The linear glycosaminoglycan (GAG) chondroitin sulfate (CS) is one of the most abundant proteoglycan (PG) carbohydrate biopolymers in the body. CS consists of repeating disaccharide units of β-D-glucuronic acid (GlcA) and β-N-acetylgalactosamine (GalNAc) and may contain hundreds of disaccharide units. Atomic-resolution simulations of large systems are computationally taxing so we aimed to find a more efficient method of modeling biologically relevant CS GAGs at atomic resolution. We therefore developed an algorithm that applies glycosidic linkage phi/psi dihedral angle probabilities from unbiased all-atom explicit-solvent molecular dynamics (MD) simulations of non-sulfated CS 20-mers. Our algorithm was used to construct non-sulfated CS 20-mer models with glycosidic linkage conformations mimicking those of the MD-generated CS 20-mers. To validate our program we compared end-to-end distances in the configurations constructed using our program and those generated by MD simulations. We noticed a subtle difference in end-to-end distances in MD-generated versus constructed 20-mer ensembles: the constructed ensemble tended toward slightly higher end-to-end distances indicating less compact polymer configurations. In an effort to explain the difference, we analyzed the dihedral angles of each monosaccharide ring in the 20-mer. Although these rings did not undergo any major conformational changes during simulation, we found that ring flexibility contributes to end-to-end distance. By incorporating glycosidic linkage dihedral angle probabilities and monosaccharide ring conformations from the MD-generated 20-mer ensemble, our program is able to construct biologically relevant CS polymer models.

Program to construct biologically relevant atomic resolution models of chondroitin sulfate biopolymers

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.