In Silico Screening for inhibitors against Apicoplast Phosphate Translocator from Toxoplama gondii

Abstract

Apicomplexa parasites, including Toxoplasma gondii and Plasmodium falciparum, contain a secondary endosymbiosis-derived plastid like organ, called apicoplast, which is an anabolic hub. This apicoplast is fueled by phosphate translocator (APT), which transport phosphorylated sugar molecules in exchange of inorganic phosphate. Disruption of APT in T. gondii was found to be lethal for parasite. Beside this, its’s plastidic nature and location in apicoplast, made it an ideal drug target. In this study two homology models of TgAPT were used for predicting putative inhibitors against this protein by combined ligand-based virtual screening (LBVS) and structure-based virtual screening (SBVS) approaches. Before doing the actual screening, a homology model of another APT, called PfoTPT from P. falciparum was generated to compare the binding pocket and the binding of known ligands by docking. The binding pocket of TgAPT was also compared with other plastidic phosphate translocator classes. The comparison revealed that there was only one amino acid different between two APTs, but several differences between the APTs and pPT classes and these differences are assumed to contribute to differences in substrate recognition and binding. Then, Known substrates, non-substrates and inhibitors were docked in two TgAPT models and PfoTPT model. The non-substrates are those which are not usually transported, nor they inhibit the transport process. The PfoTPT model did not show good result in terms of scoring and rank ordering of compounds. Of the two TgAPT models, TgAPT_5y79 showed comparatively better result, so induced fit docking (IFD) was done in this model with 3- phosphoglyceric acid (3-PGA), phosphoenol pyruvate (PEP), pyridoxal-5-phosphate (PLP) and 2,4,6- trinitrobenzene sulfonate (TNBS) for generating better conformation. Then one of the poses generated with 3-PGA IFD was selected for the SBVS approach. In VS approach, analogs of substrates and inhibitors were retrieved from PubChem database and docked into the IFD generated pose. From this docking, 318 compounds were sorted from different analog groups and compounds of each group were clustered by hierarchical clustering. Finally, 29 compounds were predicted as putative inhibitor of TgAPT based on the docking score and their interaction with the protein. These compounds will be tested in vitro for the inhibition potential.