Multiple Low-Lying States for Compound I of P450cam and Chloroperoxidase Revealed from Multireference Ab Initio QM/MM Calculations.

The hybrid CASPT2/MM approach is employed to systematically study the ground and low-lying excited states of the ultimate active species of the enzymes P450cam and chloroperoxidase (CPO): the oxoiron(IV)-porphyrin cation-radical Por(•+)Fe(IV)═O(Cys) species, the so-called Compound I (Cpd I). The results underscore the fact that the B3LYP/MM method is quite accurate on the most part. However, the CASPT2/MM energies for the ferryl-pentaradicaloid quartet state and the perferryl Fe(V)O doublet and quartet states are significantly lower than the B3LYP/MM results. Thus, while the present CASPT2/MM may still overestimate the stability of these states, nevertheless, taken at its face value, the result raises the question whether these states actually contribute to the reactivity of Cpd I. Our paper tries to grapple with this question in view of (a) the recent speculations that the perferryl Fe(V)O states may be involved in unusual reactivities of Cpd I species (Pan, Z. Z.; Wang, Q.; Sheng, X.; Horner, J. H.; Newcomb, M. J. Am. Chem. Soc. 2009, 131, 2621-2628) and (b) the DFT/MM results which show that the pentaradicaloid states have intrinsically low barriers for H-abstraction (Altun, A.; Shaik, S.; Thiel, W. J. Am. Chem. Soc. 2007, 129, 8978-8987). The application of CASPT2/MM to high valent transition metal states like the perferryl are far from being trivial, and the experience and insight gained in this study are expected to be helpful for future successful application of this type of method to resolve key issues in P450 reactivity.