DTU Chemistry - PhD 2016

Short-cut Methodology for Drug Discovery

Methods for the isolation and synthesis of complex carbohydrates with pharmaceutical relevance have been improved.
Carbohydrates are highly important biomolecules involved in numerous diseases, but on the other hand also with great potential as medicine. For instance, synthetic versions of glyco-molecules have been developed as carbohydrate-based therapeutic agents. The field is known as glyco-science. The project contributes to the field by providing improved methods for the isolation and synthesis of complex carbohydrates.

In nature, the majority of carbohydrates are polysaccharides (cellulose, starch, chitin) or glyco-conjugates (glyco-peptides, glyco-lipids) in which monosaccharides are connected via glycosidic linkages. The reaction creating these linkages is known as glycosylation.

Creating synthetic version of carbohydrates – either as drug candidates or for scientific purposes – requires glycosylation, but this is not always straightforward. Especially now, when the glyco-biology field is expanding, there is a need for glycosylation methods that are both reliable and stereo-controlled (meaning that not just the chemical composition but also the physical structure of the molecule is controlled).

A major trend in the field is extensive use of protecting groups in order to achieve the desired control. These groups suppress glycosylation at undesired positions. The drawback is that additional steps are required to install and remove the protecting groups. Therefore clever manipulations which limit the need for protecting groups are in high demand.

In the project, the role of tin was investigated in regioselective glycosylation of 2,3,4,6-unprotected hexopyranosides with perbenzylated glycosyl bromide donors. Reactions with phenyl 1-thio-β-glucopyranoside and phenyl 1-thio-β-galactopyranoside with glucosyl and galactosyl bromide donors afforded exclusively the corresponding α(1→6)-linked disaccharides in decent yields. The coupling was highly dependent on the solubility of the acceptors with dibutyltin oxide in dichloromethane. Further, it was successful for glucose and galactose, while no coupling occurred with mannose. The same behavior was observed for donors, where no conversion took place with mannosyl bromide.

In a second stage of the project, diphenylborinic acid catalyzed glycosylation was conducted with the NIS/CSA promoter system. Yet again, poor solubility of the acceptors prevented the desired reactions. The only positive result was achieved when methyl 3-O-benzoyl-α-D-mannopyranoside was used as an acceptor. The glycosylation with borinic catalyst and this acceptor led to the (1→6)-linked disaccharide in a 45 % yield. However, glycosylation under the same conditions but without the catalyst afford the (1→6)-linked product in a 15 % yield. Overall, the project has shown that tin- and boron-mediated glycosylation with unprotected carbohydrates is successful, when the components of the reaction mixture are soluble in a given solvent. Insolubility of the unprotected acceptors puts a considerable limitation on the developed method. Therefore further research on the matter would be valuable for the field.

Illustration: Project 1 Tin-mediated glycosylation of unprotected hexopyranosides.

DTU Chemistry - PhD 2016

Robert Madsen

Funded by:
The project was funded by the Novo Nordisk Foundation.