Fe-only hydrogenases contain a di-iron active site complex, in which the two Fe atoms have carbon monoxide and cyanide ligands and are linked together by a putative di(thiomethyl)amine molecule. We have cloned, purified and characterized the HydE and HydG proteins, thought to be involved in the biosynthesis of this peculiar metal site, from the thermophilic organism Thermotoga maritima. The HydE protein anaerobically reconstituted with iron and sulfide binds two [4Fe-4S] clusters, as characterized by UV and EPR spectroscopy. The HydG protein binds one [4Fe-4S] cluster, and probably an additional one. Both enzymes are able to reductively cleave S-adenosylmethionine (SAM) when reduced by dithionite, confirming that they are Radical-SAM enzymes.

Compound I of Proteus mirabilis and bovine liver catalases (PMC and BLC, respectively) were studied combining EPR spectroscopy and the rapid-mix freeze-quench techniques. Both enzymes, when treated with peroxyacetic acid, form a catalytic intermediate which consists of an oxoferryl porphyrin pi-cation radical. In PMC this intermediate is semistable, and an unexpected reversible equilibrium under pH influence takes place between two forms of compound I with different coupling between the oxoferryl and the porphyrin pi-cation radical. At acid pH, one form has a ferromagnetic character as in Micrococcus luteus compound I. At neutral pH, another form with a much smaller coupling, reminiscent of the horse radish peroxidase compound I, is detected. The approximate midpoint, estimated for these changes in the range 5.3 < pH < 6.0, approaches the pKa value of an histidyl residue. The residues possibly involved in the transformation are discussed in terms of the known structure of PMC compound I. The EPR spectrum of BLC compound I (pH 5.6), obtained in the millisecond time scale (40 ms), also showed a mixture of two forms which, most probably, correspond to two different magnetic exchange interactions, as in the case of PMC. Taken together, the low-temperature electronic absorption and the EPR spectra of BLC compound I formed in the 0.04-15 s range show that the porphyrin pi-cation radical disappears and, instead, a tyrosyl radical is formed. ENDOR experiments confirm our previously estimated hyperfine couplings to the C2,6 and C3,5 ring protons and the beta-methylene protons of the purported tyrosyl radical. Candidates for such a tyrosyl radical are discussed in connection with the possible electron transfer pathways between the heme active site and the NADPH cofactor.