Site-specific glycosylation in animal cells. Substitution of glutamine for asparagine 293 in chicken ovalbumin does not allow glycosylation of asparagine 312.

It has been shown previously that chicken ovalbumin synthesized and secreted in a heterologous cell system is glycosylated at the correct site and that the oligosaccharides at that site, similar to the protein made in hen oviduct, are predominantly of the hybrid type (Sheares, B. T., and Robbins, P. W. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 1993-1997). This site-specific glycosylation of Asn293, but not Asn312, suggested a prominent role for the nascent protein chain rather than the specific cell type in directing the proper attachment of oligosaccharide chains. In the present study, the effect of glycosylation at Asn293 on the glycosylation of Asn312 has been investigated. Using a 20-base oligodeoxynucleotide primer containing a 2-base mismatch, the codon for Asn293 in the chicken ovalbumin gene (AAC) was changed to that for Gln (CAA), thereby preventing glycosylation at amino acid 293. Constructions containing this mutation were transfected into mouse L (tk-) cells which were subsequently labeled with [35S]methionine. Ovalbumin secreted by these cells was recovered by immunoaffinity chromatography and analyzed for the presence of an oligosaccharide attached at Asn312. Treatment of the material with peptide:N-glycosidase F demonstrated that ovalbumin molecules containing Gln substituted for Asn293 were not glycosylated. This further supports our earlier hypothesis that the nascent protein chain is responsible for directing site-specific glycosylation of ovalbumin, and that the presence of an oligosaccharide chain at the first site has no influence on glycosylation at the second site.

Modulation of two distinct galactosyltransferase activities in populations of mouse peritoneal macrophages.

We have examined two galactosyltransferase activities in membrane preparations obtained from resident macrophages, from resident macrophages maintained in culture for 24 hr, and from thioglycollate (TG)-elicited macrophages. Transfer of galactose from uridine diphosphate (UDP)-galactose to N-acetylglucosamine is 2.6 times higher in membranes prepared from TG macrophages (107 +/- 5.5 nmol/hr/mg) than in membranes prepared from resident macrophages (41 +/- 2.0 nmol/hr/mg). Membranes obtained from resident macrophages cultured for 24 hr exhibit a 2.5 times higher activity (102 +/- 4.4 nmol/hr/mg) than membranes from resident cells plated for 4 hr. Transferase activity in membranes derived from TG macrophages is not significantly affected by overnight culture. The transferase reaction product, isolated on Bio-Gel P-4 and analyzed by galactosidase treatments, was identified as galactosyl-beta 1, 4-N-acetylglucosamine. The enzyme, therefore, is UDP-galactose:2-acetamido-2-deoxy-D-glucose 4 beta-galactosyltransferase. This is supported by the fact that this galactosyltransferase activity is specifically inhibited by high concentrations of N-acetylglucosamine (200 mM). We have also examined the transfer of galactose to N-acetyllactosamine. Membranes from TG-elicited macrophages contain a UDP-galactose:galactosyl-beta 1, 4-N-acetylglucosamine 3 alpha-galactosyltransferase which synthesizes the trisaccharide, galactosyl-alpha 1, 3-galactosyl-beta 1,4-N-acetylglucosamine. This product was identified by gel filtration chromatography, high performance liquid chromatography, and galactosidase digestions. This alpha-galactosyltransferase activity was not detected in membranes prepared from resident macrophages. These results indicate that glycosyltransferase activities are modulated in populations of mouse macrophages, and that these changes correlate with changes in cell surface lactosaminoglycans reported previously.

Characterization of UDP-galactose:2-acetamido-2-deoxy-D-glucose 3 beta-galactosyltransferase from pig trachea.

We have previously reported the enzymatic synthesis of galactosyl-beta 1,3-N-acetylglucosamine using membrane preparations from pig trachea (Sheares, B. T., Lau, J. T. Y., and Carlson, D. M. (1982) J. Biol. Chem. 257, 599-602). The enzyme catalyzing the synthesis of this disaccharide, UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase, has been solubilized and contaminating galactosyltransferases, including the UDP-galactose:N-acetylglucosamine 4 beta-galactosyltransferase and the UDP-galactose:N-acetylgalactosamine-mucin 3 beta-galactosyltransferase, were removed by affinity chromatography on alpha-lactalbumin-agarose, N-acetylglucosamine-agarose, and asialo-ovine submaxillary mucin bound to DEAE-Sephacel. UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase and a yet unidentified UDP-galactose:N-acetylglucosamine 4 beta-galactosyltransferase, both not retained by the alpha-lactalbumin-agarose, were further purified by adsorption to a UDP-hexanolamine-Sepharose column and these two galactosyltransferase activities were finally resolved by gel filtration on Sephacryl S-200. The purified UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase displays an absolute requirement for a divalent cation (Mn2+ or Co2+) and is most active at pH 6.0. Unlike the UDP-galactose:N-acetylglucosamine 4 beta-galactosyltransferase, UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase is not inhibited by high concentrations of N-acetylglucosamine. Apparent Km values are UDP-galactose, 0.23 mM, and N-acetylglucosamine, 385 mM. The apparent Km for a synthetic acceptor, N-acetylglucosaminyl-beta 1, 3-N-acetylgalactosamine-O-benzyl, was 2.4 mM. Acceptor specificity suggests that this 3 beta-galactosyltransferase is responsible for elongation of oligosaccharide chains in mucin glycoproteins and in glycolipids.

Two distinct UDP-galactose: 2-acetamido-2-deoxy-D-glucose 4 beta-galactosyltransferases in porcine trachea.

In previous studies on glycosyltransferase activities in porcine trachea, we demonstrated the presence of two galactosyltransferases which transfer galactose from UDP-galactose to N-acetylglucosamine (Sheares, B.T. and Carlson, D.M. (1983) J. Biol. Chem. 258, 9893-9898). One enzyme, UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase, synthesized galactosyl-beta 1,3-N-acetylglucosamine while the other, UDP-galactose:N-acetylglucosamine 4 beta-galactosyltransferase, synthesized galactosyl-beta 1,4-N-acetylglucosamine. A third galactosyltransferase has now been demonstrated utilizing a solubilized membrane preparation from pig trachea, which also synthesizes galactosyl-beta 1,4-N-acetylglucosamine as determined by gas-liquid chromatography and Diplococcus pneumoniae beta-galactosidase treatment. This new UDP-galactose:N-acetylglucosamine 4 beta-galactosyltransferase is distinct from the lactose synthetase A protein in that it does not bind to alpha-lactalbumin-agarose or to N-acetylglucosamine-agarose. The enzyme is separable from the UDP-galactose:N-acetylgalactosaminyl-mucin 3 beta-galactosyltransferase by affinity chromatography on asialo ovine submaxillary mucin adsorbed to DEAE-Sephacel. This newly discovered 4 beta-galactosyltransferase binds to UDP-hexanolamine-Sepharose and is partially separated from UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase by Sephacryl S-200 gel filtration chromatography. Neither high concentrations of N-acetylglucosamine (200 mM) nor alpha-lactalbumin inhibits the incorporation of galactose into galactosyl-beta 1,4-N-acetylglucosamine by this enzyme.

Glycosylation of ovalbumin in a heterologous cell: analysis of oligosaccharide chains of the cloned glycoprotein in mouse L cells.

In an effort to understand the factors that influence protein glycosylation, we are studying the expression of the chicken ovalbumin gene in a heterologous cell. Ovalbumin synthesized in mouse L-353 cells is glycosylated, as judged by incorporation of [3H]mannose and susceptibility to endo-beta-N-acetylglucosaminidases. Sequential digestion of ovalbumin synthesized in L cells with trypsin and chymotrypsin yields material migrating as one peak on HPLC coincident with similarly treated material from chicken ovalbumin, suggesting that the protein synthesized in L cells is glycosylated at the correct site. As in the case of chicken ovalbumin, the protein synthesized in L cells contains large amounts of hybrid oligosaccharides. Approximately 50% of the [3H]mannose incorporated into ovalbumin secreted by L cells is found in such hybrid structures. These results suggest that it is the polypeptide chain of ovalbumin that is responsible for proper glycosylation and subsequent processing of a substantial fraction of the oligosaccharide chains to hybrid structures. However, differences do exist between ovalbumin synthesized in L cells and the chicken glycoprotein. The hybrid oligosaccharides of ovalbumin secreted by L cells are completely sialylated and do not contain a bisecting GlcNAc residue, distinguishing them from hybrid chains in chicken ovalbumin. In addition to high-mannose and hybrid oligosaccharide chains, ovalbumin synthesized in L cells contains oligosaccharides of the complex type. To date, this type of sugar chain has not been observed in chicken ovalbumin. These differences in fine structure, between the oligosaccharides derived from ovalbumin secreted by L cells and those known to be present in the chicken egg glycoprotein, suggest that the cell type also plays a role in oligosaccharide processing.

Biosynthesis of galactosyl-beta 1,3-N-acetylglucosamine.

The biosynthesis of galactosyl-beta 1,3-N-acetylglucosamine has been demonstrated using membrane preparations from pig trachea. Unlike the UDP-galactose:2-acetamido-2-deoxy-D-glucose 4 beta-galactosyltransferase, which is inhibited by high levels of N-acetylglucosamine, the UDP-galactose:N-acetylglucosamine 3 beta-galactosyltransferase shows no inhibition at 200 mM N-acetylglucosamine. About 80% of the total disaccharide synthesized at 200 mM N-acetylglucosamine was base-labile suggesting the 1,3-linkage, alpha-Lactalbumin inhibits galactose incorporation into galactosyl-beta 1,4-N-acetylglucosamine but has little or no effect on the activity of the 1,3-galactosyltransferase. Escherichia coli beta-galactosidase readily hydrolyzed the base-stable product, but not the base-labile component. The apparent 1,3-linked disaccharide was reduced with NaBH4 and was isolated by Bio-Gel P-2 column chromatography. Methylation analysis by gas chromatography/mass spectrometry showed tetramethyl galactose and a 3-substituted N-acetylglucosaminitol. Neither the beta 1,4 nor the beta 1,3 disaccharide was hydrolyzed by green coffee bean alpha-galactosidase. Both disaccharides were readily hydrolyzed by bovine testes beta-galactosidase. This is the first report on the galactosyltransferase which catalyzes the synthesis of the galactosyl-beta 1,3-N-acetylglucosamine linkage such as found in the Type I chain of human blood group substances. A tissue survey in rats showed only rat intestine to have readily detectable UDP-galactose: N-acetylglucosamine 3 beta-galactosyltransferase activity. The intestinal membrane fraction like the tracheal enzyme catalyzes the synthesis of two disaccharides as judged by base treatment, and these appear to be the beta 1,3 and beta 1,4 isomers of galactosyl-N-acetylglucosamine.

Substructure of granules from serous cells of porcine tracheal submucosal glands.

The ultrastructure of serous cells from porcine tracheal submucosal glands was studied by conventional transmission electron microscopy (TEM), and by cytochemical methods to stain for complex carbohydrates. In tissue fixed and processed for TEM, and stained with uranyl acetate and lead citrate, the condensing granules of serous cells occasionally possessed a hexagonal and sometimes a lamellar substructure. Tissue fixed in paraformaldehyde-glutaraldehyde and stained with periodic acid-thiocarbohydrazide-silver proteinate (PTS) or with phosphotungstic acid (PTA) showed secretory granules stained for complex carbohydrates and revealed a substructure similar to that noted in the condensing granules. The dark staining substructure revealed by either the PTS or the PTA technique appeared to correspond to electron-lucent areas observed in the condensing granules by conventional TEM. The PTS staining probably demonstrated the presence of neutral glycoprotein, since the serous-cell granules did not react with a dialyzed iron stain for acidic glycoproteins. Treatment of periodic acid oxidized thin sections with pronase or pepsin prior to thiocarbohydrazide and silver proteinate treatment decreased the intensity of the PTS staining, but did not digest away any components of the granules. The substructure revealed by the carbohydrate stains may be a reflection of the mechanism of packaging or the macromolecular structure of the glycoproteins in the serous-cell granules.

Induction of proline-rich glycoprotein synthesis in mouse salivary glands by isoproterenol and by tannins.

Glycoproteins which contain about 45 mol% proline were dramatically induced in mouse parotid and submandibular glands by isoproterenol treatment, but these unusual proteins were not detected in control animals. These acid-soluble substances were obtained by extracting tissues with 10% trichloroacetic acid, as reported previously for isolating proline-rich proteins from rat submandibular glands (Mehansho, H., and Carlson, D.M. (1983) J. Biol. Chem. 258, 6616-6620). Three major proline-rich glycoproteins were induced in parotid glands with apparent molecular weights of 66,000 (GP-66p), 45,000 (GP-45p), and 27,000 (GP-27p), whereas only one such protein was expressed by the submandibular glands (66,000 (GP-66sm]. Both GP-66p and GP-66sm contained about 19% carbohydrate with the following molar ratios, respectively; GalNAc, 1.0, 1.0; Gal, 1.6, 2.3; GlcNAc, 0.8, 1.1; sialic acid, 0.9, 1.9. The peptide chains of GP-66p and GP-66sm appear to be identical by amino acid compositions, glycopeptide analysis, and preliminary amino acid sequencing data. Northern blot analysis of RNAs from parotid glands of normal and isoproterenol-treated rats, probed with a 32P-labeled proline-rich protein cDNA, confirmed that control animals were devoid of mRNAs encoding these proteins and that isoproterenol treatment dramatically induced expression of these genes. Feeding sorghum high in tannins caused changes in the parotid glands similar to those observed upon isoproterenol treatment, as noted earlier with rats (Mehansho, H., Hagerman, A., Clements, S., Butler, L., Rogler, J., and Carlson, D.M. (1983) Proc. Natl. Acad. Sci. U.S.A. 80, 3948-3952). These glycoproteins have high affinities for tannins as demonstrated by competitive binding curves.

Purification and characterization of recombinant human farnesyl diphosphate synthase expressed in Escherichia coli.

We previously reported the isolation of a partial-length human fetal-liver cDNA encoding farnesyl diphosphate (FPP) synthase (EC 2.5.1.10) and the expression of an active FPP synthase fusion protein in Escherichia coli. The expressed human FPP synthase fusion protein has now been purified to apparent homogeneity by using two chromatographic steps. The purification scheme allowed the preparation of 1.8 mg of homogeneous protein from 149 mg of crude extract in a 64% yield with a 52-fold enrichment. A single band with a subunit molecular mass of 39 kDa was observed by Coomassie Blue staining after SDS/PAGE. A molecular mass of 78-80 kDa was calculated for the native form of the fusion protein by h.p.l.c. on a SEC-250 column, suggesting that the active fusion protein is a dimer. The purified fusion protein has FPP synthase condensation activities in the presence of both substrates, isopentenyl diphosphate and geranyl diphosphate. Enzyme activity was inhibited by a bisubstrate analogue of isopentenyl diphosphate and dimethylallyl diphosphate, and a small amount of higher prenyltransferase was observed. Michaelis constants for isopentenyl diphosphate and geranyl diphosphate were 0.55 and 0.43 microM respectively, and Vmax for synthesis of farnesyl diphosphate from these substrates was 1.08 mumol/min per mg. These results suggest that the structure and catalytic properties of the expressed FPP synthase fusion protein are virtually identical with those of the native human liver enzyme.

Structure of the 22-residue somatostatin from catfish. An O-glycosylated peptide having multiple forms.

The 22-residue somatostatin (SST-22) from channel catfish, purified by an improved method, is shown to be a glycopeptide. This represents the first report of a glycosylated somatostatin. Multiple forms of SST-22 exist with the major form containing 1 mol of galactose and 1 mol of N-acetylgalactosamine/mol of peptide attached via an O-glycosidic linkage to Thr-5. The position of the carbohydrate was determined by trapping the reactive peptide following beta-elimination of the carbohydrate with [35S]beta-mercaptoethanol followed by sequencing of the radiolabeled protein. All forms of SST-22 that have been purified are identical in amino acid composition. The heterogeneity resides in the carbohydrate portion of the glycopeptide with at least one of the minor forms containing sialic acid. The sequence for SST-22 obtained by automated Edman degradation is Asp X Asn X Thr X Val X Thr X Ser X Lys X Pro X Leu X Asn X Cys X Met X Asn X Tyr X Phe X Trp X Lys X Ser X Arg X Thr X Ala X Cys. This sequence differs at positions 5 and 19 from that published by Oyama et al. (Oyama, H., Bradshaw, R. A., Bates, O.J., and Permutt, A. (1980) J. Biol. Chem. 255, 2251-2254). The amino acid sequence reported here is identical to that deduced from the cDNA. The mass ion of SST-22 was determined by fast atom bombardment/mass spectrometry and shown to be 2943 +/- 1 (m/z). The observed mass ion is consistent with the molecular weight predicted from the amino acid sequence plus 1 mol of galactose and 1 mol of N-acetylgalactosamine.

Cloning, analysis, and bacterial expression of human farnesyl pyrophosphate synthetase and its regulation in Hep G2 cells.

A partial length cDNA encoding farnesyl pyrophosphate synthetase (hpt807) has been isolated from a human fetal liver cDNA library in lambda gt11. DNA sequence analysis reveals hpt807 is 1115 bp in length and contains an open reading frame coding for 346 amino acids before reaching a stop codon, a polyadenylation addition sequence, and the first 14 residues of a poly(A+) tail. Considerable nucleotide and deduced amino acid sequence homology is observed between hpt807 and previously isolated rat liver cDNAs for farnesyl pyrophosphate synthetase. Comparison with rat cDNAs suggests that hpt807 is about 20 bp short of encoding the initiator methionine of farnesyl pyrophosphate synthetase. The human cDNA was cloned into a prokaryotic expression vector and Escherichia coli strain DH5 alpha F'IQ was transformed. Clones were isolated that express an active fusion protein which can be readily observed on protein gels and specifically stained on immunoblots with an antibody raised against purified chicken farnesyl pyrophosphate phosphate synthetase. These data confirm the identity of hpt807 as encoding farnesyl pyrophosphate synthetase. Slot blot analyses of RNA isolated from Hep G2 cells show that the expression of farnesyl pyrophosphate synthetase mRNA is regulated. Lovastatin increases mRNA levels for farnesyl pyrophosphate synthetase 2.5-fold while mevalonic acid, low-density lipoprotein, and 25-hydroxycholesterol decrease mRNA levels to 40-50% of control values.