Computational investigation of structure-function relationship in metalloenzymes

Important processes such as carbon dioxide capture, hydrogen evolution and oxidation as well as bioremediation are of paramount importance for the development of an environmental-friendly economy. In nature, the reactions underlying these processes are naturally performed by metalloenzymes. The understanding of how they work will ease the development of important technological and biotechnological applications, such as bioinspired inorganic catalysts or more effective enzymes. In the present work, we have used computational methods to study structure-function relationships in metalloenzymes which catalyze important reactions involved in environment-friendly processes. In particular, we selected three cases of study; the formate dehydrogenase (FDH) from Escherichia coli, which catalyzes the conversion between carbon dioxide and formate, the [NiFe]-hydrogenase from Allochromatium vinosum, which performs hydrogen oxidation, and the phosphotriesterase (PTE) from Agrobacterium radiobacter that is able to hydrolyse organophosphates. Since the study of structure-function relationship in enzymes generally requires both a view on the reaction mechanism and on the functional dynamics, we had to employ different levels of theory. At first, we focused on the study of metallo-enzymes for which the catalytic mechanisms still have to be clarified in details, as FDH and a [NiFe]-hydrogenase. In particular, we studied the oxidation of formate to carbon dioxide at the Mo-binding cofactor in the active site of FDH, using Density Functional Theory (DFT). We employed for the first time a cluster model for the molybdopterin cofactor which is meant to best reproduce the stereoelectronic properties of the whole moiety. By testing several reaction pathways for the ratedetermining step of FDH catalytic cycle, we were able to propose a novel reaction mechanism which includes a β-hydride elimination step with a metal hydride intermediate. We then employed similar methods to study hydrogen oxidation in [NiFe]-hydrogenase. We have investigated the dihydrogen coordination mode and subsequent oxidation, using model clusters of different sizes. It turned out that the spin state and the distorted seesaw coordination geometry of the Ni ion are two crucial factors that tune the energetics and regiochemistry of H2 binding. We were able to propose a reaction pathway involving the oxidative addition of H2 followed by proton transfer to the sulfur atom of one of the terminally coordinated cysteines as the lowest energy one. We then focused on another metalloenzyme, in which experimental evidences support a role of conformational dynamics for the catalysis, PTE. In this enzyme, it has been shown that the substrate hydrolysis can occur on fast time-scales, whereas one of the rate limiting steps seems to be the transition between so-called "open" and "closed" conformations with respect to the metal-binding site. Indeed, in PTE mutations of residues, which are not directly involved in the active site, are known to affect the population of the closed and open states, as well as to alter the kcat. We thus studied this enzyme by atomistic Molecular Dynamics (MD) simulations to investigate the long-range structural communication routes between mutation sites and active site residues. We did so by considering both wild-type PTE (arPTE-WT) and a multiple mutant (arPTE-8M) in which mutations alter the equilibrium between the “closed” and “open” populations. By employing both freely available and in-house produced MD trajectories analysis tools, we were able to identify the major communication routes in the enzymes as well as to point out which mutation sites are more likely to be involved in the transmission of structural information. Comparing the communication pathways in the two variants allowed us to investigate how the presence of mutations influence the communication pathways in the protein and to relate them to the effects induce by the mutations on protein function. In conclusion, we have studied function-structure relationships in three different metalloenzymes, FDH, [NiFe] hydrogenase and PTE, by using high level DFT.