Electronic Structure of the H Cluster in [Fe]-Hydrogenases

Abstract

[Fe]-Hydrogenase II isolated from C. pasteurianum contains 14 Fe which are distributed among the so-called H cluster (the catalytic center) and two [4Fe-4S] clusters. Insights gained from Mössbauer studies of M-[4Fe-4S]2+ cluster assemblies (M is a paramagnetic center) in sulfite reductase and carbon monoxide dehydrogenase have suggested that the H cluster contains a [4Fe-4S]2+ cluster covalently linked to a smaller Fe-containing cluster. Recent X-ray studies of two [Fe]-hydrogenases, combined with the results of FTIR studies, have revealed that the H cluster contains a novel binuclear Fe cluster, [2Fe]H, that is linked by a cysteinyl sulfur to a [4Fe-4S] cluster; [2Fe]H was found to have CO, CN-, and thiolate ligands. The analysis of the Mössbauer spectra of Hydrogenase II in the oxidized, reduced, and the CO-inhibited states has enabled us to assign the 57Fe magnetic hyperfine tensors observed by ENDOR and Mössbauer spectroscopy to the two subclusters. Thus, AI = +25.3 MHz and AII = −28.4 MHz of Hox-CO can be assigned to the two delocalized pairs of [4Fe − 4S . In our coupling model these A-values result for j ≈ 100 cm-1 where j describes the exchange interaction between [4Fe − 4S and [2Fe]H. The 18 MHz A-value of Hox obtained by ENDOR must result from one Fe site of [2Fe]H, while the 7.5 MHz ENDOR A-value seems to be associated with [4Fe-4S]H. Analysis of the Mössbauer spectra of Hred shows that the 4Fe cluster is in the 2+ state and that [2Fe]H contains presumably two low-spin FeII sites with ΔEQ ≈ 0.85 mm/s and δ ≈ 0.08 mm/s. The observation that the [4Fe-4S] cluster is in the 2+ state in Hox, Hox-CO, and Hred suggests that the [2Fe]H subcluster is in the mixed-valent FeIIIFeII state in Hox and Hox-CO. Given the environment of strong-field ligands in [2Fe]H, the FeIII site must have low-spin configuration. While such an assignment is compatible with the EPR g-values, low-spin FeIII sites with g ≈ 2 commonly exhibit very anisotropic 57Fe A-tensors (due to spin-dipolar interactions) and thus the isotropic A-values of Hox and Hox-CO observed by ENDOR are difficult to explain. This point is discussed in some detail.