Dolichyl phosphate linked sugars as intermediates in the synthesis of yeast mannoproteins: an in vivo study

Protein glycosylation in yeast

S. cerevisiae contains many mannose-rich glycoproteins that possess N- and O-linked carbohydrate chains, and both types may even occur on one and the same protein. The steps in the synthesis of asparagine-linked chains begin with assembly and transfer of the lipid-linked precursor to protein in a way common to all eucaryotes. Subsequent modifications lead to mannosyl extensions of various lengths, but complex type carbohydrate structures are not formed. Oligosaccharides O-linked to serine/threonine consist exclusively of mannose in S. cerevisiae. The mannose residue attached directly to the protein is transferred from Dol-P-Man in a unique way, which has been observed so far for fungal cells only. The cellular localization of the glycosylation reactions is summarized and the problem of transmembrane translocation of the sugar precursors at the ER and the Golgi is discussed. Some aspects of secretory (sec) and asparagine linked glycosylation (alg) mutants have been covered, and the various hypotheses related to the possible functions of this costly protein modification process are discussed. The article may also be helpful for those, who want to exploit the yeast's protein synthesizing machinery by genetically manipulating the cells.

Synthesis and possible role of carbohydrate moieties of yeast glycoproteins

The pathways for protein N- and O-glycosylation in yeast cells are summarized. Evidence is presented that the terminal glucosyl residues of the dolichyl-PP-oligosaccharide intermediate are responsible for decreasing the Km for the peptide to be N-glycosylated. A liposomal model system is introduced that allows the study of a dolichyl phosphate (Dol-P) dependent transmembrane transport of mannosyl residues. The results obtained so far suggest that the mannosylation of Dol-P and the transmembrane translocation of Dol-P-Man are catalysed by the enzyme more or less simultaneously. However, only about 8-10% of the enzyme molecules incorporated into the liposomes seem to carry out the 'coupled' reaction. The glycosylation of carboxypeptidase Y is not required for this protein to reach the vacuole, its target organelle. In the presence of low concentrations of tunicamycin, however, yeast cells do stop growth. This does not seem to be due to the inhibition of secretion of glycoproteins like external invertase. It is postulated that protein glycosylation is crucial for a cell cycle event during the G1 phase.

Solubilization and characterization of the initial enzymes of the dolichol pathway from yeast

Preparation and purification of substrate amounts of radioactive as well as non-radioactive dolichyl diphosphate N-acetylglucosamine and dolichyl diphosphate chitobiose made it possible to test and characterize tentatively the first three reactions of the dolichol pathway (enzyme I-III). The test conditions are described in detail. All three enzymes were solubilized from yeast membranes with detergents. Enzyme II and III were purified to give a purification factor of 35-fold and 70-fold, respectively. The reactions required divalent metal ions with an optimum concentration of 10 mM Mg2+. Enzyme II was stimulated almost to the same extent also by Ca2+. The Km values for UDP-N-acetylglucosamine for enzyme I and II were 15 and 10 muM, respectively, and for GDP-mannose (enzyme III) 7 muM. The apparent Km values for the lipophilic acceptor was 180 muM for enzyme I (dolichyl phosphate), 40 muM for enzyme II (dolichyl diphosphate N-acetylglucosamine) and 17 muM for enzyme III (dolichyl diphosphate chitobiose). The corresponding V values were approximately 1, 10, and 50 nmol X h-1 X mg protein-1. All reactions were inhibited by nucleoside diphosphates.

Membranes of protein bodies. II. Detection and partial characterization of membrane glycoproteins

The synthesis of organelle proteins was studied in cotyledons of Cucumis sativus. Protein constituents of protein-body membranes were shown to be synthesized and assembled, at a stage characterized by mobilization of the storage globulin. Besides L-[35S]methionine various labelled hexoses were incorporated into protein bodies of cucumber cotyledons. While D-[U-14C] glucose functioned as precursor of a broad spectrum of glycoproteins, D-[6-3H]glucosamine was selectively incorporated into four glycoproteins of the protein-body membrane. Labelled galactose and mannose, respectively, were preferentially transferred into another set of membrane glycoproteins. The four glycoproteins revealed by labelling with glucosamine were solubilized and purified by chromatography on concanavalin-A-Sepharose.

[In vitro incorporation of 14C-hexose-6-phosphate in mannan, beta-glucan and glycogen in Candida spec. H and its mutants]

Mannose-6-p is an activator of the 14C-mannose incorporation from GDP-14C-mannose in the mono- and oligosaccharides and in the mannopolymers of the cell wall proteophosphomannan produced by the food protein yeast Candida spec. H. Moreover, mannose-6-p is a precursor of proteophosphomannan: 14C-Mannose-6-p has been incorporated in absence of GTP. Corresponding behaviour shows glucose-6-p by synthesis of beta-glucan and glycogen. Mutants of Candida spec. H with different efficiency in the biosynthesis of mannan, beta-glucan and glycogen incorporate hexose-6-p in a different extent.

N-Glycosylation of yeast proteins. Characterization of the solubilized oligosaccharyl transferase

The enzyme transferring the oligosaccharide from DolPP-(GlcNAc)2(Glc)3 to asparagine residues of glycoproteins has been solubilized from yeast membranes by extraction with detergents. Enzyme activity was tested by measuring transfer of the glycosyl moiety from DolPP-[14C]saccharides to the hexapeptide Tyr-Asn-Leu-Thr-Ser-Val. The rate of transfer was linear for 20 min, with about 40% of dolichyl-diphosphate-bound radioactivity transferred to the peptide. The solubilized enzyme has been characterized as follows: 1. The enzyme is most efficiently solubilized (60% of the membrane-associated activity) by 0.5% Nonidet P40 at a protein/detergent ratio of 2. Octylglucoside solubilizes one third of the activity, but strongly inhibits the reaction if present in the test at a concentration of 1%. 2. Divalent cations are absolutely required. 1 mM Mn2+ is optimal; Mg2+ at a concentration of 10mM yields only one third the activity observed with Mn2+. 3. The enzyme transfers besides dolichyl-diphosphate-bound (GlcNAc)2(Man)9(Glc)3 also (GlcNAc)2(Man)1 and (GlcNAc)2; the rate decreases in this order. No transfer is observed from DolPP-(GlcNAc)2(Man)9 and from DolPP-GlcNAc. 4. The Km value for DolPP-(GlcNAc)2(Man)9(Glc)3 of 0.5 microM does not differ significantly from that for DolPP-(GlcNAc)2 of 1.2 microM. A broad pH-optimum for the reaction with both substrates was found between 6.5 and 7.7. 5. However, a clear difference in Km values for the hexapeptide was observed with difference dolichol-linked sugar derivatives. With DolPP-(GlcNAc)2 a peptide concentration of 0.6 mM resulted in half-maximal transfer rate, whereas 0.05 mM peptide were sufficient with DolPP-(GlcNAc)2(Man)9(Glc)3 as donor.