The iron-sulfur cluster-free hydrogenase (Hmd) from methanogenic archaea harbors an iron-containing cofactor of yet unknown structure. X-ray absorption spectroscopy of the active, as isolated enzyme from Methanothermobacter marburgensis (mHmd) and of the active, reconstituted enzyme from Methanocaldococcus jannaschii (jHmd) revealed the presence of mononuclear iron with two CO, one sulfur and one or two N/O in coordination distance. In jHmd, the single sulfur ligand is most probably provided by Cys 176 , as deduced from a comparison of the activity and of the x-ray absorption and Mössbauer spectra of the enzyme mutated in any of the three conserved cysteines. In the isolated Hmd cofactor, two CO, one sulfur, and two nitrogen/oxygen atoms coordinate the iron, the sulfur ligand being most probably provided by mercaptoethanol, which is absolutely required for the extraction of the iron-containing cofactor from the holoenzyme and for the stabilization of the extracted cofactor. In active mHmd holoenzyme, the number of iron ligands increased by one when one of the Hmd inhibitors (CO or KCN) were present, indicating that in active Hmd, the iron contains an open coordination site, which is proposed to be the site of H 2 interaction.Hydrogenases are enzymes that catalyze the reversible oxidation of molecular hydrogen (1). Their structure and catalytic mechanism are of considerable applied interest as models for the development of efficient catalysts for hydrogen-fueled processes. Despite intensive efforts, however, the understanding of how hydrogenases react with H 2 is only in its infancy.Two of the three known types of hydrogenases are ironsulfur proteins that contain, besides one or several iron-sulfur clusters, a dinuclear metal center, either [NiFe] or [FeFe], which was shown to be the site of H 2 reaction. Both dinuclear hydrogenases catalyze the reversible formation of 2e Ϫ and 2Hϩ from H 2 , the electrons being transferred one by one to electron acceptors via the iron-sulfur clusters. One iron ion in both dinuclear centers is coordinated by sulfur, CO, and cyanide ligands, as supported by crystal structures of the [NiFe]-hydrogenase (2, 3) and of the [FeFe]-hydrogenase (4 -7). During the catalytic cycle, the iron ion in the [NiFe] center remains in a low spin ferrous state and is not redoxactive (8, 9). In the case of the [FeFe]-hydrogenases, the oxidation states of the dinuclear iron center are still under discussion; however, the involvement of a low spin Fe(II) has been predicted based on the spectroscopic and DFT studies (9 -11). Despite similarities in structures and properties, [FeFe]-and [NiFe]-hydrogenases are phylogenetically not related.[FeFe]-hydrogenases are found in bacteria and eucarya, whereas [NiFe]-hydrogenases are found in bacteria and archaea. For recent reviews on these two types of hydrogenases, see Frey (9), Zhou et al.