Novel monogalactoside galectin inhibitor scaffolds : Guiding selectivity with heteroaromatic interactions

Abstract: Carbohydrates are involved in many cellular processes, and most biomolecules are glycosylated. Thesemodifications are used in biological systems as information carriers, helping regulate organization on the cell surfaceand interactions between cells and the environment. Galectins are a family of carbohydrate binding proteins thatbind to polysaccharides containing a galactose. Galectins have the ability to crosslink glycosylated proteins –especially on the cell surface – giving galectins a role in modulating cell signalling and environmental interactions,influencing angiogenesis, immune regulation and cell adhesion. This implicates galectins in diseases like cancerand immune related disorders. Subsequently, many glycomimetics have been developed as galectin inhibitors,based on a variety of scaffolds, many with very high affinities, but selectivity between galectins remains a challenge.The galectin family of proteins has a very conserved binding motif, hence the differences in the binding pocket aresmall, making designing a selective inhibitor a challenge.We investigated C1-galactosides as possible galectin inhibitor scaffolds, exploiting one of the few differencesbetween galectin-1 and galectin-3 – histidine 52. We used C1-arylheterocycles to control the selectivity via theinteraction between the anomeric heterocycle and the histidine, an approach which turned out to be fruitful resultingin the inhibitors 1-naphthyloxazole galactose, a galectin-3 selective inhibitor with 90μM affinity and 2-fluorophenyltriazole galactose, a galectin-1 selective inhibitor with a 170 μM affinity with fivefold and eightfoldselectivity respectively. Extending the C1- system with a methylene linker resulted in the galectin-1 selective 4-fluorophenyltriazole 2-deoxygalactoheptulose, an inhibitor with 170 μM affinity and fourfold selectivity. In order topursue these 2-deoxyheptulose scaffolds we developed a diastereoselective hydroboration method for C1-exomethylene glycopyranosides. Combining C1-substitutions with substitution in position three on galactose with aphenyltriazole motif did not increase affinities in a straightforward way; instead of increasing affinity and perservingselectivity patterns set by the C1-substitutents, the disubstituted molecules emerged as galectin-4 selectiveinhibitors with affinities down to 2.3 μM and up to thirty-eightfold or better selectivity for galectin-4. This shows thatC1-galactosides can be selective galectin inhibitors with good affinities, but more work needs to be done tounderstand the interaction between substitution patterns. We also investigated aminpyrimidine substitutedgalactosides and identified compounds with a threehundred-fold selectivity for galectin-3 over galectin-1 andaffinities down to 1.7 μM. These results show that careful selection of heterocycles with an aim towards exploitingeven minute differences in the binding pocket can be effective in achieving selectivity.