Search

Saccharomyces cerevisiae possesses two glycogenin isoforms (designated as Glg1p and Glg2p) that both contain a conserved tyrosine residue, Tyr232. However, Glg2p possesses an additional tyrosine residue, Tyr230 and therefore two potential autoglucosylation sites. Glucosylation of Glg2p was studied using both matrix-assisted laser desorption ionization and electrospray quadrupole time of flight mass spectrometry. Glg2p, carrying a C-terminal (His6) tag, was produced in Escherichia coli and purified. By tryptic digestion and reversed phase chromatography a peptide (residues 219–246 of the complete Glg2p sequence) was isolated that contained 4–25 glucosyl residues. Following incubation of Glg2p with UDPglucose, more than 36 glucosyl residues were covalently bound to this peptide. Using a combination of cyanogen bromide cleavage of the protein backbone, enzymatic hydrolysis of glycosidic bonds and reversed phase chromatography, mono- and diglucosylated peptides having the sequence PNYGYQSSPAM were generated. MS/MS spectra revealed that glucosyl residues were attached to both Tyr232 and Tyr230 within the same peptide. The formation of the highly glucosylated eukaryotic Glg2p did not favour the bacterial glycogen accumulation. Under various experimental conditions Glg2p-producing cells accumulated approximately 30% less glycogen than a control transformed with a Glg2p lacking plasmid. The size distribution of the glycogen and extractable activities of several glycogen-related enzymes were essentially unchanged. As revealed by high performance anion exchange chromatography, the intracellular maltooligosaccharide pattern of the bacterial cells expressing the functional eukaryotic transgene was significantly altered. Thus, the eukaryotic glycogenin appears to be incompatible with the bacterial initiation of glycogen biosynthesis.

Saccharomyces cerevisiae possesses two glycogenin isoforms (designated as Glg1p and Glg2p) that both contain a conserved tyrosine residue, Tyr232. However, Glg2p possesses an additional tyrosine residue, Tyr230 and therefore two potential autoglucosylation sites. Glucosylation of Glg2p was studied using both matrix-assisted laser desorption ionization and electrospray quadrupole time of flight mass spectrometry. Glg2p, carrying a C-terminal (His6) tag, was produced in Escherichia coli and purified. By tryptic digestion and reversed phase chromatography a peptide (residues 219-246 of the complete Glg2p sequence) was isolated that contained 4-25 glucosyl residues. Following incubation of Glg2p with UDPglucose, more than 36 glucosyl residues were covalently bound to this peptide. Using a combination of cyanogen bromide cleavage of the protein backbone, enzymatic hydrolysis of glycosidic bonds and reversed phase chromatography, mono- and diglucosylated peptides having the sequence PNYGYQSSPAM were generated. MS/MS spectra revealed that glucosyl residues were attached to both Tyr232 and Tyr230 within the same peptide. The formation of the highly glucosylated eukaryotic Glg2p did not favour the bacterial glycogen accumulation. Under various experimental conditions Glg2p-producing cells accumulated approximately 30% less glycogen than a control transformed with a Glg2p lacking plasmid. The size distribution of the glycogen and extractable activities of several glycogen-related enzymes were essentially unchanged. As revealed by high performance anion exchange chromatography, the intracellular maltooligosaccharide pattern of the bacterial cells expressing the functional eukaryotic transgene was significantly altered. Thus, the eukaryotic glycogenin appears to be incompatible with the bacterial initiation of glycogen biosynthesis.

Higher plants possess two types of glucan phosphorylase (EC 2.4.1.1). One isozyme type, designated as Pho1, is located in the plastid whereas the other type, Pho2, is restricted to the cytosol. For Solanum tuberosum L. two Pho1 type phosphorylases have been sequenced [Nakano, K. & Fukui, T. (1986) J. Biol. Chem. 261, 8230-8236; Sonnewald, U., Basner, A., Greve, B. & Steup, M. (1995) Plant Mol. Biol. 27, 567-576]. Both proteins (referred to as Pho1a and Pho1b, respectively) are highly similar (81-84% amino acid identity over most parts of the two sequences) with the exception of the N-terminal transit peptide and the large insertion located between the N- and the C-terminal domains. In this communication antibodies that bind specifically to either Pho1a or Pho1b were used to study both isoforms at the protein level. The antibodies were applied to both potato tuber and leaf extracts following either denaturing or non-denaturing electrophoresis. Pho1a but not Pho1b was immunochemically detectable in tuber extracts whereas leaf extracts contained both the Pho1a and Pho1b protein. During denaturing electrophoresis the two antigens comigrated. When the leaf Pho1 isoforms were separated by affinity electrophoresis three bands of activity were resolved; all of them were recognized by the anti-Pho1a antibodies, but only two of these reacted with the anti-Pho1b antibodies. The isoform binding exclusively to the anti-Pho1a antibodies comigrated with the Pho1 isozyme from potato tubers. Immunoprecipitation experiments performed with anti-Pho1a antibodies removed the entire Pho1 phosphorylase activity from both tuber and leaf extracts. Addition of anti-Pho1b antibodies to tuber extracts did not affect the enzyme pattern, whereas in leaf extracts one isoform remained unchanged but the two other bands were strongly retarded. This indicates that the Pho1a protein is present in all three forms and Pho1b is associated with Pho1a. Association of Pho1a and Pho1b was further demonstrated by cross-linking experiments using bis(sulfosuccinimidyl)suberate as linker. Immunoprecipitation experiments were also performed using extracts of transformed Escherichia coli cells that expressed either Pho1a or Pho1b or both simultaneously. Under these conditions a homodimeric Pho1b phosphorylase was observed that had a lower electrophoretic mobility than the heterodimer from leaves. In leaves of transgenic potato plants antisense inhibition of the Pho1a gene affected the formation of (Pho1a)2 more strongly than that of the heterodimer. Thus, in leaves, Pho1a exists both as a homodimer, (Pho1a)2 and as heterodimer, (Pho1a-Pho1b); a part of it appears to be covalently modified. Pho1b, in the homodimeric form, is often below the limit of detection. In tubers the homodimer, (Pho1a)2, is the only detectable Pho1-type enzyme. To our knowledge this is the first report on a heterodimeric structure of plant phosphorylase.

Purple acid phosphatase of the common bean Phaseolus vulgaris (KBPase), a dimeric 110-kDa glycoprotein related to the mammalian purple acid phosphatases with a two-metal cluster at the active site contains five oligosaccharide side chains/monomer. The N-linked glycan structures were characterized by selective enzymic degradation in combination with matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). The purified protein was cleaved by cyanogen bromide. One 30-kDa large methionine-free fragment required a further tryptic digest. The peptides were separated by HPLC and the glycosylated species were identified both by their heterogeneous mass spectra and by an immunoassay. None of the glycopeptides proved to have more than one glycosylation site. The composition of the carbohydrate moieties were calculated by comparing the mass spectra of the glycopeptides before and after enzymic deglycosylation. These results were complemented by data from a carbohydrate composition analysis. In four of the five peptides an α1–3 fucose attached to the asparagine-linked N-acetylglucosamine prevented removal of the glycan by peptide N-glycosidase F; peptide N-glycosidase A removed all carbohydrates from the peptides. To reveal the sequence of the carbohydrate moiety including the linkage positions between the different saccharides, one of the glycopeptides was degraded by specific exoglycosidases. The enzymic degradations by these hydrolases were monitored by mass spectrometry of small aliquots taken at intervals during the reaction. The detailed structure of this one glycan in conjunction with the respective mass spectra and the composition analysis were used to infer the structure of the other four glycans. All glycans of the KBPase have a complex-type xylose-containing structure with four of the five having an additional fucose.

Extracts of Synechococcus leopoliensis (Anacystis nidulans) contain two forms of D-fructose-1,6-bisphosphatase (EC 3.1.3.11) previously designated as forms A and B [Gerbling, K.-P., Steup, M., and Latzko, E. (1984) Arch. Microbiol. 137, 109-114]. Form B, which probably represents the major part of the total extractable fructose-1,6-bisphosphatase activity, has been purified to apparent homogeneity. Gel filtration, non-denaturing polyacrylamide gel electrophoresis, and cross-linking with bis(sulfosuccinimidyl)suberate revealed that the fructose-1,6-bisphosphatase B exists in either a dimeric or in a tetrameric subform, depending upon the absence or presence of fructose-1,6-bisphosphate and Mg2+. The dimer--tetramer interconversion was readily reversible. The results provide evidence for a two-step activation of fructose-1,6-bisphosphatase B involving the reduction of the dimeric subform and the subsequent substrate-dependent conversion of the reduced dimer to a reduced tetramer, which is the only catalytically active state. In contrast to form B, no substrate-dependent interconversion was detected with form A from S. leopoliensis.