Galectin-1 (Gal-1), a member of a family of highly conserved carbohydrate-binding proteins, has been reported to be a potent immunosuppressive and anti-inflammatory factor. It has been proposed that the presence of galectin-1 in organs such as the eye, the placenta, reproductive organs (testis, ovaries) and also in tumours might confer a status of immune privilege to these vulnerable sites. Up-regulated expression of this protein may contribute to preserve an immunosuppressive microenvironment by inducing apoptosis of inflammatory and effector T cells and skewing the balance of the local immune response towards a Th2 cytokine profile. Different glycosyltransferase enzymes are involved in the generation of specific saccharide ligands required for Gal-1 binding and Gal-1-induced cell death. Interestingly, expression of these enzymes is highly regulated during T cell development, activation and apoptosis. The aim of this review is to explore the potential link between glycosylation of immune cell subsets, susceptibility to Gal-1 and the generation of immune privilege, particularly in the context of tumor-immune escape.
Supported by progress in the synthesis of glycopeptides and detailed analyses by NMR in the last decade, a better understanding has developed as to how glycosylation affects the peptide backbone conformation. In short, the attachment of carbohydrate moieties to a peptide affects the three-dimensional structure of the peptide backbone through interactions between the carbohydrate and the peptide portions. Although, superficially, the outcomes range from a real change in the local three-dimensional structure with specific contacts between the carbohydrate and the peptide atoms, which is easy to observe by analytical methods, to a subtle change in the conformational space caused by the excluded volume effect of carbohydrate without specific atomic contacts, which is difficult to observe by analytical methods, it is generally understood that glycosylation has a certain effect on the peptide backbone conformation. The apparent difference in conformational change depends on the structure of the glycosylated amino acid residue, the structure of the attached carbohydrate, and the glycosylation site or the amino acid sequence around it. Long range interactions between the carbohydrate and the peptide portions depend on the larger three-dimentional structure of the peptide or protein. In view of these general perspectives, our own results, obtained for glycosylated calcitonin derivatives, were revisited.
Although computer modeling of carbohydrates has had a long history, almost comparable to that of protein modeling, there are still some technical problems to be solved at present due to their complexity and versatility. Carbohydrates which comprise many polar functional groups, as represented by the hydroxyl group, exhibit significant flexibility in their three-dimensional structures. They change the electronic configurations depending on the three-dimensional configurations and conformations resulting in the specific stereochemical features, such as anomeric effect, exoanomeric effect, and gauche effect. In computer modeling, these stereochemical features of carbohydrates have required developments of the force fields and/or the parameter sets designed to reproduce them. The hydroxyl groups on a sugar residue form a hydrogen bond with water molecules and the solvation environment affects conformational behavior of the carbohydrate molecule. In fact, a stable water-mediated interresidue hydrogen bond has been detected in the molecular dynamics simulations for solvated carbohydrate molecules. Since carbohydrate molecules readily exhibit various conformations arising from the intrinsic flexibility of glycosidic linkages, conformational analysis was required to systematically explore possible conformations and evaluate their steric energies. On the other hand, carbohydrate molecules often generate a large and complex conformational space derived from diversities of sugar residues and substitutents, and formation of branching structures. Several algorithms intended for an effective search for such a conformational space have been examined.
Glycosyltransferases are found in most living organisms, and they are involved in the biosynthesis of carbohydrates catalyzing the transfer of a saccharide residue to a specific acceptor. Despite the important roles that carbohydrates play in numerous biological events, the molecular details of the catalytic mechanism of these enzymes are generally not well understood, and consequently the progress in development of therapeutics based on their inhibitions has been relatively slow. High-level quantum mechanical calculations can be used to gain some insight into characteristics of the enzymatic reactions, and they lead to a deeper understanding of glycosyltransferases catalysis. In addition, the computed transition state structures provide a rational basis for design of transition state analog inhibitors.