Spectroscopic insights into the mechanism of anammox hydrazine synthase

Abstract

Anaerobic ammonium oxidizing bacteria make a living oxidizing ammonium with nitrite as electron acceptor, intermediates nitric oxide and hydrazine, and end product dinitrogen gas. Hydrazine is a biologically unique free intermediate in this metabolism, and is produced by the enzyme hydrazine synthase. Crystallization of Candidatus Kuenenia stuttgartiensis hydrazine synthase allowed for an initial hypothesis of its reaction mechanism. In this hypothesis, nitric oxide is first reduced to hydroxylamine after which hydroxylamine is condensed with ammonium to form hydrazine. Hydrazine synthase is a tetraheme cytochrome c, containing two proposed active site hemes ({gamma}I & I) in the {gamma}- and -subunit, respectively, connected by an intra-enzymatic tunnel. Here we combined the data from electrochemistry-induced Fourier transform infrared (FTIR) spectroscopy, EPR and optical spectroscopy to shed light on the redox properties and protein dynamics of hydrazine synthase in the context of its reaction mechanism. Redox titrations revealed two low potential low spin hemes with midpoint potentials of [~]-360 mV and [~]-310 mV for heme II and {gamma}II, respectively. Heme {gamma}I showed redox transitions in the range of 0 mV, consisting of both low spin and high spin characteristics in optical and EPR spectroscopy. Electrochemistry-induced FTIR spectroscopy indicated an aspartic acid ligating a OH-/H2O at the heme {gamma}I axial site as a possible candidate for involvement in this mixed spin characteristic. Furthermore, EPR spectroscopy confirmed the ability of heme {gamma}I to bind NO in the reduced state. Heme I exhibited a rhombic high spin signal, in line with its ligation by a proximal tyrosine observed in the crystal structure. Redox titrations down to -610 mV nor addition of dithionite resulted in the reduction of heme I, indicating a very low midpoint potential for this heme. In vivo chemistry at this heme I, the candidate for the comproportionation of hydroxylamine and ammonium, is thus likely to be initiated solely on the oxidized heme, in contrast to previously reported DFT calculations. The reduction potentials of the {gamma}-subunit hemes were in line with the proposed electron transfer of heme {gamma}II to heme {gamma}I for the reduction of NO to hydroxylamine (E0 = - 30 mV).