Academic literature on the topic 'Carbohydrate force field'

The emergence of ST131 extra-intestinal pathogenic Escherichia coli that are resistant to multiple antibiotics is a growing international health concern. Infections are common, treatment options for antibiotic resistant bacteria are limited and there is no vaccine available. Polysaccharides serve key functions in immune response to bacterial infection. The Opolysaccharides present on the cell surface of gram negative bacteria are antigenic and are associated with specific bacterial serogroups. These are, therefore, a potentially effective target for vaccines. Most ST131 E. coli isolates express the O25b antigen and monoclonal antibodies that are specific to it have been isolated. The chemical structure of O25b has been characterized and differentiated from that of the previously known O25 (or O25a) variety. Relatively little is known about the conformations of O25a and O25b and how they differ, however. As conformation is a factor in antigen-antibody binding, differences between the conformations of these two antigens may be relevant to further research into carbohydrate targeted vaccines and diagnosis techniques for ST131:O25b bacteria. The conformations of polysaccharides are typically dynamic in solution and are difficult to determine empirically. Molecular dynamics simulation provides a means of estimating polysaccharide conformation but the results are critically dependent on the quality of the selected force field. Carbohydrate force fields have matured over the past few decades and CHARMM36 and GLYCAM06 are used extensively for the analysis of bacterial polysaccharides. Studies that compare results from these two widely used force fields are, however, still quite rare. Here we use molecular dynamics simulations of unacetylated, 3 RU oligosaccharide extensions to compare the CHARMM36 and GLYCAM06 force fields and to present an initial analysis of the conformations of the O25a and O25b E. coli antigens. We then apply CHARMM36 molecular dynamics simulation to analogous O- and N- acetylated oligosaccharide extensions to gauge the effect of these groups on the conformations of the two antigens and to compare O25a and O25b. Despite some differences, our CHARMM36 and GLYCAM06 simulations are largely in agreement regarding the conformation of O25a trimers without O- or N-acetylation. Both force fields predict extended, linear antigen conformations. Differences between the two force fields are noted in our analogous study of O25b however: GLYCAM06 favors a collapsed, globular oligosaccharide over a more extended molecule favored by CHARMM36; CHARMM36 and GLYCAM06 predict different preferred dihedral values for a conformationally important, main-chain ɑ-L-Rhap-(1->3)- β-D-Glcp bond; GLYCAM06 favors an anti-Ψ, anti-ω orientation of a side-chain β-D-Glc-(1->6)-ɑD-Glc bond over an anti-Ψ, syn-ω orientation favored by CHARMM36. These findings are in agreement with other studies that indicate the collapse of some oligosaccharides into metastable globular conformations during simulations with GLYCAM06. Our CHARMM36 simulations of O- and N-acetylated, 3 RU oligosaccharide extensions of O25a and O25b indicate large differences between the conformations of the two antigens: First, the O25b trimer favors either a compressed or extended helical conformation in solution whereas the O25a trimer favors a single, extended conformation. Second, O25a and O25b exhibit notably different dihedral values for conformationally important glycosidic bonds that correspond with the reported structural differences between the two antigens. Third, O- and N-acetylation is found to facilitate rotation about a key ɑ-D-Glcp-(1->3)-ɑ-L-Rhap2Ac bond in O25b that, in turn, facilitates the formation of compressed, helical O25b conformations. These compressed conformations are stabilized by intramolecular hydrogen bonds that involve O- and N-acetyl groups. Finally, N-acetyl groups appear to be shielded on the inside of the compressed O25b helix whereas O-acetyl groups appear to be exposed on the outside of the molecule. We postulate that these large conformational differences provide a rationale for the clinically noted differences in cross reactivity of monoclonal antibodies for the O25a and O25b antigens.