Converged Structural and Spectroscopic Properties for Refined QM/MM Models of Azurin.

Blue copper proteins continue to challenge experiment and theory with their electronic structure and spectroscopic properties that respond sensitively to the coordination environment of the copper ion. In this work, we report state-of-the art electronic structure studies for geometric and spectroscopic properties of the archetypal "Type I" copper protein azurin in its Cu(II) state. A hybrid quantum mechanics/molecular mechanics (QM/MM) approach is used, employing both density functional theory (DFT) and coupled cluster with singles, doubles, and perturbative triples (CCSD(T)) methods for the QM region, the latter method making use of the domain-based local pair natural orbital (DLPNO) approach. Models of increasing QM size are employed to investigate the convergence of critical geometric parameters. It is shown that convergence is slow and that a large QM region is critical for reproducing the short experimental Cu-SCys112 distance. The study of structural convergence is followed by investigation of spectroscopic parameters using both DFT and DLPNO-CC methods and comparing these to the experimental spectrum using simulations. The results allow us to examine for the first time the distribution of spin densities and hyperfine coupling constants at the coupled cluster level, leading us to revisit the experimental assignment of the 33S hyperfine splitting. The wavefunction-based approach to obtain spin-dependent properties of open-shell systems demonstrated here for the case of azurin is transferable and applicable to a large array of bioinorganic systems.

Structure-Spectroscopy Correlations for Intermediate Q of Soluble Methane Monooxygenase: Insights from QM/MM Calculations.

The determination of the diiron core intermediate structures involved in the catalytic cycle of soluble methane monooxygenase (sMMO), the enzyme that selectively catalyzes the conversion of methane to methanol, has been a subject of intense interest within the bioinorganic scientific community. Particularly, the specific geometry and electronic structure of the intermediate that precedes methane binding, known as intermediate Q (or MMOHQ), has been debated for over 30 years. Some reported studies support a bis-µ-oxo-bridged Fe(IV)2O2 closed-core conformation Fe(IV)2O2 core, whereas others favor an open-core geometry, with a longer Fe-Fe distance. The lack of consensus calls for a thorough re-examination and reinterpretation of the spectroscopic data available on the MMOHQ intermediate. Herein, we report extensive simulations based on a hybrid quantum mechanics/molecular mechanics approach (QM/MM) approach that takes into account the complete enzyme to explore possible conformations for intermediates MMOHox and MMOHQ of the sMMOH catalytic cycle. High-level quantum chemical approaches are used to correlate specific structural motifs with geometric parameters for comparison with crystallographic and EXAFS data, as well as with spectroscopic data from Mössbauer spectroscopy, Fe K-edge high-energy resolution X-ray absorption spectroscopy (HERFD XAS), and resonance Raman 16O-18O difference spectroscopy. The results provide strong support for an open-core-type configuration in MMOHQ, with the most likely topology involving mono-oxo-bridged Fe ions and alternate terminal Fe-oxo and Fe-hydroxo groups that interact via intramolecular hydrogen bonding. The implications of an open-core intermediate Q on the reaction mechanism of sMMO are discussed.

Systematic High-Accuracy Prediction of Electron Affinities for Biological Quinones.

Quinones play vital roles as electron carriers in fundamental biological processes; therefore, the ability to accurately predict their electron affinities is crucial for understanding their properties and function. The increasing availability of cost-effective implementations of correlated wave function methods for both closed-shell and open-shell systems offers an alternative to density functional theory approaches that have traditionally dominated the field despite their shortcomings. Here, we define a benchmark set of quinones with experimentally available electron affinities and evaluate a range of electronic structure methods, setting a target accuracy of 0.1 eV. Among wave function methods, we test various implementations of coupled cluster (CC) theory, including local pair natural orbital (LPNO) approaches to canonical and parameterized CCSD, the domain-based DLPNO approximation, and the equations-of-motion approach for electron affinities, EA-EOM-CCSD. In addition, several variants of canonical, spin-component-scaled, orbital-optimized, and explicitly correlated (F12) Møller-Plesset perturbation theory are benchmarked. Achieving systematically the target level of accuracy is challenging and a composite scheme that combines canonical CCSD(T) with large basis set LPNO-based extrapolation of correlation energy proves to be the most accurate approach. Methods that offer comparable performance are the parameterized LPNO-pCCSD, the DLPNO-CCSD(T0 ), and the orbital optimized OO-SCS-MP2. Among DFT methods, viable practical alternatives are only the M06 and the double hybrids, but the latter should be employed with caution because of significant basis set sensitivity. A highly accurate yet cost-effective DLPNO-based coupled cluster approach is used to investigate the methoxy conformation effect on the electron affinities of ubiquinones found in photosynthetic bacterial reaction centers. © 2018 Wiley Periodicals, Inc.