On the Mössbauer spectra of isopenicillin N synthase and a model {FeNO}⁷ (S = 3/2) system

We have carried out a series of density functional theory (DFT) calculations to predict the ⁵⁷Fe Mössbauer quadrupole splittings (ΔE_Q) and isomer shifts (δ_Fe) for the nitrosyl complex of isopenicillin N synthase with the substrate δ-(L-α-aminoadipoyl)-L-cysteinyl-D-valine (IPNS·ACV·NO) and an {FeNO}⁷ (S = 3/2) model system, FeL(NO)(N₃)₂ (L = N,N′,N″-trimethyl-1,4,7-triazacyclononane). B3LYP predictions on the model compound are in almost exact agreement with experiment. The same DFT methods did not enable the prediction of the experimental ΔE_Q and δ_Fe results for IPNS·ACV·NO when using the experimental protein crystal structure but did permit good predictions of ΔE_Q, δ_Fe, and the asymmetry parameter (η) when using a fully optimized structure. This optimized structure also enabled good predictions of the Mössbauer spectra of the photodissociation product of IPNS·ACV·NO. Mulliken and natural bonding orbital (NBO) spin density analyses indicate an electronic configuration of Fe^II (S = 2) anti-ferromagnetically coupled to NO (S = 1/2) in the protein as well as in the model system and the geometry optimized structure helps explain part of the enzyme reaction.