Computer-assisted Design of Novel Polyketide Synthase 13 of Mycobacterium tuberculosis Inhibitors Using Molecular Modeling and Virtual Screening

Abstract

Aims: Polyketide synthase 13 (Pks13) is an essential enzyme in the synthesis of mycolic acids biosynthesis pathway of Mycobacterium tuberculosis (Mtb). Therefore, Pks13 is a promising drug target for tuberculosis treatment. Here we report the in silico design and evaluation of novel Pks13 inhibitors made of benzofuran derivatives with favorable predicted pharmacokinetic profiles. Methodology: A 3D model of Pks13-TAMx complexes was prepared for a training set of 18 TAMs with experimentally determined inhibitory potencies (half-maximal inhibitory concentrations IC \({ }_{50}^{\text {exp }}\) ) by using in situ modifications of the crystal structure of the TAM1-Pks13 complex (PDB entry 5V3X). A linear QSAR model was built, correlating computed gas phase enthalpies of formation \(\left(\Delta \Delta \mathrm{H}_{\mathrm{MM}}\right)\) of Pks13-TAMx complexes with the \(\mathrm{IC}_{50}^{\exp }\) in order to find active conformations of the 18 TAMs. Furthermore, taking into account the implicit solvent effect and entropy changes upon ligand binding, a superior QSAR model was brought forth, correlating computed complexation of Gibbs' free energies \(\Delta \Delta \mathrm{G}_{\text {com }}\). Using the active conformations of the training set TAMs, we built a pharmacophore model ( \(\mathrm{PH} 4)\) which was used to virtually screen novel analogs included in a virtual combinatorial library (VCL) of compounds containing benzofuran scaffolds. The PH4 model screened the VCL, which was formerly filtered by Lipinski's rule-of-five, in order to identify new benzofuran analogs. Results: Gas phase QSAR model: \(-\log _{10}\left(\mathrm{IC}_{50}^{e \mathrm{exp}}\right)=\mathrm{pIC}_{50}^{e \mathrm{xp}}=-0.1822 \times \Delta \Delta \mathrm{H}_{\mathrm{MM}}+6.9135, \mathrm{R}^2= 0.89\); superior aqueous phase QSAR model pIC\({ }_{50}^{e x p}=-0.2182 \times \Delta \Delta \mathrm{G}_{\mathrm{com}}+7.0853, \mathrm{R}^2=0.98\) and PH4 pharmacophore model: \(\mathrm{pIC}_{50}^{e \mathrm{exp}}=1.0003 \times \mathrm{pIC}_{50}^{\mathrm{pre}}-0.0019, \mathrm{R}^2=0.91\). The PH4-based screening retained 109 new TAM analogues. Finally, the predicted pharmacokinetic profiles of these new analogues were compared to current orally administered antituberculosis drugs, and the former were found to be almost 92 times more active than TAM2 \(\left(\mathrm{IC}_{50}^{e \mathrm{exp}}=0.12 \mu \mathrm{M}\right)\). Conclusion: This computational approach, which combines molecular mechanics and the Poisson-Boltzmann (PB) implicit solvation theory, the pharmacophore model, the analysis of Pks13-TAMs interaction energies, the in silic screening of VCL compounds, and the inference of ADME properties resulted in a set of new suggested Pks13 inhibitors.