Identification of anti-cancer agents using integrated computational tools.
Mbatha, Sbongile Happyness.
MetadataShow full item record
Cancer is a heterogeneous disease that is responsible for various molecular changes and pathological entities that play vital roles in its response to treatment, survival, and growth. Better understanding into the critical pathways and molecular events involved in cancer has enabled the identification of novel targets and development of anti-cancer therapies. In this study, two enzymes that have been shown to be involved in different stages of cancer were used, namely cathepsin B and Hsp90. The main aim of this study was to employ integrated in-silico approaches to study inhibitory routes of these enzymes to develop novel anti-cancer agents. Cathepsin B is the most well studied of the cathepsin family as a potential therapeutic target to treat cancer. To accomplish practical aim 1 of this study, the Michael-acceptor type compounds obtained from a chemical database that irreversibly inhibit cathepsin B were investigated. Validation was carried out using compounds with experimentally determined anti-cathepsin B activity. Four novel compounds exhibited better covalent binding affinity when compared against the experimentally determined prototypes. Molecular dynamics simulations were performed to ensure the stability of the docked complexes and to allow further analysis. Per-residue interaction decomposition analysis provided deeper insight into the interaction themes with active site residues. It was found that polar and hydrophobic interactions had the highest contribution towards drug binding. Recent experimental studies have documented FDA-approved protease inhibitors involvement in anti-cancer activity, however, there was limited understanding to the mode of inhibition. To accomplish practical aim 2 of this study, the mode of inhibition of protease inhibitors against Hsp90 cysteine protein was investigated. The lack of an X-ray crystal structure of human Hsp90 prompted the creation of its homology model for subsequent simulations. Two possible binding sites, C-terminal and N-terminal domains, were identified and considered in this study. Molecular docking followed by molecular dynamic simulations and post-dynamic analyses were performed to elaborate on the binding mechanism and relative binding affinities of nine FDA-approved HIV-1 protease inhibitors against human Hsp90. Our findings from thermodynamics calculations revealed that these inhibitors were more likely to bind to the N-terminal domain (~ 54.7 -83.03 kcal/mol) when compared to C-terminal domain. This appears to be the first account of a detailed computational investigation aimed to understand the binding mechanism of HIV protease inhibitors binding to Hsp90. Information gained from this study should also provide a significant route map towards the design and optimisation of potential derivatives of protease inhibitors to treat breast cancer. The results obtained will serve as a powerful tool in the drug design and development process. However, further experimental investigations will be useful to improve our computational findings.