Refining the active site structure of iron-iron hydrogenase using computational infrared spectroscopy.

Iron-iron hydrogenases ([FeFe]H2ases) are exceptional natural catalysts for the reduction of protons to dihydrogen. Future biotechnological applications based on these enzymes require a precise understanding of their structures and properties. Although the [FeFe]H2ases have been characterized by single-crystal X-ray crystallography and a range of spectroscopic techniques, ambiguities remain regarding the details of the molecular structures of the spectroscopically observed forms. We use density functional theory (DFT) computations on small-molecule computational models of the [FeFe]H2ase active site to address this problem. Specifically, a series of structural candidates are geometry optimized and their infrared (IR) spectra are simulated using the computed C-O and C-N stretching frequencies and infrared intensities. Structural assignments are made by comparing these spectra to the experimentally determined IR spectra for each form. The H red form is assigned as a mixture of an Fe(I)Fe(I) form with an open site on the distal iron center and either a Fe(I)Fe(I) form in which the distal cyanide has been protonated or a Fe(II)Fe(II) form with a bridging hydride ligand. The Hox form is assigned as a valence-localized Fe(I)Fe(II) redox level with an open site at the distal iron. The Hox(air)(ox) form is assigned as an Fe(II)Fe(II) redox level with OH(-) or OOH(-) bound to the distal iron center that may or may not have an oxygen atom bound to one of the sulfur atoms of the dithiolate linker. Comparisons of the computed IR spectra of the (12)CO and (13)CO inhibited form with the experimental IR spectra show that exogenous CO binds terminally to the distal iron center.