In this dissertation research, two new computational tools were developed to model the biotransport of small organic molecules into the active site of broad-substrate specificity (BSS) enzymes. The biological organism selected to develop, test and validate these tools were Rieske non-heme iron dioxygenases. Members of this family of enzymes are known to have biocatalytic activity on more than three hundred different substrates. The large diversity of substrates that can be acted upon makes these enzymes very attractive in biotechnological processes such as bioremediation. In addition, the highly specific chirality of the products obtained makes these enzymes attractive for the potential synthesis of pharmaceutical precursors. Currently, the most common way to identify new substrates requires formulating an educated guess followed by the arduous task of testing each possible compound individually. This slows down the pace at which new industrial processes can be formulated or current ones further developed. The tools presented in this research provide fundamental and practical scientific contributions. For the basic science studies of my dissertation, an all-atom and, a coarse-grained (CG) model of Rieske non-heme iron dioxygenases were used to investigate the factors that affect the biotransport of small organic molecules into their active sites. From the all-atom model I discovered a gating mechanism that allows aromatic substrates into the active site and blocks other compounds. The key to these gates are T-stacked pi-pi interactions between hydrophobic amino acids and the aromatic substrates. On the other hand, from the CG model I discovered that the shape of tunnel modulates the hydrophobicity level of the surface. As the tunnels become more concave, the hydrophobicity increases causing the formation of a water exclusion zone which increases the diffusivity of aromatic substrates. The CG models also revealed that convex tunnels prevent the adhesion of hydrophobic substrates to the tunnel walls; providing a possible explanation for the evolution of bottlenecks at the entrance of Rieske active sites. For the practical contributions of my dissertation, I developed two new computational tools for the prediction of Rieske substrates. The first tool is an all-atom algorithm that models the stochastic roto-translational movement of small organic molecules along the Rieske enzyme tunnels. This algorithm has a 92% prediction accuracy of Rieske substrates. In addition, it is capable of elucidating the location of high-energy barriers along the tunnel, allowing the formulation of possible protein engineering sites. The second tool is a CG non dimensional model of the Rieske enzyme tunnels. This algorithm has a 90% prediction accuracy of Rieske substrates. The processing time of 1ms/substrate combined with its high accuracy allows for the high-throughput screening of possible Rieske substrates.
University of Minnesota Ph.D. dissertation. July 2019. Major: Mechanical Engineering. Advisors: Alptekin Aksan, Lawrence Wackett. 1 computer file (PDF); xii, 248 pages.
Development of computational tools for modeling the biotransport of small organic molecules into the active site of broad-substrate specificity enzymes.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.