Structural insights in galectin-1-glycan recognition: Relevance of the glycosidic linkage and the N-acetylation pattern of sugar moieties

Galectins, soluble lectins widely expressed intra- and extracellularly in different cell types, play major roles in deciphering the cellular glycocode. Galectin-1 (Gal-1), a prototype member of this family, presents a carbohydrate recognition domain (CRD) with specific affinity for beta-galactosides such as N-acetyllactosamine (beta-D-Galp-(1 -> 4)-D-GlcpNAc), and mediate numerous physiological and pathological processes. In this work, Gal-1 binding affinity for beta-(1 -> 6) galactosides, including beta-D-Galp-(1 -> 6)-beta-D-GlcpNAc-(1 -> 4)-D-GlcpNAc was evaluated, and their performance was compared to that of beta-(1 -> 4) and beta-(1 -> 3) galactosides. To this end, the trisaccharide beta-D-Galp-(1 -> 6)-beta-D-GlcpNAc-(1 -> 4)-D-GlcpNAc was enzymatically synthesized, purified and structurally characterized. To evaluate the affinity of Gal-1 for the galactosides, competitive solid phase assays (SPA) and isothermal titration calorimetry (ITC) studies were carried out. The experimental dissociation constants and binding energies obtained were compared to those calculated by molecular docking. These analyses evidenced the critical role of the glycosidic linkage between the terminal galactopyranoside residue and the adjacent monosaccharide, as galactosides bearing beta-(1 -> 6) glycosidic linkages showed dissociation constants six- and seven-fold higher than those involving beta-(1 -> 4) and beta-(1 -> 3) linkages, respectively. Moreover, docking experiments revealed the presence of hydrogen bond interactions between the N-acetyl group of the glucosaminopyranose moiety of the evaluated galactosides and specific amino acid residues of Gal-1, relevant for galectin-glycan affinity. Noticeably, the binding free energies (Delta G(bind)(calc)) derived from the molecular docking were in good agreement with experimental values determined by ITC measurements (Delta G(bind)(exp)), evidencing a good correlation between theoretical and experimental approaches, which validates the in silico simulations and constitutes an important tool for the rational design of future optimized ligands.