Spectroscopic analysis of the trinuclear cluster in the Fet3 protein from yeast, a multinuclear copper oxidase

The Fet3 protein (Fet3p) is a multinuclear copper oxidase essential for hig h-affinity iron uptake in yeast. Fet3p contains one type 1, one type 2, and a strongly antiferromagnetically coupled binuclear Cu(II)-Cu(II) type 3 co pper. The type 2 and type 3 sires constitute a structurally distinct trinuc lear cluster at which dioxygen is reduced to water. In Fet3p, as in cerulop lasmin, Fe(II) is oxidized to Fe(III) at the type 1 copper; this is the fer roxidase reaction that is fundamental to the physiologic function of these two enzymes. Using site-directed mutagenesis, we have generated type 1-depl eted (TID), type 2-depleted (T2D), and T1D/T2D mutants. None were active in the essential ferroxidase reaction catalyzed by Fe3p. However, the spectro scopic signatures of the remaining Cu(II) sites in any one of the three mut ants were indistinguishable from those exhibited by the wild type. Although the native protein and the TID mutant were isolated in the completely oxid ized Cu(II) form, the T2D and T1D/T2D mutants were found to be completely r educed. This result is consistent with the essential role of the type 2 cop per in dioxygen turnover, and with the suggestions that cuprous ion is the valence state of intracellular copper. Although stable to dioxygen, the Cu( I) sites in both proteins were readily oxidized by hydrogen peroxide. The d ouble mutant was extensively analyzed by X-ray absorption spectroscopy. Edg e and near-edge features clearly distinguished the oxidized from the reduce d form of the binuclear cluster. EXAFS was strongly consistent with the exp ected coordination of each type 3 copper by three histidine imidazoles. Als o, copper scattering was observed in the oxidized cluster along with scatte ring from a ligand corresponding to a bridging oxygen. The data derived fro m the reduced cluster indicated that the bridge was absent in this redox st ate. In the reduced form of the double mutant, an N/O ligand was apparent t hat was not seen in the reduced form of the T1D protein. This ligand in T1D /T2D could be either the remaining type 2 copper imidazole ligand (from His 416) or a water molecule that could be stabilized at the type 3 cluster by H-bonding to this side chain. If present in the native protein, this H2O co uld provide acid catalysis of dioxygen reduction at the reduced trinuclear center.