The Valence Bond Way: Reactivity Patterns of Cytochrome P450 Enzymes and Synthetic Analogs

Abstract

The preceding decade has witnessed an immense surge of activity in the bioinorganic chemistry of transition metal enzymes and synthetic analogs that model their operation. The wide range of research covers both experimental and theoretical investigations of structure and reactivity patterns. Theory, and especially density functional theory (DFT), has become a very useful tool, an important partner of experiment in resolving structural and mechanistic issues. This flare of activity has generated a great deal of knowledge on intermediates, transition states, barriers, rate constants, rate-equilibrium relationships, stereoselectivity, and so forth. This abundance of acquired knowledge has created the need for establishing order, namely, the outlining of broad generalizations, as well as the creation of a more-intuitive interface between experimental and theoretical data. The valence bond (VB) diagram model, originally developed for organic reactions, is such a theoretical framework that has the potential to guide the requisite generalizations in the field of bioinorganic chemical reactivity. In this Account, we briefly describe the principles of construction of VB diagrams for bioinorganic reactions, detailing applications in the booming research area of heme enzyme (specifically cytochrome P450) reactivity, and particularly two archetypal reactions of these enzymes, alkane hydroxylation and thioether sulfoxidation. For congruence with the lingua franca of bioinorganic chemistry, the VB model is formulated to create bridges to (i) the molecular orbital (MO) description, (ii) the oxidation state formulation of transition metal complexes, and (iii) widely used concepts such as the Bell-Evans-Polanyi (BEP) principle. The VB diagram model reveals the origins of the barrier, describes the formation of transition states and reaction intermediates, and allows the prediction of barrier heights and structure-reactivity relationships. Thus, from the VB diagram model, we can rationalize the mechanistic selection during alkane hydroxylation compared with thioether sulfoxidation, as well as the different behaviors of the spin states during the reactions with the active species of P450, the high-valent iron oxo species called compound I (Cpd I). Furthermore, the VB model leads to expressions that enable us to estimate barrier heights from easily accessible reactant properties, such as bond energies, ionization potential, and electron affinities. We further show that the model is not limited to these archetypal processes: its applicability is wider and more general. Accordingly, we outline the potential applications of these principles to other reactions of P450 (such as olefin epoxidation and arene hydroxylation) and to similar reactions of nonheme enzymes and synthetic models. The VB diagram model leads to a unified understanding of complex bioinorganic transformations, creates order in the data, and provides an important framework for making useful predictions.