Evolutionary Guided Molecular Dynamics Driven Protein Design

Natures hardworking machines, proteins, are dynamic beings. Comprehending the role of dynamics in mediating allosteric effects is paramount to unraveling the intricate mechanisms underlying protein function and devising effective protein design strategies. Thus, the essential objective of this thesis is to elucidate ways to use protein dynamics based tools integrated with evolution and docking techniques to investigate the effect of distal allosteric mutations on protein function and further rationally design proteins. To this end, I first employed molecular dynamics (MD) simulations, Dynamic Flexibility Index (DFI) and Dynamic Coupling Index (DCI) on PICK1 PDZ, Butyrylcholinesterase (BChE), and Dihydrofolate reductase (DHFR) to uncover how these proteins utilize allostery to tune activity. Moreover, a new classification technique (“Controller”/“Controlled”) based on asymmetry in dynamic coupling is developed and applied to DHFR to elucidate the effect of allosteric mutations on enzyme activity. Subsequently, an MD driven dynamics design approach is applied on TEM-1 β-lactamase to tailor its activity against β-lactam antibiotics. New variants were created, and using a novel analytical approach called "dynamic distance analysis" (DDA) the degree of dynamic similarity between these variants were quantified. The experimentally confirmed results of these studies showed that the implementation of MD driven dynamics design holds significant potential for generating variants that can effectively modulate activity and stability. Finally, I introduced an evolutionary guided molecular dynamics driven protein design approach, integrated co-evolution and dynamic coupling (ICDC), to identify distal residues that modulate binding site dynamics through allosteric mechanisms. After validating the accuracy of ICDC with a complete mutational data set of β-lactamase, I applied it to Cyanovirin-N (CV-N) to identify allosteric positions and mutations that can modulate binding affinity. To further investigate the impact of mutations on the identified allosteric sites, I subjected putative mutants to binding analysis using Adaptive BP-Dock. Experimental validation of the computational predictions demonstrated the efficacy of integrating MD, DFI, DCI, and evolution to guide protein design. Ultimately, the research presented in this thesis demonstrates the effectiveness of using evolutionary guided molecular dynamics driven design alongside protein dynamics based tools to examine the significance of allosteric interactions and their influence on protein function.