The Pharmacophore Model for the Antistaphylococcal Activity Screening Among Thiazolidinone-Related Structures

Abstract

Aim. To develop a pharmacophore model suitable for the antistaphylococcal activity screening among thiazolidinone, thiopyrano[2,3-d]thiazole and thiazolo[4,5-b]pyridine derivatives. Results and discussion. The best pharmacophore model in the series of models developed has a planar structure and consists of an aromatic ring (or cycle with π-bonds), a hydrophobic region, a projection of a hydrogen bond donor and two projections of a hydrogen bond acceptor. Its classification accuracy is 72.4 %. The diameter line of the model is formed by the projection of the hydrogen bond donor and the projection of the hydrogen bond acceptor, and its length is 8.05 Å. The analysis of conformations of active compounds consistent with the pharmacophore mode revealed two different ways of spatial arrangement of active molecules, in which the conditions of the pharmacophore model are fully met. Experimental part. The antistaphylococcal activity was determined by the agar diffusion method against methicillin-resistant strain of Staphylococcus aureus (MRSA) and evaluated by measuring the diameter of the microbial growth inhibition zone. The plates were incubated for 24 h at 37 °C. Compounds with the growth inhibition diameter of more than 7.5 mm were considered to be active. Computer processing of the results of the microbiological experiment and modeling of the probable pharmacophore were performed in the MOE software environment version 2007.09. The geometry of the compounds was optimized by molecular mechanics using the MMFF94x force field. Accuracy of classification was used as the main quality criterion of the pharmacophore model. Conclusions. The pharmacophore model developed can be used for virtual screening of the antistaphylococcal activity for the compounds similar to the training sample. When applying it to the in-home database of compounds the enrichment factor is EF = 2.05. Received: 27.09.2020 Revised: 23.10.2020 Accepted: 03.11.2020

Keywords

• chemoinformatics
• 4-thiazolidinones
• pharmacophore modeling
• antibacterial activity