Computational tools to aid the design of glycomimetic agents

S4.2 Chemistry and biology of multivalent glycomimetics
Location (hall): 
Start/end time: 
Tuesday, July 2, 2019 - 11:45 to 12:00

Robert Woods1

1University Of Georgia, Athens, United States

Carbohydrate-binding proteins (human, bacterial or viral lectins) and carbohydrate-processing enzymes (glycosyltransferases and glycosidases) are important targets for therapeutic intervention, however the creation of drug-like molecules that can competitively inhibit carbohydrate-binding sites is uniquely challenging.  Computational approaches that are specifically designed to screen analogs of carbohydrates could be invaluable aids in both increasing the objectivity of the synthetic choices and in prioritizing the synthetic effort required for glycomimetic development.  The creation of such a tool is the focus of this presentation.

We present the development of an alternative strategy to ligand docking that leverages the benefits of computational modeling and structural biology.   Specifically, we are developing a computational approach that uses carbohydrate-protein co-crystal  structures as the basis for lead glycomimetic discovery by modifying the bound oligosaccharide in situ.  Interaction energies are computed and compared for several theoretical models (AutoDock v4.2, AutoDock Vina, and AMBER MM-GBSA).

We compare the performance of both AutoDock scoring functions and the MM-GBSA method applied to inhibitors of Galectin-3 and FimH, and conclude that empirical scoring methods offer important benefits over the more physics-based MM-GBSA approach.  While the preliminary results are encouraging, they also indicate that additional energy terms, such as cation-pi, will be necessary for improving performance.

The presentation will illustrate the degree to which in silico modeling can be applied in the screening and development of glycomimetic compounds.

A glycomimetic created by the addition of a derivative moiety (grey) to a carbohydrate scaffold (beige) shown in the binding site (blue surface) (from PDB ID 5NFA).  Once generated, the conformations available to the moiety will be searched (indicated by bond rotation arrows), and the lowest energy conformation identified and scored.