Uncovering Cryptic Pockets in Biologically Relevant Proteins: An Improved Computational Methodology

Abstract

Revealing and targeting novel cryptic pockets in a protein has become a major challenge in the field of drug design. Static experimental protein structures represent ensemble averaged lacking the dynamics of the real biological system. Molecular dynamics (MD) simulations serve as a “virtual microscope” to probe the dynamics of the biological structures based on Newton’s equations of motion. In this thesis I have demonstrated an improved simulation based technique that uses benzene molecules as probes and enables revealing cryptic pockets hidden in the experimental structures of proteins. In this method benzene rings contain additional dummy atom in its geometric center with an optimized Lennard-Jones potential. Initially the method has been benchmarked against experimentally well-defined wild-type and mutant K-Ras (G12D) protein, essential bio-substrate in development of many cancers. The application of the method resulted in the formation of “saddle” like benzene clusters around the pockets of protein. With the novel protocol I have shown that all pockets previously defined by experimental and in silico methods for both wild-type and mutant K-Ras have been observed with enlarged volume together with the exposure of a novel pocket. The method was shown to be efficient within only 100 ns of sampling. Subsequently the novel approach was applied on envelope (E) protein of all dengue virus (DENV) serotypes as well as Zika virus (ZIKV). The novel consensus pocket exposed was positioned in between two monomers and found in all dimers studied. This crucial region has a great potential for blocking dimer-to-trimer transition and viral infection process. Hence, it was a subject to subsequent virtual screening. Ten Food and Drug Administration (FDA) approved drugs (based on binding affinity and key interaction with both the monomers) have been shortlisted for the experimental validation. The most promising molecule named Itraconazole was shown to possess the highest affinity for ZIKV. In summary a general novel method for screening cryptic sites