Theoretical studies of cyclic adenosine monophosphate dependent protein kinase: native enzyme and ground-state and transition-state analoguesElectronic supplementary information (ESI) available: Complete citation for ref. 25; further computational and model details; extended comparisons of results (relative energies and geometries) from larger basis sets, DFT functionals including empirical dispersion corrections, and continuum-solvation model and gas-phase data; structural representation and 3-

The mechanisms of phosphoryl transfer enzymes have garnered considerable attention. Cyclic AMP-dependent protein kinase (cAPK) catalyzes the transfer of the γ phosphoryl group of ATP to the serine hydroxyl group of a peptide chain. Metal-containing fluoro species have been used as transition-state and ground-state analogues in a variety of phosphoryl transfer enzymes and have shed light on the nature of the requirements in the active site to catalyze phosphoryl transfer. For cAPK, we present computational studies of the mechanism of phosphoryl transfer and the structure and 19 F NMR spectra of various ground- (BeF 3 − ) and transition-state (MgF 3 − , AlF 4 − , and AlF 3 0 ) analogues. With native substrate, the phosphoryl transfer proceeds through a five-coordinate phosphorane transition state, i.e. , there is not a five-coordinate phosphorane intermediate. Comparisons of simulated and experimental 19 F NMR spectra show cAPK prefers a monoanionic analogue (MgF 3 − or AlF 4 − ) over a neutral analogue (AlF 3 ), supporting the charge balance hypothesis. The computed phosphoryl transfer in cAPK involves a five-coordinate phosphorus transition state and 19 F NMR shows metal-containing fluorine analogues of the phosphoryl species are monoanionic.