Cell surface glycosyl transferases and acceptors in normal and RNA- and DNA-virus transformed fibroblasts

A unifying model of the cell proliferation emphasizing plasma membrane fluxes

The regulation of cellular growth and proliferation is perhaps the most investigated and elusive problem in cell biology and seems to be possible to solve from almost any angle of study chosen. Among the non-systemic factors that have been discussed are genetic damage, genomic control, regulation by stimulatory and inhibitory peptide factors such as EGF, chalones, and fibronectin, protein kinase activation with tyrosine phosphorylation, adenylylcyclase and cAMP, cGMP, membrane perturbations and specifically in tumours the failure of the Pasteur effect in control of glycolysis, excessive membrane ATPase activity, and excessive hydrolytic and proteolytic activities at the cell surface. This article focuses on the central role of fluxes within the plasma membrane and re-examines the possibility that changes of flux of metabolites, ions, and reducing equivalents may be the common denominator regulating cellular proliferation.

Binding sites for horseradish peroxidase on the cell surface. Suppression of binding by gangliosides and effects of some bivalent cations

The cytochemical reaction for surface-bound horseradish peroxidase (HRP) on cultured HeLa cells, GH3 cells, and isolated rat liver cells was suppressed by 30 microM monosialoganglioside, by 30 microM trisialoganglioside, or by 5 mM CMP-neuraminic acid. The reaction was also suppressed by 10 mM chitotriose or by 10 mM UDP-galactose, a galactose acceptor and donor, respectively, for galactosyl-transferase. The addition of 2 mM Mn2+ to the incubation medium with HRP suppressed the reaction for surface-bound HRP, and the addition of 10-20 mM Ca2+ intensified the reaction. The addition of 2 mM Zn2+ caused less inhibition than that of 2 mM Mn2+, and the addition of 2 mM Co2+ caused either a slight inhibition, or no inhibition. These observations support the hypothesis that HRP may be bound to a glycosyltransferase at the cell surface.

Unique antigen of cultured Hodgkin's cells. A putative sialyltransferase

Hodgkin's disease-derived giant cell lines (HD-cells) express high levels of ectosialyltransferase activity presumed to be a galactose-specific lectin recognizing the desialylated 3-fucosyl-N-acetyllactosamine structure (X-hapten). Both the anti-X-hapten monoclonal antibody VIM-D5 and a polyclonal antiserum to another galactose-lectin, the hepatic asialoglycoprotein receptor (HBP), recognize a 55,000-mol wt HD-cell protein (Paietta, E., R. J. Stockert, A. G. Morell, V. Diehl, and P. H. Weirnik. 1986. Proc. Natl. Acad. Sci. USA. 83:3451-3455.) That the expression of the 55,000-mol wt protein is restricted to HD-cells among X-hapten positive cells lines is confirmed in this study. The 55,000-mol wt protein is shown to be present on the cell surface and intracellularly, where an additional immunocrossreactive 150,000-mol wt protein is recognized. Extraction of the 55,000 mol wt protein from HD-cell lysates by affinity chromatography results in the loss of sialyltransferase activity. While evidence for a single protein possessing both the antigenic and the enzymatic activity is not direct, these results suggest that the ectosialyltransferase unique to HD-cells is a 55,000-mol wt membrane glycoprotein possessing the X-hapten oligosaccharide.

Lectin activity as a marker for Hodgkin disease cells

Treatment of cultured Hodgkin disease (HD) cells with neuraminidase results in decreased reactivity of monoclonal antibody VIM-D5 with its antigen, the X hapten, a fucosyl-N-acetyllactosamine. The other feature characteristic of HD cells is the expression of high levels of ectosialyltransferase activity. We present evidence for a cause-effect relationship between these two findings in that VIM-D5 antigenicity can be restored on neuraminidase-treated HD cells by modulating transferase activity. This can be interpreted in terms of a lectin activity of the ectosialyltransferase that binds the X hapten's desialylated galactosyl residues, thereby preventing antigen recognition by VIM-D5 antibody. This proposed mechanism is indistinguishable from the autoinhibition phenomenon described for another galactophilic binding protein, the hepatic binding protein (HBP), which binds its own terminal galactosyl residues following neuraminidase treatment. We establish a close relationship between the HD galactophilic binding site and HBP in that antiserum to HBP (i) inhibits the neuraminidase-induced loss of VIM-D5 antigenicity, (ii) blocks the binding of asialoglycoprotein to hepatocytes after being absorbed by and eluted from HD cells, and (iii) recognizes a single HD protein, which in its high level of expression is unique to HD cells. The presence of lectin activity in its classic sense on the surface of HD cells is confirmed by the erythrocyte-agglutinating ability of these cells. This lectin activity, which appears to be related to an ectosialyltransferase on the surface of HD cells, may serve as a marker for the abnormal cells characteristic of HD.

Incorporation of sialic acid into Trypanosoma cruzi macromolecules. A proposal for a new metabolic route

Sialo- and asialoglycoconjugates were isolated from Trypanosoma cruzi epimastigotes and their composition determined. Sialoglycoconjugates bound to wheat germ agglutinin (WGA)-Sepharose and were precipitated by concanavalin A, Wistaria floribunda hemagglutinin and WGA. Asialoglycoconjugate bound to concanavalin A-Sepharose and precipitated with concanavalin-A and W. floribunda hemagglutinin but not with WGA. Cells grown in the presence of fetal calf serum were agglutinated by WGA but not by peanut agglutinin. The reverse was true for cells grown without fetal calf serum. Neuraminidase-treated cells incorporated sialic acid or its 7-carbon analog, 5-acetamido-3,5-dideoxy-L-arabino-2-heptulosonic acid (AcNeu7) from sialylated compounds such as fetuin or sialyl-lactose but did not incorporate free sialic acid. Restoration of the WGA sialylreceptors in neuraminidase-treated cells, as determined by cell agglutination with WGA, was also obtained by incubation with fetuin or sialyl-lactose but not with free sialic acid. Moreover, restoration of agglutinability by WGA in neuraminidase-treated cells or cells grown in medium without fetal calf serum occurred equally well in energy-rich or energy-depleted cells. A transglycosilase reaction for sialic acid incorporation in T. cruzi epimastigotes is suggested.

Endogenous lectins of bovine pancreas

Affinity chromatography of salt and detergent extracts from bovine pancreas on glycosylated or glycoprotein-linked Sepharose 4B resulted in purification of different carbohydrate-binding proteins. Three species of proteins with molecular masses of 16 kDa, 35 kDa and 64 kDa exhibiting specificity for beta-galactosides, but none with preferential specificity for alpha-galactosides, were isolated from salt and detergent extracts. No Ca2+ was required for binding. Mannan-binding proteins of 37 kDa, 47 kDa and 94 kDa without Ca2+-requirement were only found in the salt extract. No other mannan-binding activity could be detected. Fucose-binding proteins of 34 kDa, 62 kDa and 70 kDa exhibiting Ca2+-requirement for binding were present in the salt extract and two proteins with 62 kDa and 70 kDa in detergent extract. The different fractions showed agglutination activity when assayed with rabbit erythrocytes. Thus they can be defined as lectins.

Multiple forms of chicken liver and hepatoma Mc-29 microsomal and plasma-membrane sialyl and fucosyltransferases

Multiple forms of microsomal and plasma membrane sialyl and fucosyltransferases from chicken liver and transplantable hepatoma Mc-29 have been separated by means of isoelectric focusing. A net different pattern was distinguished between liver and hepatoma microsomal and plasma-membrane associated transferases. Microsomal sialyltransferase from hepatoma Mc-29 has typical forms with pI = 5.69, 7.43, 8.05 and 8.56, while in plasma membrane, enzymes with pI = 5.00 and 8.70 occur. The presence of 9 forms of fucosyltransferase within the pH range 3.46-9.57 for hepatoma microsomes and within pH 4.52-9.60 for plasma membranes was detected. Forms with pI 5.10, 5.75 and 7.87 could be considered specific for the hepatoma microsomal enzyme, and forms with pI 4.52, 4.85 and 5.20 for the plasma-membrane associated enzyme.

Comparative study on fucosylation of chicken liver and virus-induced hepatoma Mc-29

Comparative studies on fucoprotein metabolism of chicken liver and hepatoma Mc-29 have been carried out and the following parameters were determined: the incorporation rate of [14C]fucose into hepatoma and liver total tissue homogenate, acid-soluble and acid-insoluble fractions, acid-soluble nucleotide fraction and into plasma-membrane acid-precipitable fraction; the activity of microsomal and plasma-membrane fucosyltransferase; the electrophoretic pattern of hepatoma and liver plasma-membrane proteins and the incorporation of [14C]fucose into the glycoprotein fractions in both plasma-membrane preparations. It was found that the labelling of hepatoma tissue homogenate and plasma membranes was higher than that of the same liver preparations 3 hr after the [14C]fucose injection. This finding was supported by a considerably elevated hepatoma fucosyltransferase activity. The labelling rate of numerous fucoproteins from hepatoma plasma membranes was greatly increased and some of the individual glycoprotein bands were labelled to a higher extent compared with liver. The data presented show specific alterations of fucose and fucoprotein metabolism which could be considered as a characteristic feature of chicken viral-induced hepatoma Mc-29.

Extensive radioactive labeling of glycolipids in cultured hamster fibroblasts by incubation of whole cells with UDP-[14C]glucose and UDP-[14C]galactose

NIL 2cl cells, a cloned hamster cell line, are known to have seven classes of neutral glycosphingolipids in addition to hematoside. In this paper, I examined the participation of cell surface glycosyltransferases in the glycolipid synthesis in NIL cells. The cells, harvested from culture bottles by the use of chelating agents of divalent cations, incorporated radioactive glucose and galactose from labeled UDPglucose and UDPgalactose into endogenous glycolipid acceptors to form all eight classes of glycosphingolipids and dolichyl phosphate glucose. The major part of the incorporation occurred by glycosyltransferase reactions from nucleotide sugars, as indicated by a slight inhibition of the incorporation observed in the presence of a large excess of sugars or sugar phosphates. Acid hydrolysis of the glycolipid products showed that 91.4% of the 14C label was in the galactose and glucose moieties of the glycolipids. Exogenously added lactosylceramide stimulated the synthesis of trihexosylceramide in the absence of detergents. Th results presented in this paper suggest that incubation of whole cells with an equilibrium mixture of radioactive UDPglucose and UDPgalactose is a useful radioactive labeling method to determine the glycolipid composition of a cultured cell.

[Mitochondrial N-acetylglucosaminyl transferase connected with viral infection]

A higher level of N-acetylglucosamine incorporation by proteinic and polyprenic endogenous acceptors is observed after infection by Myxovirus. This phenomenon occurs in whole mitochondria as in outer mitochondrial membranes, where it is particularly obvious with proteinic endogenous acceptors. Under viral infection, no new N-acetylglucosaminylated polyprenol is detected. In the case of infected animals as in the case of control animals, compounds P1 (extracted by chloroform/methanol 2:1) are identified by thin layer chromatography as a N-acetylglucosamyl-pyrophosphoryl-dolichol and a N, N'-diacetylchitobiosyl-pyrophosphoryl-dolichol. In the case of infected animals, biosynthesis of proteinic acceptors and dolichol is not modified; therefore the increase of N-acetylglucosamine incorporation is not due to a modification of the endogenous acceptors level. By the use of exogenous dolichol-monophosphate we demonstrate that the increased transfer of [14C] N-acetylglucosamine into polyprenic acceptors is the result of a higher activity of the mitochondrial N-acetylglucosaminyltransferase after viral infection.

The repair of the surface structure of animal cells

Experiments were performed to determine if animal cells in culture possess specific mechanisms to repair surface molecules damaged by enzymes. The surface membranes of a primary cell culture, chick fibroblasts, a permanent hamster cell line, BHK21/C13, and its virally transformed counterpart, C13/B4 were damaged by exposure to trypsin or to neuraminidase. Following digestion with trypsin, the incorporation of radioactive amino acids or sugars into purified surface membrane of cells was monitored. No differences were noted in rates of incorporation when control and trypsin-damaged cells were compared. Neuraminidase damage to the surface of BHK21/C13 and C13/B4 cells was evidenced by altered gel filtration profiles of surface glycopeptides, i.e., delayed elution because of reduction in size. By labelling cells with 14C-L-fucose prior to neuraminidase treatment and following the incorporation of 3H-L-fucose into cell surface glycopeptides after neuraminidase digestion, we were able to monitor the synthesis and turnover of fucose-containing glycopeptides in the same cells. Gel filtration profiles indicated that little or no desialylated glycoproteins were resialylated (repaired) by specific replacement of sialic acid. Comparing neuraminidase-digested and control cells we observed no difference in rates of 3H-L-fucose incorporation or of 14C-L-fucose loss from these cells; nor did we find differences in the rate of incorporation of isotopic glucosamine into sialic acid. Neuraminidase treatment failed to alter the rate of cell growth or the pattern of isotopic incorporation into various cell surface components. These results support the suggestion that return of sialic acid (repair) was effected by turnover which serves as a non-specific repair mechanism to replace damaged cell surface molecules (Warren and Glick '68; Warren, '69).

Cell contact dependence of surface galactosyltransferase activity as a function of the cell cycle

Mitotic, nonmalignant Balb/c 3T3 cells exhibit endogenous, surface galactosyltransferase activity that does not require intercellular contact throughout the assay period. In this respect, mitotic 3T3 cells resemble malignant Balb/c 3T12 cells which similarly show no contact requirement for optimum transferase activity in any phase of their cell cycle. Previously, it was shown that randomly growing populations of 3T3 cells have lower galactosyltransferase activity when assayed under conditions which decreased cell contact. This led to the conclusion that these normal (3T3) and malignant (3T12) cells differed in that intercellular contact is required for optimum activity of surface galactosyltransferases on the normal cell type. The present data indicate that mitotic 3T3 cells may be capable of expressing enzyme activities exhibited at all times by malignant cells. That is, mitotic 3T3 cells and randomly growing 3T12 cells may readily catalyze galactosyltransferase reactions between enzymes and acceptors on the same cell. Interphase 3T3 cells, on the other hand, might require that enzymes glycosylate acceptors on adjacent cells. A model is proposed that suggests that changes in the spatial arrangement of surface enzymes and acceptors or variations in the fluidity of the cell membrane can account for this contact-related glycosylation.

Isolation of a major cell surface glycoprotein from fibroblasts

A cell surface component has been isolated in partially purified form from cultured chick embryo and chick heart fibroblasts. This glycoprotein is similar to a protein recently reported to be present at the surface of normal cells, and missing after neoplastic transformation. It is a major cell surface glycoprotein that is synthesized by cultured fibroblasts, but is not collagen. It is shown to be markedly trypsin-sensitive, and its recovery from the cell surface is dependent on cell density. It is excluded from Sephadex G-200, but is not rapidly sedimented by ultracentrifugation, and has an apparent molecular weight of 220,000. The isolation of this cell surface glycoprotein may now provide a means of determining its function.

Differentiation between intracellular and cell surface glycosyl transferases: galactosyl transferase activity in intact cells and in cell homogenate

Intact BHK (baby hamster kidney) cells catalyze the hydrolysis of UDP-galactose to free galactose. The generation of galactose from UDP-galactose and its intracellular utilization impede the detection of possible galactosyl transferases on the cell surface of intact cells. Several independent procedures have been used to distinguish between intracellular and cell surface glycosyl transferases. With these procedures, no evidence was obtained for the presence of detectable amounts of galactosyl transferase activity on the surface of BHK cells. The data suggest that galactosyl transferases do not play a general role in the phenomena of cell adhesion and contact inhibition.

Enzyme activity in invasive tumors of human breast and colon

Elevated levels of glycoprotein:sialyltransferase activity (EC 2.4.99.1; CMP-N-acetylneuraminate: D-galactosyl-glycoprotein N-acetylneuraminyltransferase) were found in human malignant neoplastic tissues compared to normal, benign, and "preneoplastic" tissues. This increase was not due to the cell density of the tissue. Elevated levels of certain proteases and glycosidases were also found. The increase in transferase activity may be associated with altered membrane synthesis in the neoplastic state; changes in the activity of degradative enzymes may be associated with tumor invasiveness and maintenance of the neoplastic state. Measurements on human tumors are possibly more directly relevant to cancer than those described for transformed fibroblastic cells in vitro.

Migration of glycoprotein from the Golgi apparatus to the surface of various cell types as shown by radioautography after labelled fucose injection into rats

A single intravenous injection of L-[(3)H]fucose, a specific glycoprotein precursor, was given to young 35-45 g rats which were sacrificed at times varying between 2 min and 30 h later. Radioautography of over 50 cell types, including renewing and nonrenewing cells, was carried out for light and electron microscope study. At early time intervals (2-10 min after injection), light microscope radioautography showed a reaction over nearly all cells investigated in the form of a discrete clump of silver grains over the Golgi region. This reaction varied in intensity and duration from cell type to cell type. Electron microscope radioautographs of duodenal villus columnar cells and kidney proximal and distal tubule cells at early time intervals revealed that the silver grains were restricted to Golgi saccules. These observations are interpreted to mean that glycoproteins undergoing synthesis incorporate fucose in the saccules of the Golgi apparatus. Since fucose occurs as a terminal residue in the carbohydrate side chains of glycoproteins, the Golgi saccules would be the site of completion of synthesis of these side chains. At later time intervals, light and electron microscope radioautography demonstrated a decrease in the reaction intensity of the Golgi region, while reactions appeared over other parts of the cells: lysosomes, secretory material, and plasma membrane. The intensity of the reactions observed over the plasma membrane varied considerably in various cell types; furthermore the reactions were restricted to the apical surface in some types, but extended to the whole surface in others. Since the plasma membrane is covered by a "cell coat" composed of the carbohydrate-rich portions of membrane glycoproteins, it is concluded that newly formed glycoproteins, after acquiring fucose in the Golgi apparatus, migrate to the cell surface to contribute to the cell coat. This contribution implies turnover of cell coat glycoproteins, at least in nonrenewing cell types, such as those of kidney tubules. In the young cells of renewing populations, e.g. those of gastro-intestinal epithelia, the new glycoproteins seem to contribute to the growth as well as the turnover of the cell coat. The differences in reactivity among different cell types and cell surfaces imply considerable differences in the turnover rates of the cell coats.