How Are Substrate Binding and Catalysis Affected by Mutating Glu127 and Arg161 in Prolyl-4-hydroxylase? A QM/MM and MD Study
High-valent metal-oxo oxidants are common reactive species in synthetic catalysts as well as heme and nonheme iron enzymes. In general, they efficiently react with substrates through oxygen atom transfer, and for a number of cases, experimental rate constants have been determined. However, because these rate constants are generally measured in a polar solution, it has been found difficult to find computational methodologies to reproduce experimental trends and reactivities. In this work, we present a detailed computational study into para-substituted thioanisole sulfoxidation by a nonheme iron(IV)-oxo complex. A range of density functional theory methods and basis sets has been tested for their suitability to describe the reaction mechanism and compared with experimentally obtained free energies of activation. It is found that the enthalpy of activation is reproduced well, but all methods overestimate the entropy of activation by about 50%, for which we recommend a correction factor. The effect of solvent and dispersion on the barrier heights is explored both at the single-point level and also through inclusion in geometry optimizations, and particularly, solvent is seen as highly beneficial to reproduce experimental free energies of activation. Interestingly, in general, experimental trends and Hammett plots are reproduced well with almost all methods and procedures, and only a systematic error seems to apply for these chemical systems. Very good agreement between experiment and theory is found for a number of different methods, including B3LYP and PBE0, and procedures that are highlighted in the paper.