PRODUCTION AND STRUCTURAL INVESTIGATION OF GLYCANS AND PROTEINS OF GLYCOBIOLOGICAL RELEVANCE
lf carbohydrates are viewed not merely as energy suppliers for sustaining life but their intriguing biological role, chiefly as modulators of cell communication, is investigated, then we approach the field of glycobiology. This nontrivial task of glycans arises from their inherent structural complexity (great variety of linkage and branching occurrence) an d their ubiquitous location in the cell and extracellular matrix. Most glycans densely cover the outer cellular surface and are exposed to an environment of many proteins (such as growth factors, cytokines, toxins, enzymes and others). This particular position enables them to mediate several cell-cell or cell-matrix interactions and act as recognition determinants in a great variety of important cellular events ( adhesion, proliferation and differentiation). Therefore, the comprehension of the unique function of carbohydrates in biology calls for a structural characterization of saccharides, considered either as sugar chains or as integral part of glycoconjugates, and their interaction partners, such as carbohydratebinding proteins. The aim of the present work was essentially methodological in kind, i.e. directed towards the development of techniques and experimental procedures for the analysis (identification/detection, characterization, synthesis) of given carbohydrate sequences, frequently occurring in nature, glycoproteins and sugar-binding proteins. l. Derivatization of saccharides for UV -visible detection on capillary electrophoresis CE). Concerning the compositional monosaccharide determination of a hydrolyzed glycan pool, which is frequently the first step in oligosaccharide mapping, we developed and optimized experimental methods aimed to improve the UV -vis detection of monoand oligosaccharides present as widespread components in mammalian and plant glycoproteins. Saccharides inherently lack chromophores and must be suitably derivatized, i.e. covalently linked to a chromophore usualiy through reductive amination, prior to detection on CE. Derivatization yield, however, depends on the nature of the sugar residue and N-acetylamino sugars, such as N-acetylgalactosamine (GalNAc) and N-acetylglucosamine (GlcNAc), occupying the reducing end of respectively 0- and N-glycan sequences, are notoriously difficult to label. A comparative study aimed at the improvement of a labelling procedure, enhancing the detectability of N -acetylamino sugars, was therefore performed. The results showed that, among frequently used tagging dyes, 2-aminobenzoic acid turned out to be the most efficient, offering high and comparable sensitivity at the CE UV -vis detector for ali saccharides tested. Choosing adequate CE running conditions, a mixture of eleven monosaccharides was efficiently separated in a very short time frame. 2. Immobilized beta-galactosidase-based synthesis ofN-acetyllactosamine (LacNAc) The abovementioned findings of suitable derivatization procedure for the UV -vis detection of reducing sugars was successfully applied to the detection of LacNAc from crude reaction mixtures via CE. This disaccharide is very commonly located at the outermost portion of complex celi surface oligosaccharides of glycoconjugates and is therefore directly involved in recognition processes between cells. The enzymatic synthesis in heterogeneous phase of the LacNAc disaccharide was here attempted. The goal of the study was to improve both product yield and recovery of the biocatalyst through its immobilization on solid supports (commercial polymers Eupergit® and Sepabeads ), in comparison with the synthesis performed with the free enzyme in solution. The biosynthesis was performed through transglycosylation reaction with beta-galactosidase from B. circulans starting from p-nitrophenyl galactopyranoside as the glycosyl donor and GlcNAc as the acceptor. The reaction was followed monitoring via CE the disaccharide production in function of time. The kinetic analysis revealed that the immobilization procedure did not suppress catalytic activity, but, on the contrary, improved the galactose transfer efficiency of the enzyme. Kinetic profiles of the reactions performed with Eupergit® or Sepabeads were qui te different, suggesting that the physico-chemical properties of the supporting matrices infuence enzyme behaviour. Maximum LacNAc molar yield (64%) was obtained using Eupergit® as solid carrier. 3. Glycan structure analysis The identification and quantitative analysis of oligosaccharide portions of glycoproteins of biotechnological, like beta-glucosidase, or therapeutical interest, such as membrane proteins of hepatic celllines, was here carried out according to two alternative glycoprotein processing strategies. (3a) Prior to commercialization, recombinant beta-glucosidase (GCase) produced in transgenic tobacco seeds must be first thoroughly characterized in order to test whether the glycosylation pattem of the glycoprotein expressed in plant resembles that of the human placenta} GCase. Treatment with trifluoroacetic acid led to exhaustive hydrolysis of the glycan moiety and o n the released monosaccharides a CE analysis was carried out. Since neither fucose nor xilose, frequent sugar residues found in plant glycoproteins, could be detected, potential immunogenicity for delivery to humans was excluded. (3b) The scientific relevance of studying the glycan structure of membrane-bound proteins, differently expressed in a healthy and hepatoma cell line, relies on the possible detection of early-stage saccharidic tumoral markers. In order to develop and optimize a methodological procedure, based on the in-solution or in-gel enzymatic release (with PNGase F) of entire oligosaccharide chains and analysis through LC/MS of the hydrolizate, that could be exported to the glycosylation pattem analysis of membrane proteins, the commerciai bovine fetuin was here used as a model of a heavily glycosylated protein. From the obtained data, a biantennary and two triantennary oligosaccharide structures, pertaining to the N-glycosylation profile of Fet, could be finally assigned. 5. Large-scale biosynthesis of a Hpylori lysozyme Much effort has been devoted to the study of a Hpylori lysozyme (Lys), an enzyme that seems to play a key role in the autolysis of the bacterium during colonization of the gastric epithelium of primates. In order to attempt structural characterization of the protein, either alone or in complexed form with its saccharidic substrate (Nacetylmuramic acid), the scaling-up of the biosynthesis and the purification procedures were required. This could be performed after identification of the gene encoding the lysozyme sequence in Hpylori and cloning the DNA sequence of interest into a suitable expression vector and promoting high-level expression of the Lys in E. coli. Two expression systems were tested: the pGEX system yielded the Lys as fusion protein with GST, which was produced preferentially as inclusion body in E. coli host cells, especially when larger culture volumes were used. The protein was isolated from insoluble cell pellet, solubilized and successfully refolded. After purification and digestion with thrombin protease to remove the GST fusion tag from the Lys protein, tests for lytic activity were positive, even though separation could not be achieved. Binding studies performed through affinophoresis on these samples, unfortunately did not provide any evidence of the affinity of Lys fora disaccharide (LacNAc) mimicking the natural sugar substrate. The pET vector was therefore chosen for an alternative cloning strategy. The expression system gave the Lys with a C-terminai 6xHis-tag. Also in this case the protein was almost totally recovered in the insoluble fraction of the celllysate. After a refolding step, directly performed on Ni-NTA agarose affinity column, highly pure, but unactive protein was obtained. The whole work was carried out at the University of Trieste, department of Biochemistry, Biophysics and Macromolecular Chemistry in the laboratory of Prof. S.Paoletti under the supervision of Dr. A.Gamini, where I had the opportunity to learn the basic issues of analytical biochemistry. Within the Hpylori lysozyme project, a four-months stay at the Martin-Luther University of Halle-Wittenberg at the Institute of Biotechnology (Germany) in the laboratory of Prof. R.Rudolph and under the supervision of Dr. C.Lange was of great importance for getting expertise in the field of protein chemistry.