In silico Approaches for Targeting an Essential Mycobacterial Protein, PrsA

Abstract

Mycobacterium tuberculosis is a leading cause of death worldwide and with increasing cases of drug-resistance, new treatments options are required. PrsA is an essential protein in mycobacteria and is necessary for arabinogalactan production and nucleotide biosynthesis. This protein is not currently targeted by any drugs in clinical development and represents a potential drug development avenue. Currently, no crystal structure of M. tuberculosis PrsA (MtPrsA) is available, hence the SWISS-MODEL structure, based on M. smegmatis (93% similarity) was utilised. Molecular Dynamic simulations were performed on the modelled structure, to gain information on the protein’s predicted flexibility. The output trajectory was clustered, and representatives of each major cluster were used in downstream analysis. Ensemble docking, molecular docking on the ensemble of MtPrsA structures, of the GSK-177 prioritised compounds was undertaken. The docking predictions were also re-evaluated using machine learning based programmes, including NNScore. This approach yielded several compounds which could be taken forward for further computational analysis. The compound rankings were also retroactively compared to experimental data studying the GSK-177 compounds’ interactions with mycobacteria. This research represents a potential workflow for future in silico drug development efforts against M. tuberculosis and further work may allow the identified compounds to be used for the treatment of TB.