Abstract:
In the present thesis a method for solid phase synthesis of oligosaccharide mimetics has been developed, tested and verified in order to probe the possibility of binding the oligosaccharide mimetics to the carbohydrate recognizing domains of proteins.
For this purpose, three different solid phases, namely glass, cellulose and oxygen-derivatized polystyrene, were tested regarding their suitability for solid phase synthesis of glycopeptides as oligosaccharide mimetics. As a result it can be found that all three selected solid phases are in theoretically suitable for the synthesis, however, glass shows the best results in terms of resistance to the solvents used and the reproducibility of the derivatisation. Despite of the change of the solvent system the solvent resistance of polystyrene was found to be problematic. Cellulose shows in general the highest degree of derivatization, but the reproducibility is low. In addition, because of swelling of the solid phase during washing operations the completeness of the washing operations is difficult to ensure. Furthermore, the washing operations need much more time consuming more solvent than needed for glass as solid phase.
Finally, a comprehensive library of glycopeptide building blocks for the synthesis of oligosaccharide mimetics was prepared. For this deoxy-azido derivatives of glucose at positions 1, 2, 3, and 6, of galactose at positions 1, 2, 3, 4 and 6 were synthesized. In addition, the anomeric azides of glucosamine and galactosamine and the 6-deoxy-azido glucosamin were prepared. Furthermore a 9-deoxy-azido derivative and a azido-ethyl glycoside of sialic acid were synthesized. The azido-ethyl glycosides of glucose and galactose also were used. Furthermore, glucose and galactose were derivatised with azido-ethyl ethers at the hydroxy groups at the positions 2,4 and 6. In addition, 6-deoxy-azido glucose was used to introduce various groups such as methyl, phenyl, allyl, thioethyl or thiophenyl groups at the anomeric center, in order to investigate their influence on binding to proteins. The azido derivatives were coupled in a 1,3-dipolar cyloaddition with an alkyne in the backbone to the corresponding triazoles. The building blocks of the glycopeptide library were synthesized in yields of 62% to 96%.
Furthermore, backbones consisting of PNA or iso-asparagine were tested concerning the selectivity and the bonding strength using a marina-blue labeled PHA-E. For this purpose different tetrameric glycopeptides were used. It has been found that in the context of the compounds tested, the PNA backbone showed higher selectivities, and rates of binding. To determine the degree of binding as well of the not bound protein, as of the bound protein after denaturation and removal from the solid phase, were measured by UV spectroscopy.
With several monomers of the glycopeptide library, a glycopeptide microarray with 96 pentameres was created. With this concept of synthesis, the binding of a complex carbohydrate recognizing domain to the mimetics was proven and thus the method was verified. The protein used was texas red labeled cholera toxin. In conclusion, the developed method is valid for the synthesis of oligosaccharide mimetics and for the screening with corresponding proteins. Furthermore, initial statements about the suitability of individual building blocks for the screening with cholera toxin can be made.