The ferrous square-pyramidal [Fe(NHis)4(SCys)] site of superoxide reductases (SORs) has been shown to reduce superoxide at a nearly diffusion-controlled rate. The final products of the reaction are hydrogen peroxide and the ferric hexacoordinated SOR site, with a carboxylate group from a conserved glutamate serving as the sixth ligand trans to the cysteine sulfur. A transient intermediate absorbing at approximately 600 nm in the reaction of the ferrous pentacoordinated site with superoxide has been proposed to be a ferric-(hydro)peroxo complex (Coulter, E.; Emerson, J.; Kurtz, D. M., Jr.; Cabelli, D. J. Am. Chem. Soc. 2000, 122, 11555-11556.). In the present study, DFT and ZINDO/S-CI results are shown to support the description of the 600-nm intermediate as an end-on, low-spin ferricperoxo or--hydroperoxo complex. Side-on peroxo coordination was found to be significantly less stable than end-on because of constraints on the imidazole ligand ring orientations imposed mostly by the protein. The modeled ferric-hydroperoxo complex had a decidedly nonplanar CysC beta-S-Fe-O-O geometry that appears to be imposed by the same constraints. A single prominent visible absorption of the (hydro)peroxo model is shown to be due mainly to a CysS-->Fe(III) pi charge transfer (CT) transition with a minor portion of His-->Fe(III) pi CT character and very little peroxo-->Fe(III) CT character. On the basis of calculations of models with various mono- and diprotonated peroxo ligands, protonation of the iron-bound peroxo oxygen is a key step in the decay of the ferric(hydro)peroxo complex favoring release of hydrogen peroxide over cleavage of the O-O bond, as occurs in the heme structural analogue, cytochrome P450.

No abstract

The details of the heme‐thiolate nitric oxide reductase (P450nor) catalytic mechanism are still controversial. One theory, supported by computational results [D. L. Harris, Int. J. Quantum Chem. 2002, 88, 183−200], assumes two sequential one‐electron transfers from NAD(P)H to an initial [FeNO]6 complex. The [FeNO]8 species thus formed would react with NO, eventually liberating the unstable ONNO2− anion (most probably in its protonated form), which decomposes to N2O and water. However, more recent experimental results [A. Daiber et al., J. Inorg. Biochem. 2002, 88, 343−352] suggest the first committed step of the mechanism to be direct hydride transfer from NAD(P)H to [FeNO]6, presumably resulting in an iron‐bound HNO unit, [Fe‐(H)NO]8, that would be readily protonated to [Fe‐(H)NOH]8. Subsequent NO addition would yield the unstable HO‐N(H)‐N=O, which would dissociate from the heme and decompose to H2O and N2O. Here, the DFT geometry optimization of all previously proposed reaction intermediates is reported. The first step of the mechanism is predicted to be hydride transfer to [FeNO]6, to produce [FeNOH]8 or [Fe‐N(H)O]8. Subsequent addition of NO to [Fe‐NOH]8 (but not to [Fe‐N(H)O]8 or [Fe‐N(H)OH]8) is predicted to lead to immediate liberation of HN2O2−, without any stable intermediates. Contrary to what would be predicted according to the “thiolate push effect” dogma, the thiolate ligand at the heme active site is shown to obstruct NO reduction, rather than facilitate it. It is in fact shown that replacement of the thiolate by a neutral nitrogen ligand (i.e., lysine, as found in the active site of cytochrome c nitrite reductase, an enzyme that can reduce NO) clearly favors, from a thermodynamic point of view, NO reduction at the heme site. (© Wiley‐VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003)