LAUSR

Flavodiiron nitric oxide reductases (FNORs) carry out the reduction of nitric oxide (NO) to nitrous oxide (N2O), allowing infectious pathogens to mitigate toxic levels of NO generated in the human immune response. We previously reported the model complex [Fe2(BPMP)(OPr)(NO)2](OTf)2 (1, OPr- = propionate) that contains two coplanar NO ligands and that is capable of quantitative NO reduction to N2O [White et al. J. Am. Chem. Soc. 2018, 140, 2562-2574]. Here we investigate, for the first time, how a distortion of the active site affects the ability of the diiron core to mediate N2O formation. For this purpose, we prepared several analogues of 1 that contain two monodentate ligands in place of the bridging carboxylate, [Fe2(BPMP)(X)2(NO)2]3+/1+ (2-X; X = triflate, 1-methylimidazole, or methanol). Structural data of 2-X show that without the bridging carboxylate, the diiron core expands, leading to elongated (O)N-N(O) distances (from 2.80 Å in 1 to 3.00-3.96 Å in 2-X) and distorted (O)N-Fe-Fe-N(O) dihedral angles (from coplanarity (5.9°) in 1 to 52.9-85.1° in 2-X). Whereas 1 produces quantitative amounts of N2O upon one-electron reduction, N2O production is substantially impeded in 2-X, to an initial 5-10% N2O yield. The main products after reduction are unprecedented hs-FeII/{Fe(NO)2}9/10 dinitrosyl iron complexes (DNICs). Even though mononuclear DNICs are stable and do not show N-N coupling (since it is a spin-forbidden process), the hs-FeII/{Fe(NO)2}9/10 DNICs obtained from 2-X show unexpected reactivity and produce up to quantitative N2O yields after 2 h. The implications of these results for the active site structure of FNORs are discussed.