On ammonia binding to the oxygen-evolving complex of photosystem†II: a quantum chemical study.

A recent EPR study (M. Perrez Navarro et al., Proc. Natl. Acad. Sci. 2013, 110, 15561) provided evidence that ammonia binding to the oxygen-evolving complex (OEC) of photosystem II in its S2 state takes place at a terminal-water binding position (W1) on the "dangler" manganese center MnA. This contradicted earlier interpretations of (14)N electron-spin-echo envelope modulation (ESEEM) and extended X-ray absorption fine-structure (EXAFS) data, which were taken to indicate replacement of a bridging oxo ligand by an NH2 unit. Here we have used systematic broken-symmetry density functional theory calculations on large (ca. 200 atom) model clusters of an extensive variety of substitution patterns and core geometries to examine these contradictory pieces of evidence. Computed relative energies clearly favor the terminal substitution pattern over bridging-ligand arrangements (by about 20-30 kcal mol(-1)) and support W1 as the preferred binding site. Computed (14)N EPR nuclear-quadrupole coupling tensors confirm previous assumptions that the appreciable asymmetry may be accounted for by strong, asymmetric hydrogen bonding to the bound terminal NH3 ligand (mainly by Asp61). Indeed, bridging NH2 substitution would lead to exaggerated asymmetries. Although our computed structures confirm that the reported elongation of an Mn-Mn distance by about 0.15 Šinferred from EXAFS experiments may only be reproduced by bridging NH2 substitution, it seems possible that the underlying EXAFS data were skewed by problems due to radiation damage. Overall, the present data clearly support the suggested terminal NH3 coordination at the W1 site. The finding is significant for the proposed mechanistic scenarios of OEC catalysis, as this is not a water substrate site, and effects of this ammonia binding on catalysis thus must be due to more indirect influences on the likely substrate binding site at the O5 bridging-oxygen position.

Computation of hyperfine tensors for dinuclear Mn(III) Mn(IV) complexes by broken-symmetry approaches: anisotropy transfer induced by local zero-field splitting.

Based on broken-symmetry density functional calculations, the (55)Mn hyperfine tensors of a series of exchange-coupled, mixed-valence, dinuclear Mn(III) Mn(IV) complexes have been computed. We go beyond previous quantum chemical work by fully including the effects of local zero-field splitting (ZFS) interactions in the spin projection, following the first-order perturbation formalism of Sage et al. [J. Am. Chem. Soc. 1989, 111, 7239]. This allows the ZFS-induced transfer of hyperfine anisotropy from the Mn(III) site to the Mn(IV) site to be described with full consideration of the orientations of local hyperfine and ZFS tensors. After scaling to correct for systematic deficiencies in the quantum chemically computed local ZFS tensors, good agreement with experimental (55)Mn anisotropies at the Mn(IV) site is obtained. The hyperfine coupling anisotropies on the Mn(III) site depend sensitively on structural distortions for a d(4) ion. The latter are neither fully reproduced by using a DFT-optimized coordination environment nor by using experimental structures. For very small exchange-coupling constants, the perturbation treatment breaks down and a dramatic sensitivity to the scaling of the local ZFS tensors is observed. These results are discussed with respect to ongoing work to elucidate the structure of the oxygen-evolving complex of photosystem II by analysis of the EPR spectra.

Density functional calculations of (55)Mn, (14)N and (13)C electron paramagnetic resonance parameters support an energetically feasible model system for the S(2) state of the oxygen-evolving complex of photosystem II.

Metal and ligand hyperfine couplings of a previously suggested, energetically feasible Mn(4)Ca model cluster (SG2009(-1)) for the S(2) state of the oxygen-evolving complex (OEC) of photosystem II (PSII) have been studied by broken-symmetry density functional methods and compared with other suggested structural and spectroscopic models. This was carried out explicitly for different spin-coupling patterns of the S=1/2 ground state of the Mn(III)(Mn(IV))(3) cluster. By applying spin-projection techniques and a scaling of the manganese hyperfine couplings, computation of the hyperfine and nuclear quadrupole coupling parameters allows a direct evaluation of the proposed models in comparison with data obtained from the simulation of EPR, ENDOR, and ESEEM spectra. The computation of (55)Mn hyperfine couplings (HFCs) for SG2009(-1) gives excellent agreement with experiment. However, at the current level of spin projection, the (55)Mn HFCs do not appear sufficiently accurate to distinguish between different structural models. Yet, of all the models studied, SG2009(-1) is the only one with the Mn(III) site at the Mn(C) center, which is coordinated by histidine (D1-His332). The computed histidine (14)N HFC anisotropy for SG2009(-1) gives much better agreement with ESEEM data than the other models, in which Mn(C) is an Mn(IV) site, thus supporting the validity of the model. The (13)C HFCs of various carboxylates have been compared with (13)C ENDOR data for PSII preparations with (13)C-labelled alanine.