Using X-ray information, Molecular Dynamics and Docking to predict Protein-monosaccharide complexes

MATIAS DI PAOLA; JUAN BLANCO; GAMARRA, MARCELO; CARLOS MODENUTTI

Buenos Aires

Simposio; 3rd Argentinian Symposium on Glycobiology; 2019

Universidad Nacional de San Martin - UNSAM -Buenos Aires

Molecular Docking application to predict lectin-carbohydrate complexes is challenging due to their low affinity, hydrophilic nature and ligand conformational diversity. In the present work we have analyzed conventional and solvent-site biased Autodock4 performance, concerning receptor conformational diversity as derived from different X-ray structures (apo and holo), Molecular Dynamics snapshots and Homology-based models, for 14 different lectin-carbohydrate complexes. In order to assess docking methods accuracy, we introduce the '2D-score' as a new technique to rank predicted poses. Our results show that both conventional and solvent-site biased docking yield accurate lectin-monosaccharide complexes when starting from Holo X-ray structures. Apo structures can be sometimes tricky, specially in such complexes with a high degree of 'induced fit', but solvent-site bias method can overcome this situation. Another strategy can be the construction of a middle-sized (50-100 structures) receptor conformational ensemble, derived either from Molecular Dynamics or from multiple homology-model building. When no structural information of the protein is reported, generally homology modelling can be successfully employed if a > %45 identity template is available. Finally, our results also show that docking methods can select the correct receptor structure within the structural ensemble, underscoring the relevance of joint evaluation of ligand pose and receptor conformation.