-glycosylation process in various biological models. To investigate which enzyme is responsible for OS-Gn1 formation, we used the Madin-Darby bovine kidney (MDBK) cell line which exhibits neither lysosomal chitobiase nor endoglucosaminidase activities. However, these cells produced OS-Gn1 and we showed that a neutral chitobiase is responsible for the transformation of OS-Gn2 into OS-Gn1. Using streptolysin O-permeabilized MDBK cells, we demonstrated that this neutral chitobiase activity is located in the cytosolic compartment and is active on oligomannoside species released during the N-glycosylation process.

Catabolism of glycan moieties of lipid intermediates leads to a single Man5GlcNAc oligosaccharide isomer: a study with permeabilized CHO cells

This paper presents kinetic and structural analyses of oligosaccharide material released during glycosylation in permeabilized Chinese hamster ovary cells incubated with sugar nucleotides. Permeabilized cells released 30 times more oligosaccharide material than metabolically labelled cells, normalized to the amount of labelled glycoprotein acceptor, making this an amenable system for study. Fifteen to forty per cent of the oligosaccharide material released by permeabilized cells was oligosaccharide-phosphate, depending on the nature and amount of the oligosaccharide-lipids synthesized. The oligosaccharide-phosphates released were recovered in the cytosol, and were exclusively Man2Glc-NAc2P and Man5GlcNAc2P, released from oligosaccharide-lipids thought to be facing the cytosol. In contrast, the structures found as neutral oligosaccharide material were similar to those attached to newly synthesized glycoproteins, indicating that the oligosaccharides were subjected to the same processing enzymes whether or not they were protein bound. Importantly, the kinetics of the transfer to protein and the release of free neutral oligosaccharide were parallel, suggesting that the same enzyme was responsible for both processes. Structural analyses demonstrated that the same Man5GlcNAc2 structure was transferred to protein and released as free oligosaccharide. Neutral oligosaccharides were found in both the cytosol and the pellet; however, oligosaccharides with one GlcNAc residue at the reducing end (OS-Gn1) were found exclusively in the supernate. The major neutral oligosaccharide produced after 2 h of metabolic labelling was Man5GlcNAc and it was found in the cytosol.

Release of oligomannoside-type glycans as a marker of the degradation of newly synthesized glycoproteins

The N-glycosylation of proteins is accompanied by the release of soluble oligosaccharide material. Besides oligosaccharide phosphates originating from the cleavage of lipid intermediates, neutral free oligosaccharides represent the major part of this material and are heterogeneous depending on whether the reducing end has one or two N-acetylglucosamine residues. The present study focuses on the intracellular origin of neutral free oligosaccharides in a CHO cell line. Kinetic and pulse-chase experiments clearly indicate that oligosaccharides possessing a chitobiosyl unit are derived from oligosaccharide pyrophosphodolichol, whereas oligosaccharides possessing one N-acetyl-glucosamine residue are derived from newly synthesized glycoprotein. This relationship is confirmed by comparing the glycosylation pattern of lipid donors and glycoproteins with those of neutral free oligosaccharides under various incubation conditions (inhibition of protein synthesis, presence of processing inhibitors, presence or absence of glucose). Degradation of newly synthesized glycoprotein and formation of neutral oligosaccharides with one N-acetylglucosamine residue are inhibited at 16 degrees C but not affected by lysosomotropic agents such as leupeptin or NH4Cl. Together with the fact that the degradation of newly synthesized glycoproteins and the subsequent release of the glycan are recovered in permeabilized cells, these results suggest that this phenomenon occurs in the rough endoplasmic reticulum or in a closely related compartment.

Effect of cell attachment and growth on the synthesis and fate of dolichol-linked oligosaccharides in Chinese hamster ovary cells

The inhibition of cellular processes in suspended anchorage-dependent Chinese hamster ovary (CHO) cell lines and their restoration upon attachment to a solid substrate has been used as a model to study the relationship between oligosaccharide-diphospho-dolichols and their metabolic products (glycoprotein and soluble oligosaccharide material, i.e. oligosaccharide phosphates and neutral oligosaccharides). Using metabolic labelling we demonstrated that suspended cells have a low incorporation rate into lipid intermediates and into glycoproteins. The oligosaccharide-lipid populations are mainly glucosylated and the neutral oligosaccharides have exclusively a chitobiosyl residue at their reducing end. In contrast, monolayer cells exhibit a high incorporation rate into lipid intermediates with a pattern dominated by two species containing either two or five mannose residues, and into glycoproteins with a pattern similar to the one observed for suspended cells (i.e. glucosylated species). In monolayer cells the neutral oligosaccharides possess either one or two GlcNAc residues at their reducing end. The variations in the nature and in the quantity of soluble oligosaccharide material as a function of the cell density reflects regulatory points in the synthesis of N-glycosyl proteins. The first regulatory point could be the control of the quantity of non-glucosylated oligosaccharide-lipids to be channelled toward the glucosylated lipid-donor pool. The level of this donor pool being constant, the oligosaccharide-transferase could utilize oligosaccharide-lipid donors at a constant rate by two different reactions: either transfer onto protein when acceptors are available, or transfer onto water generating neutral oligosaccharides possessing two GlcNAc residues at the reducing end. Another regulatory point would be the degradation of a part of neoglycoproteins leading to the release of neutral oligosaccharides possessing one GlcNAc residue at the reducing end.

Phytobezoar after vagotomy with drainage or resection

Two cases of bezoar formation after vagotomy and drainage are reported. A review of the literature reveals the stomach to be the most common site of bezoar formation after operations involving vagotomy, whereas most bezoars occurring after gastrectomy are found in the small intestine. The reasons for this difference are discussed, as well as the clinical features and treatment of bezoar.