The gene encoding the cyanobacterial ferritinSynFtn is up-regulated in response to copper stress. Here, we show that, whileSynFtn does not interact directly with copper, it is highly unusual in several ways. First, its catalytic diiron ferroxidase center is unlike those of all other characterized prokaryotic ferritins and instead resembles an animal H-chain ferritin center. Second, as demonstrated by kinetic, spectroscopic, and high-resolution X-ray crystallographic data, reaction of O2with the di-Fe2+center results in a direct, one-electron oxidation to a mixed-valent Fe2+/Fe3+form. Iron–O2chemistry of this type is currently unknown among the growing family of proteins that bind a diiron site within a four α-helical bundle in general and ferritins in particular. The mixed-valent form, which slowly oxidized to the more usual di-Fe3+form, is an intermediate that is continually generated during mineralization. Peroxide, rather than superoxide, is shown to be the product of O2reduction, implying that ferroxidase centers function in pairs via long-range electron transfer through the protein resulting in reduction of O2bound at only one of the centers. We show that electron transfer is mediated by the transient formation of a radical on Tyr40, which lies ∼4 Å from the diiron center. As well as demonstrating an expansion of the iron–O2chemistry known to occur in nature, these data are also highly relevant to the question of whether all ferritins mineralize iron via a common mechanism, providing unequivocal proof that they do not.

This work describes the identification of two residues, D137 and E62, that are critical for, respectively, the transport of Fe2+ into, and Fe3+ out of, the catalytic sites of a prokaryotic ferritin.

The iron redox cycle in ferritins is not completely understood. Bacterioferritins are distinct from other ferritins in that they contain haem groups. It is acknowledged that the two iron motifs in bacterioferritins, the di-nuclear ferroxidase centre and the haem B group, play key roles in two opposing processes, iron sequestration and iron mobilisation, respectively, and the two redox processes are independent. Herein, we show that in Escherichia coli bacterioferritin, there is an electron transfer pathway from the haem to the ferroxidase centre suggesting a new role(s) haem might play in bacterioferritins.

Both O 2 and H 2 O 2 can oxidize iron at the ferroxidase center (FC) of Escherichia coli bacterioferritin (EcBfr) but mechanistic details of the two reactions need clarification. UV/ Vis, EPR, and Mçssbauer spectroscopies have been used to follow the reactions when apo-EcBfr, pre-loaded anaerobically with Fe 2+ , was exposed to O 2 or H 2 O 2. We show that O 2 binds di-Fe 2+ FC reversibly, two Fe 2+ ions are oxidized in concert and a H 2 O 2 molecule is formed and released to the solution. This peroxide molecule further oxidizes another di-Fe 2+ FC, at a rate circa 1000 faster than O 2 , ensuring an overall 1:4 stoichiometry of iron oxidation by O 2. Initially formed Fe 3+ can further react with H 2 O 2 (producing protein bound radicals) but relaxes within seconds to an H 2 O 2-unreactive di-Fe 3+ form. The data obtained suggest that the primary role of EcBfr in vivo may be to detoxify H 2 O 2 rather than sequester iron.