Influence of intracellular nucleotide and nucleotide sugar contents on recombinant interferon-gamma glycosylation during batch and fed-batch cultures of CHO cells.

Both the macroheterogeneity of recombinant human IFN-gamma produced by CHO cells and intracellular levels of nucleotides and sugar nucleotides, have been characterized during batch and fed-batch cultures carried out in different media. Whereas PF-BDM medium was capable to maintain a high percentage of the doubly- glycosylated glycoforms all over the process, mono-glycosylated and non-glycosylated forms increased during the batch culture using SF-RPMI medium. Intracellular level of UTP was higher in PF-BDM all over the batch culture compared to the SF-RPMI process. UDP-Gal accumulated only during the culture performed in PF-BDM medium, probably as a consequence of the reduced UDP-Glc synthesis flux in SF-RPMI medium. When the recombinant CHO cells were cultivated in fed-batch mode, the UTP level remained at a relatively high value in serum-containing RPMI and its titer increased during the fed-phase indicating an excess of biosynthesis. Besides, an accumulation of UDP-Gal occurred as well. Those results all together indicate that UTP and UDP-Glc syntheses in CHO cells cultivated in SF-RPMI medium in batch process, could be limiting during the glycosylation processes of the recombinant IFN-gamma. At last, the determination of the energetic status of the cells over the three studied processes suggested that a relationship between the adenylate energy charge and the glycosylation macroheterogeneity of the recombinant IFN-gamma may exist.

Protein C-mannosylation is enzyme-catalysed and uses dolichyl-phosphate-mannose as a precursor.

C-mannosylation of Trp-7 in human ribonuclease 2 (RNase 2) is a novel kind of protein glycosylation that differs fundamentally from N- and O-glycosylation in the protein-sugar linkage. Previously, we established that the specificity determinant of the acceptor substrate (RNase 2) consists of the sequence -x-x-W, where the first Trp becomes C-mannosylated. Here we investigated the reaction with respect to the mannosyl donor and the involvement of a glycosyltransferase. C-mannosylation of Trp-7 was reduced 10-fold in CHO (Chinese hamster ovary) Lec15 cells, which are deficient in dolichyl-phosphate-mannose (Dol-P-Man) synthase activity, compared with wild-type cells. This was not a result of a decrease in C-mannosyltransferase activity. Rat liver microsomes were used to C-mannosylate the N-terminal dodecapeptide from RNase 2 in vitro, with Dol-P-Man as the donor. This microsomal transferase activity was destroyed by heat and protease treatment, and displayed the same acceptor substrate specificity as the in vivo reaction studied previously. The C-C linkage between the indole and the mannosyl moiety was demonstrated by tandem electrospray mass spectrometry analysis of the product. GDP-Man, in the presence of Dol-P, functioned as a precursor in vitro with membranes from wild-type but not CHO Lec15 cells. In contrast, with Dol-P-Man both membrane preparations were equally active. It is concluded that a microsomal transferase catalyses C-mannosylation of Trp-7, and that the minimal biosynthetic pathway can be defined as: Man -> -> GDP-Man -> Dol-P-Man -> (C2-Man-)Trp.

Trafficking of oligomannosides released during N-glycosylation: a clearing mechanism of the rough endoplasmic reticulum.

The main reaction of N-glycosylation of proteins is the transfer 'en bloc' of the oligosaccharide moieties of lipid intermediates to an asparagine residue of the nascent protein. For the past 15 years, a few laboratories including ours have shown that the process was accompanied by the release of oligosaccharide-phosphates and of neutral oligosaccharides possessing one GlcNAc (OS-Gn(1)) or two GlcNAc (OS-Gn(2)) at the reducing end. The aim of this review is to gather the evidence for the different origins of these soluble oligomannosides, to examine their subcellular location and intracellular trafficking. Furthermore, using Brefeldin A we demonstrated that this released oligomannoside material could be the substrate for the Golgi glycosidases and glycosyltransferases. Indeed, released oligomannoside never reach the Golgi vesicles either because they are directly produced in the cytosol as has been demonstrated for oligosaccharide-phosphates and for neutral oligosaccharides possessing one GlcNAc at the reducing end or because they are actively transported out of the rough endoplasmic reticulum to the cytosol. One of the functions of oligomannoside trafficking between rough endoplasmic reticulum, cytosol and lysosomes could be to prevent these oligosaccharides for competing with glycosylation in the Golgi.

Detection of ectosiallyltransferase activity using whole cells. Correction of misleading results due to the release of intracellular CMP-N-acetylneuraminic acid.

An inhibitory effect due to broken cells is observed when sialyltransferase (CMP-N-acetylneuraminate:D-galactosyl-glycoprotein N-acetylneuraminyltransferase, EC 2.4.99.1) is measured with mixture of intact and homogenized lymphocytes. This intracellular inhibitory factor ib purified and characterized as CMP-N-acetylneuraminic acid (CMP-NeuNAc) by its behavior in various chromatographic and electrophoretic systems and by its susceptibility to CMP-NeuNAc hydrolase. This endogenous CMP-NeuNAc leads to an isotopic dilution of the exogenous labelled CMP-NeuNAc explaining the apparently lower activity of homogenate when compared to whole cells. Consequently, the radioactivity bound to acceptors may not be related to a known number of sialyl residues transferred, calling into question the validity of comparing the incorporation of [14C]NeuNAc by homogenate and whole cells in order to assign sialyltransferase activity to ectoenzyme. A new approach is developed to detect ectoglycosyltransferases with whole cells, taking into account that both intracellular enzymes and endogenous precursor may be introduced by the small percentage of broken cells.

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.

Mechanism of UDP-sugar transport into intracellular vesicles. Occurrence of UDP-GlcNAc/UDP and UDP-Gal/UDP antiports.

The mechanism of translocation of UDP-GlcNAc, UDP-Gal and UDP-Glc into intracellular vesicles has been studied using thymocytes whose plasma membranes have been permeabilized with isotonic ammonium chloride. It has been previously shown that the intracellular vesicles have specific carriers for UDP-GlcNAc and UDP-Gal. We now report that the translocation of these two sugar nucleotides occurs via UDP-GlcNAc/UDP and UDP-Gal/UDP antiports. The entry of UDP-GlcNAc or UDP-Gal into vesicles was specifically dependent on the exit of UDP from these vesicles. In contrast, no antiport mechanism has been recovered with UDP-Glc for which no transport and accumulation into intracellular vesicles were observed.

Transport of CMP-N-glycoloylneuraminic acid into mouse liver Golgi vesicles.

CMP-Neu5Gc has been shown to be transported into mouse liver Golgi vesicles by a specific carrier the characteristics of which were investigated in detail. In the system employed, CMP-Neu5Gc enters the Golgi vesicles within 2 min; transport was saturable with high concentrations of the sugar-nucleotide and was dependent on temperature. A kinetic analysis gave an apparent Km of 1.3 microM and a maximal transport velocity of 335 pmol/mg protein per min. Almost identical values were obtained with CMP-Neu5Ac, under the same incubation conditions. Furthermore, the uptake of CMP-Neu5Gc was inhibited by CMP-Neu5Ac, a substrate analogue. Conversely, the uptake of CMP-Neu5Ac was inhibited by CMP-Neu5Gc to the same extent, leading to the conclusion that the transport of CMP-Neu5Ac and CMP-Neu5Gc is mediated by the same carrier molecule. This transport system for CMP-Neu5Gc involves both CMP and CMP-Neu5Gc since intravesicular CMP induced the entry of external CMP-Neu5Gc.

Permeabilized cells as a way of gaining access to intracellular organelles: an approach to glycosylation reactions.

The selective plasma membrane permeabilization of animal cells is a way of introducing non permeable substrates into the cytoplasmic space. This technique facilitates the introduction of a wide range of labelled precursors and avoids the drawbacks of subcellular fractionation. We review here various physical and chemical methods successfully used in different metabolic studies, and as an example, note the advantages of permeabilized cells in glycosylation studies.

Membrane transport of sugar donors to the glycosylation sites.

The assembly of N-linked glycoproteins in eukaryotic cells begins with the segregation of these molecules within the lumen of intracellular vesicles. Since the sugar nucleotides are cytoplasmic molecules, translocation of the sugar moiety across the membrane appears as a crucial event in the glycoprotein synthesis. This N-glycosylation process occurs in two different cytological sites: in the rough endoplasmic reticulum, the stepwise synthesis of a large lipid-linked oligosaccharide takes place, as well as its transfer to protein; then after trimming the immature glycoprotein is further elongated in the Golgi apparatus. In this paper, a brief review will be given of the present knowledge on the sugar donor transport across the membrane barrier to the glycosylation site. Based upon the transmembrane orientation of oligosaccharide lipid intermediates and on the localization of the glycosyltransferase active sites, the different processes required to translocate the sugar moieties during the preassembly of the dolichyl-pyrophosphate-oligosaccharides will be examined. Combining the different results, obtained in several laboratories, it is suggested that the Man5-GlcNAc2-lipid is synthesized on the cytoplasmic side directly from the sugar-nucleotides and then translocated to the lumenal face where the Glc3-Man9-GlcNAc2-lipid is completed using Man-P-Dol and Glc-P-Dol as transmembrane carriers of these sugars. Concerning the elongation process leading to assembly of the antennae of N-acetyllactosamine type oligosaccharides, specific carriers for sugar nucleotides have been described as Golgi markers. Several authors have characterized such carriers for UDP-Gal, GDP-Fuc, CMP-NeuAc, UDP-GlcNAc and UDP-Glc using microsomal vesicles and similar results have been obtained in our laboratory using plasma membrane permeabilized cells. This carrier-mediated process leads to the formation of an intralumenal pool whose biological significance will be discussed. The translocation process of sugar donors occurring in the rough endoplasmic reticulum via lipid intermediates as well as in the Golgi apparatus via specific carriers would represent a regulation step based on the availability of the substrates for the glycosylation.

Inhibition of glycosyltransferases by bis-(p-nitrophenyl)phosphate: general effect and relation to their membrane integration.

The effect of bis-(p-nitrophenyl)phosphate on various glycosyltransferases involved in protein glycosylation (sialyl-, fucosyl-, galactosyl-, mannosyl- and glucosyltransferases) have been studied using crude enzyme preparations solubilized from rat spleen lymphocytes. Bis-(p-nitrophenyl)phosphate appears as a common inhibitor for every glycosyltransferase reaction utilizing sugar nucleotides as direct donors. In most cases 10 mM inhibitor is sufficient to obtain a 90 per cent inhibition. Kinetic studies achieved with a purified galactosyltransferase preparation reveal that bis-(p-nitrophenyl)phosphate exerts a competitive inhibition towards UDP-galactose binding. Concerning membrane-bound enzymes, the interaction of bis-(p-nitrophenyl)phosphate depends on its accessibility to the enzyme active site. This is shown by the different effect obtained with two UDP-Glc utilizing membrane-bound enzymes : UDP-Glc : phospho-dolichyl glucosyltransferase and UDP-Glc : ceramide glucosyltransferase : the first one not being affected but the second one being markedly inhibited under the same condition, although both are inhibited when the membrane environment is disturbed by detergent. Bis-(p-nitrophenyl)phosphate appears to be a tool to study membrane topology of glycosyltransferases.

Viral neuraminidase and cellular ectosialyltransferase in human lymphoblastoid cells infected with influenza virus.

In human lymphoblastoid cells, infected with an influenza virus, Fowl Plague Virus (FPV), glycoproteins (such as secreted IgM) are hyposialylated, through the action of viral neuraminidase. In this study, the modulation of the cellular ectosialyltransferase activity during viral infection was investigated. This activity was detectable in FPV-infected cells, was shown to be 2.5-fold higher than that of uninfected cells, and to be able to restore, at least partially, the level of sialylation of the cell surface acceptors.

Effect of cyclic nucleotides in vitro assays of rat liver galactosyltransferase.

Experiments with rat liver microsomal galactosyltransferase has been developed to test the effect of cyclic nucleotides on the transfer activity. An overall stimulation is observed when cAMP or cGMP (concentration higher that 10(-6) M) are added to the incubation medium. However, more detailed experiments show that the cyclic nucleotides do not act as direct effectors of the enzyme but present the precursors degradation by the glycosylnucleotide pyrophosphatases.

'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.

Dilatation of Golgi vesicles by monensin leads to enhanced accumulation of sugar nucleotides.

Incubation of mouse thymocytes with 10 microM monensin for 1 hour induces morphological alterations characterized by the extensive dilatation and vacuolization of the Golgi complex. This effect is used to study the transport and utilization of labelled sugar nucleotides into intracellular vesicles by using thymocytes whose plasma membrane has been permeabilized by ammonium chloride treatment. It is demonstrated that monensin stimulates the incorporation of labelled sialyl, fucosyl, galactosyl, and N-acetylglucosaminyl residues. This enhanced incorporation is not due to a direct effect of monensin on glycosyltransferase activities themselves but is a consequence of a higher entry and accumulation of labelled sugar nucleotides in the dilated vesicles.

Retention and degradation of N-glycoproteins in the rough endoplasmic reticulum.

Recent studies have shown that newly synthesized proteins and glycoproteins are submitted to a quality control mechanism in the rough endoplasmic reticulum (ER). In this report we present two models: One model will illustrate a transient retention in rough ER leading to a further degradation of glycoproteins in the cytosol, (soluble alkaline phosphatase expressed in Man-P-Dol deficient CHO cells lines). The second model will illustrate a strict retention of glycoproteins in rough ER without degradation nor recycling through the Golgi (E1, E2 glycoproteins of Hepatitis C virus in stably transfected UHCV-11.4 cells and in infected Hep G2 cells). In both cases, oligomannoside structures are markers of these phenomena, either as free soluble released oligomannosides in the case of degradation, or as N-linked oligomannosides for strict retention in rough ER.