Statistical analyses of the oxidized P-clusters in MoFe proteins using the bond-valence method: towards their electron transfer in nitrogenases

Abstract

26 well selected oxidized P-clusters (P2+) from the crystallographic data deposited in the Protein Data Bank have been analysed statistically by the bond-valence sum method with weighting schemes for MoFe proteins at different resolutions. Interestingly, the oxidation states of P2+ clusters correspond to Fe2 3+Fe6 2+ with high electron delocalization, showing the same oxidation states as the resting states of P-clusters (PN) in nitrogenases. The previously uncertain reduction of P2+ to PN clusters by two electrons was assigned as a double protonation of P2+, in which decoordination of the serine residue and the peptide chain of cysteine take place, in MoFe proteins. This is further supported by the obviously shorter α-alkoxy C—O bond (average of 1.398 Å) in P2+ clusters and longer α-hydroxy C—O bond (average of 1.422 Å) in PN clusters, while no change is observed in the electronic structures of Fe8S7 Fe atoms in P-clusters. Spatially, the calculations show that Fe3 and Fe6, the most oxidized and most reduced Fe atoms, have the shortest distances of 9.329 Å from the homocitrate in the FeMo cofactor and 14.947 Å from the [Fe4S4] cluster, respectively, and may well function as important electron-transport sites.