The main objective of this project was to design biomimetic compounds with hydrogenase (H2ase) activity. H2ases are capable of both H2 production and oxidation and thorough understanding of these phenomena would have a significat impact in strategies aimed at using H2 as a future, clean, fuel. The stated immediate goals included 1) the elucidation of the crystal structure of at least one Ni-containing hydrogenase in different redox states; 2) the correlation between the redox-dependent structural changes and the Ni Electron Paramagnetic Resonance spectra and 3) the use of this information to understand the catalytic mechanism of Ni-hydrogenases and to design effective biomimetic catalysts.
1) Crystallographic analyses of as-prepared hydrogenase from the sulfate-reducing bacterium Desulfovibrio gigas carried out at 2.85 Å and 2.54 Å resolution have yielded detailed images of the active site. Unexpectedly, the site contains, besides Ni, a second metal center. Furthermore, this second metal ion was found to bind three non-exchangeable diatomic ligands of unknown nature. 2) Our combined effort involving metal content analysis, novel EPR and ENDOR studies and the use of X-ray anomalous scattering effects, has established that the hydrogenase active site contains a Ni-Fe center, and that the Fe ion is probably low-spin Fe(II) in all the EPR active redox states. 3) However, a likely redox role for the active site Fe center is deduced from the fact that a set of three very high frequency infrared bands, that we attribute to the diatomic ligands, shifts as a function of active site redox state (including those which are EPR-silent); thus reporting on electron density changes taking place at the Fe center. 4) A plausible catalytic mechanism has emerged from these studies and from an analysis of previous model chemistry concerning Ni-containing hydrogenases. In this hypothetical mechanism, Ni is the catalytic site, accomplishing the cleavage of the hydrogen molecule; whereas Fe appears to play a central role in the redox processes. 4) In an effort to improve our understanding of the structural aspects of hydrogen catalysis, we have, on the one hand, developed state-of-the-art reduction and anaerobic cryocrystallographic techniques in order to characterize the reduced enzyme (study under way) and its complexes with CO and C2H2, and on the other, we have cloned and expressed wild type and mutated D. fructosovorans hydrogenase (and solved its crystal structure) to probe electron and proton transfer pathways. In addition, mutations at, or near to, the active site have been deviced. This enzyme has also been used to probe gas accessibility to the active site by exposing the crystals to a high pressure Xe atmosphere 5) The fact that the hydrogenase active site is heterobinuclear has prompted us and others to synthetize model compounds containing Fe-Ni centers. So far, we have concentrated our efforts on generating Ni compounds with S-rich coordination because in the enzyme the Ni ion is coordinated by four cysteine residues. A second step would consist on adding an Fe ion to the model compounds. The fact that the active site appears much more complex that originally envisioned, implies that it will be difficult to obtain compounds which are truly biomimetic. However, many of the structural features of the active site can be modelled and they should further significantly our understanding of hydrogen biocatalysis.
MAJOR SCIENTIFIC BREAKTHROUGHS:
1) The structure of the first Ni-containing enzyme has been determined; 2) A previously considered mononuclear Ni center has been shown to be binuclear and the identity of the second metal center (Fe) has been determined beyond any doubt; 3) Three diatomic ligands to the active site Fe center have been characterized by IR. This is a completely unprecedented finding, as this is the first time that such endogenous ligands are described in a protein.
Funding SchemeCSC - Cost-sharing contracts
SE1 9NN London
2300 RA Leiden