Pharmacophore-based screening of autoinducer-2 inhibitor in bacteria for prevention of oral biofilm formation: An exploratory study

Aim: The aim of this study was to predict the 3D structure of luxS of Fusobacterium nucleatum and identification of key residues by in silico homology modeling and molecular docking for inhibiting autoinducer-2 (AI-2), responsible for biofilm formation. Materials and Methods: The current study was to explore 3D structure of luxS protein along with its major residues that are directly responsible for the activity by homology modeling employing Discovery Studio V4.0. Further, a ligand-based pharmacophore was generated, employing the known ligands to control AI-2 producing enzymes and to ultimately regulate biofilm formation in the oral environment subjecting to key bioinformatics tools. Results: The multiple sequence alignment study showed 26 and 56% similarity in target and template sequences of luxS protein, respectively. Homology modeling indicated high structural similarity between the two luxS proteins (homocysteine and 4, 5-dihydroxy-2, 3 pentanedione) with a low root mean square value of 0.6 angstrom, in addition to leading to major mutations that are responsible for the production of AI-2. A pharmacophore model was generated as a template for virtual screening to explore potent luxS inhibitor. ZINC15 database was used for similarity search algorithms and screening against a generated pharmacophore model. During molecular docking, a molecule was structurally identified with a high libdock score and significant binding energy, in addition to establishing a regulatory mechanism with the receptor protein. Conclusions: This investigation paves way for the high throughput virtual screening to characterize luxS and associated proteins resulting in considerable minimization of time and funds before taking up biological confirmatory tests in the wet laboratories during exploration of possible inhibitors.