Development of new inhibitors for the bacterial glycosyltransferase : computational docking, synthesis and bioassays

Abstract

Bacterial drug resistance caused by abuse and misuse of antibiotics represents a widespread problem in our modern society. There is an urgent need to discover new and novel antibacterial agents with new targets, otherwise the drug resistant bacteria will become a great threat to human health. The bacterial glycoltransferase (GT), which plays a vital role in the biosynthesis of cell wall using lipid II as substrate, has potential to become a novel antibacterial target as the sequence of its encoding gene is highly conserved in both wild-type and drug-resistant strains. Currently, there is no clinical drug targeting at this enzyme. The only known glycosyltransferase inhibitor, moenomycin, is used as a growth promoter in animal feed due to its poor bioavailability. After all, discovery of novel glycosyltransferase inhibitors is a challenging but rewarding research area. Computer-aided drug design (CADD) has been widely used in the discovery of new drugs against specific binding targets. This technique can simulate the binding process between receptors and ligands, and the binding affinity could be estimated by score function. Compared to the traditional method of drug development, CADD accelerates discovery speed, saves time, reduces cost and improves hit-rate. To discover new antibacterial agents with glycosyltransferase as the target, over 3,000,000 compounds were computationally screened against the GT domain of S. aureus penicillin binding protein 2 (PDB: 2OLV, 2.8Å) by Internal Coordinate Mechanics (ICM) software. A hit compound with final score -34.5, 2-(3-(2-carbamimidoylhydrazono)-2-oxoindolin-1-yl)-N-(m-tolyl) acetamide (compound GT10) was found to possess a weak antibacterial ability to S. aureus and B. subtilis (MIC, 192 μg mL¹). Saturation-transfer difference (STD)-NMR confirmed the interaction between GT10 and the GT domain of S. aureus. In order to optimize the structure of GT10, analogues of the hit compound with the basic isatin core were designed and successfully synthesized with acceptable yields. The structures of all these derivatives were confirmed by ¹H-NMR and mass spectrometry. The structure activity relationship was rationalized according to the MIC results of these GT10 analogues against S. aureus, B. subtilis and E. coli. The results revealed that substituents such as nitro, fluorine or ether at the meta-position on the phenyl ring of the N-phenyl amide group can improve the antibacterial ability. Besides, the guanidyl group attached to the isatin core is essential to the activity. To better understand the binding between GT domain and GT10 analogues, a docking simulation between GT10-22, GT10-27 and S. aureus GT domain was performed by ICM. This model shows that the phenyl group in the N-phenyl amide group and the isatin core establish hydrophobic interaction with Ile195, Val233 and Pro234 whereas the guanidyl group enhances binding by forming three H-bonds with Lys155 and Ser160 of the GT domain. Interaction between the phenyl rings of GT10-27 and the GT domain was confirmed by STD-NMR.