Identification of PBP3 Inhibitors by Covalent Docking-Based Virtual Screening

Background/Aim: Antimicrobial resistance to medications is recognized as a worldwide public health issue. Penicillin Binding Proteins (PBPs) are found in all bacteria and are important targets in anti-biotherapy, particularly for commonly used β-lactam antibiotics.
Staphylococcus aureus

is a common gram-positive pathogen and is difficult to treat clinically, especially a methicillin-resist- ant form.

S. aureus

PBP3 serves an important function in accelerating the cross-linking of cell wall peptidoglycan dur- ing cell division. PBP3 was thus selected as a promising therapeutic target in order to find effective inhibitors that may be used as antibacterial medication options. In the light of this information, it is aimed to computationally determine PBP3 covalent inhibitors in drug molecules (FDA-approved, experimental, and research) in the Drugbank database.

Materials and Methods: All molecules (~10767) in the Drugbank database were downloaded and 2D structures were converted into 3D structures by preparing them with the LigPrep module. PBP3 crys- tal structure was prepared using the Protein Preparation Wizard. Virtual screening was performed with the Glide covalent docking algorithm and then the clustering with the Tanimoto similarity ma- trix resulting from ligand-based similarities. Hit compound were found by evaluating clustering and covalent docking results. Next, Desmond was used to analyze molecular dynamics (200 ns) of the hit and Cefotaxim (Reference)-PBP3 complexes.

Results: As a result of the virtual screening, docking scores were examined, interactions in the clus- ters were evaluated, and Terbogrel was selected as the hit drug molecule. It was determined that Ter- bogrel formed a covalent bond with Ser392 and found to be in the binding cavity during the period of simulation to be stable and to have at lower RMSD (Cα and (lig) fit on prot) value compared to the reference compound.

Conclusion: In this study, it was found that Terbogrel could be a potential novel PBP3 covalent inhib- itor for the treatment of bacterial infections.

Atatürk University BAP unit (ProjectID: 10785) financed this research. Keywords: PBP3, Covalent Doking, Molecular Simulation, Virtual Screening