Enzymatic synthesis of non-natural trisaccharides and galactosides; Insights of their interaction with galectins as a function of their structure

Abstract

Galectins are a family of carbohydrate-recognizing proteins that by interacting with specific glycoepitopes can mediate important biological processes, including immune cell homeostasis and activation of tolerogenic circuits. Among the different members of this family, Galectin 1 and 3 have shown pro-tumorigenic effects, being overexpressed in numerous neoplasic diseases, proving to be relevant in tumor immune escape, tumor progression and resistance to drug-induced apoptosis. Thus, generation of specific glycosides that could inhibit their pro-tumorigenic ability by blocking their carbohydrate recognition domain is one of the current major challenges in the field. Considering that galectin-ligand binding strength is closely related to the ligand structure, analysis of this relationship provides valuable information for rational design of high-affinity ligands that could work as effective galectin inhibitors. Taking profit of the ability of glycosidases to catalyze transglycosylation reactions we achieved the enzymatic synthesis of β- D -Galp-(1 → 6)-β- D -Galp-(1 → 4)- D -Glcp (2), a mixture of β- D -Galp-(1 → 6)-β- D -Glcp-(1 → 4)- D -Glcp (5) and β- D -Galp-(1 → 3)-β- D -Glcp-(1 → 4)- D -Glcp (6), and finally benzyl β-D -galactopyranoside (9), with reaction yields between 16 and 27%. All the galactosides were purified, and characterized using 1 H and 13 C nuclear magnetic resonance spectroscopy. Docking results performed between the synthesized compounds and human Galectin 1 (hGal-1) and human Galectin 3 (hGal-3) showed that the replacement of a glucose moiety linked to the terminal galactose with a galactose moiety, decreases the affinity for these galectins. Moreover, regarding the interglycosidic bond the most favorable β-Gal linkage seems to be β(1 → 4) followed by β(1 → 3) and β(1 → 6) for hGal-1, and β(1 → 4) followed by β(1 → 6) and β(1 → 3) for hGal-3. These results were in accordance with the IC50 values obtained with in vitro solid phase inhibition assays. Therefore, docking results obtained in this work proved to be a very good approximation for predicting binding affinity of novel galactosides.