A Computational Chemistry Approach to Cytochrome P450 Metabolism

Abstract

Drugs and other xenobiotic compounds exhibit different transformations upon entering the human body, most often starting with an oxidative reaction involving P450 enzymes. Hence, the effectiveness and half-life, and even the toxicity of these drugs is determined in part to their metabolism. Thus, it is imperative to produce better models, either experimental or theoretical, to facilitate the understanding and predict the behaviour of these processes. Herein, I present a computational study using hybrid quantum mechanics/molecular mechanics (QM/MM) and density functional theory (DFT) methodologies to model enzymatic metabolism reactions. The work particularly emphasises on the reactivity patterns of the active species of cytochrome P450, i.e. Compound I or the iron(IV)-oxo heme cation radical species, through the use of biomimetic models and enzymatic structures. I initially started the work with a thorough benchmark study on the reproducibility and limitations of DFT methods and compare a set of calculated free energy of activations against experimental reaction rates for a biomimetic FeIVO model. In particular, I focus on a range of density functional theory methods, basis sets, empirical dispersion corrections and solvation as well as entropic effects on the free energy of activation. Based on these studies a recommended set of methods and procedures is proposed. Thereafter, a series of projects explore the reactivity of biomimetic models of Compound I against a number of model substrates. Starting with a model of a carbene ligated iron(IV)-oxo system that shows catalytic properties dramatically altered with respect to P450 CpdI. Through DFT characterization and reactivity with a set of common substrates I establish the electronic and catalytic differences. A subsequent set of projects explore the chemistry of a set of iron(IV)-oxo biomimetic models of Compound I with either a pure computational or a mixed computational-experimental approach. Sound characterizations of the catalytic properties and mechanistic descriptions of such systems are presented often with comparisons to P450-Compound I. A gas phase electron deficient metalloporphyrin model, TPFFP+• is explored giving comprehensive evidence and rationalization into the distinctions between the mechanisms of aromatic vs aliphatic hydroxylation with common aromatic substrates. Further experiments and modelling were performed in a follow up project where a stable cationic intermediate is formed in contrast to the more common radical intermediate chemistry as seen in CpdI with the same substrate. Following this investigations, a rather more applied set of projects is presented. First the capabilities of a simplified P450 model are explored for the prediction of toxic metabolic products and sites of metabolism (SOMs) derived from P450-oxidations on a common phthalate derived substrate found frequently on cosmetic products and pharmaceutical formulations, revealing reaches and limitations of this approach. Lastly, a comprehensive DFT and QM/MM approach is employed to investigate the chemistry behind the co-factor independent oxidation of nogalamycin, a naturally occurring antibiotic. Thermochemical cycles, DFT models as well as QM/MM studies highlight possible oxidants in the reaction mechanism and propose relevant factors affecting the chemical reaction.

Date of Award	31 Dec 2018
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Samuel De Visser (Supervisor) & Andrew Munro (Supervisor)