Michel Frey

Hydrogenases: Hydrogen-activating enzymes

Many microorganisms, such as methanogenic, acetogenic, nitrogen-fixing, photosynthetic, or sulfate-reducing bacteria, metabolize hydrogen. 2a] Hydrogen activation is mediated by a family of enzymes, termed hydrogenases, which either provide these organisms with reducing power from hydrogen oxidation or act as TMelectron sinks∫, following the reaction: H2 2H 2e . Not surprisingly, hydrogenases are mostly studied with a view to designing chemical or biochemical processes to produce molecular hydrogen more abundantly and cheaply than with platinum catalysts ; molecular hydrogen is an ideally clean fuel. 5] Hydrogenases (cytochrome c3 oxidoreductase, EC 1.18.99.1) are classified into two major families in the present paper on the basis of the metal content of their respective dinuclear catalytic centers, that is nickel ± iron (NiFe) hydrogenases and TMiron only∫ (FeFe) hydrogenases. Some NiFe hydrogenases also contain selenium at their catalytic center in the form of selenocysteine (Table 1). The two hydrogenases families differ functionally from each other in that NiFe hydrogenases tend to be more involved in hydrogen oxidation and FeFe hydrogenases in hydrogen production. Moreover, NiFe hydrogenases are approximately 10 1 ± 10 2 times less active, show 10 times more affinity for hydrogen, and are less sensitive to inhibition by oxygen and carbon monoxide than FeFe hydrogenases (Table 2). A TMmetalfree∫ hydrogenase, found in methanogenic bacteria, catalyzes the reversible reduction of a methenyltetrahydromethanopterin (methenyl-H4MPT) methanogenic cofactor with H2 to form methylene-H4MPTand a proton during methane formation from CO2 and 4H2. The three-dimensional atomic models of four NiFe, one NiFeSe, and, more recently, two FeFe 14] hydrogenases (Table 3) have been elucidated by X-ray crystallography on the basis of gene sequencing and a wealth of biochemical and spectroscopic studies, in some cases coupled with isotopic labeling. 2, 16] These results represent a considerable impetus to research the catalytic mechanism of hydrogenases and the design of organometallic compounds which mimic their structural or functional properties, or both. The aim of this short review is to highlight some recent works and trends in hydrogenase research.

The crystal structure of a reduced [NiFeSe] hydrogenase provides an image of the activated catalytic center

BACKGROUNDn[NiFeSe] hydrogenases are metalloenzymes that catalyze the reaction H2<-->2H+ + 2e-. They are generally heterodimeric, contain three iron-sulfur clusters in their small subunit and a nickel-iron-containing active site in their large subunit that includes a selenocysteine (SeCys) ligand.nnnRESULTSnWe report here the X-ray structure at 2.15 A resolution of the periplasmic [NiFeSe] hydrogenase from Desulfomicrobium baculatum in its reduced, active form. A comparison of active sites of the oxidized, as-prepared, Desulfovibrio gigas and the reduced D. baculatum hydrogenases shows that in the reduced enzyme the nickel-iron distance is 0.4 A shorter than in the oxidized enzyme. In addition, the putative oxo ligand, detected in the as-prepared D. gigas enzyme, is absent from the D. baculatum hydrogenase. We also observe higher-than-average temperature factors for both the active site nickel-selenocysteine ligand and the neighboring Glu18 residue, suggesting that both these moieties are involved in proton transfer between the active site and the molecular surface. Other differences between [NiFeSe] and [NiFe] hydrogenases are the presence of a third [4Fe4S] cluster replacing the [3Fe4S] cluster found in the D. gigas enzyme, and a putative iron center that substitutes the magnesium ion that has already been described at the C terminus of the large subunit of two [NiFe] hydrogenases.nnnCONCLUSIONSnThe heterolytic cleavage of molecular hydrogen seems to be mediated by the nickel center and the selenocysteine residue. Beside modifying the catalytic properties of the enzyme, the selenium ligand might protect the nickel atom from oxidation. We conclude that the putative oxo ligand is a signature of inactive unready [NiFe] hydrogenases.

Gas access to the active site of Ni-Fe hydrogenases probed by X-ray crystallography and molecular dynamics

The 2.54 Å resolution structure of Ni-Fe hydrogenase has revealed the existence of hydrophobic channels connecting the molecular surface to the active site. A crystallographic analysis of xenon binding together with molecular dynamics simulations of xenon and H2 diffusion in the enzyme interior suggest that these channels serve as pathways for gas access to the active site.

A redox-dependent interaction between two electron-transfer partners involved in photosynthesis

Ferredoxin:NADP+:reductase (FNR) catalyzes one terminal step of the conversion of light energy into chemical energy during photosynthesis. FNR uses two high energy electrons photoproduced by photosystem I (PSI) and conveyed, one by one, by a ferredoxin (Fd), to reduce NADP+ to NADPH. The reducing power of NADPH is finally involved in carbon assimilation. The interaction between oxidized FNR and Fd was studied by crystallography at 2.4 Å resolution leading to a three‐dimensional picture of an Fd–FNR biologically relevant complex. This complex suggests that FNR and Fd specifically interact prior to each electron transfer and disassemble upon a redox‐linked conformational change of the Fd.

Rubredoxin from Desulfovibrio gigas. A molecular model of the oxidized form at 1.4 A resolution.

The crystal structure of rubredoxin from the sulfate-reducing bacterium Desulfovibrio gigas has been determined at 1.4 A resolution (1 A = 0.1 nm) by X-ray diffraction methods; starting with a model of the isostructural rubredoxin from Desulfovibrio vulgaris. Refinement of the molecular model has been carried out by restrained least-squares techniques and Fourier series calculations. The present model includes a formyl at the N-terminal end and 121 possible sites for solvent molecules with full or partial occupancy, which corresponds to the modeling of nearly all the solvent medium. The crystallographic R factor against the data with 10 A greater than d greater than 1.4 A with F greater than 2 sig(F), is 0.136; and R = 0.140 when all the data are considered. The estimated average root-mean-square (r.m.s.) error on the positional parameters is about 0.12 A. The overall structural features of this molecule are close to those of the two highly refined rubredoxins from Clostridium pasteurianum and D. vulgaris. Superposition of these two molecules on the rubredoxin from D. gigas shows in both cases an overall r.m.s. deviation of 0.5 A for the atoms in the main-chain and of 0.4 A for the atoms in the side-chains that make up the hydrophobic core. The iron atom is co-ordinated to four cysteine sulfur atoms forming an almost regular tetrahedron, with Fe-SG distances ranging from 2.27 A to 2.31 A and angles varying from 103 degrees to 115 degrees. The intramolecular hydrogen-bonding pattern is quite comparable to those found in other proteins refined at high resolution. All the polar groups are involved in hydrogen bonds: intramolecular, intermolecular or with solvent molecules. The main structural differences from the other rubredoxins are in the nature and the distribution of some of the charged residues over the molecular surface. The possible influence of several structural factors on the intramolecular and intermolecular electron transfer properties such as the NH...SG bonds, the solvent exposure of the redox center, and the aromatic core is discussed. The conservation, during evolution, of a ring of acidic residues in the proximity of the FeSG4 center suggests that this ring may be implicated in the recognition processes between rubredoxins and their functional partners.

X-ray crystal structure determination and refinement at 1.9 A resolution of isolectin I from the seeds of Lathyrus ochrus.

Orthorhombic crystals of isolectin I (LOLI) from the seeds of Lathyrus ochrus were first obtained during the STS 29 space shuttle mission. Subsequently, isostructural crystals were also obtained in the laboratory. They belong to the space group P2(1)2(1)2, with cell dimensions a = 135.84 A, b = 63.12 A and c = 54.54 A with one functional entity, a dimer, in the asymmetric unit (Vm = 2.2 A3/Da). The three-dimensional structure of LOLI, which was solved by the molecular replacement method using a 3 A resolution model of pea lectin, has subsequently been refined by using crystallographic data between 8.0 A and 1.9 A resolution, coupled to molecular dynamics and energy minimization techniques. The conventional R-factor is 0.185 for Fo greater than 1 sigma(Fo). The final model includes 220 well-defined water molecules and has root-mean-square deviations from ideal bond distances and angles of 0.004 A and 3 degrees, respectively. Only slight conformation differences have been found between the two alpha beta monomers. The molecular structure of LOLI, the first to be determined from the genus Lathyrus, is very similar to those of concanavalin A, pea lectin and favin. Differences are confined to the loop regions and beta-chain termini. Comparison of equivalent C alpha atom positions between our final model and the pea lectin structure shows slight differences in the association of the two monomers, which are probably due to the different environments in the crystals. The root-mean-square deviation between C alpha atoms of LOLI and pea lectin is 0.40 A. The metal binding sites are very similar in pea lectin, concanavalin A and LOLI. The sugar-binding sites of LOLI are occupied by four well-ordered water molecules each. The cleavage site for one of the monomers is specially well defined in the final electron density map: the amino group of Glul (alpha) seems to form a salt bridge with the carboxylate group of the terminal Asn181 (beta). A detailed analysis of the difference in crystal packing contacts between pea lectin and LOLI shows that, as might be expected, several of the intermolecular interactions are mediated by residues that correspond to substitutions in the LOLI amino acid sequence.

Hydrogenase: A hydrogen-metabolizing enzyme. What do the crystal structures tell us about its mode of action?

Hydrogenases are proteins which metabolize the most simple of chemical compounds, molecular hydrogen, according to the reaction H2<-->2H+ + 2e-. These enzymes are found in many microorganisms of great biotechnological interest such as methanogenic, acetogenic, nitrogen fixing, photosynthetic or sulfate-reducing bacteria. The X-ray structure of a dimeric [NiFe] hydrogenase together with a wealth of biophysical, biochemical and genetic studies have revealed that the large subunit contains the bimetallic [Ni-Fe] active site, with biologically uncommon CO and CN ligands to the iron, whereas the small subunit contains three iron-sulfur cluster. During catalysis, the nickel atom is most likely responsible for a base-assisted heterolytic cleavage of the hydrogen molecule whereas the iron atom could be redox active. Specific channels are probably required for the transfer of the chemical reaction partners (H2, H+ and e-) between the active site, deeply buried inside the protein, and the molecular surface. The generation of a functional enzyme, including the assembly of the complex catalytic center, requires maturation and involves a large number of auxiliary proteins which have been partly characterized by molecular biology.

Crystallographic studies of the interaction between the ferredoxin-NADP+ reductase and ferredoxin from the cyanobacterium Anabaena: looking for the elusive ferredoxin molecule.

Ferredoxin-NADP(+) reductase (FNR) and its physiological electron donor ferredoxin (Fd) from the cyanobacterium Anabaena PCC7119 have been co-crystallized. The unit-cell parameters are a = b = 63.72, c = 158.02 A and the space group is P2(1)2(1)2(1). The crystal structure has been solved with 2.4 A resolution synchrotron data by molecular replacement, anomalous dispersion and R(min) search methods. For the computations, the crystal was treated as a merohedral twin. The asymmetric unit contains two FNR molecules and one ferredoxin molecule. The packing of the FNR molecules displays a nearly tetragonal symmetry (space group P4(3)2(1)2), whereas the ferredoxin arrangement is orthorhombic. This study provides the first crystallographic model of a dissociable complex between FNR and Fd.

Novel metal sites in protein structures

Recently, the three-dimensional structures of several novel metalloenzymes have been solved. Of special interest are those containing uncommon and/or not well characterized metals such as molybdenum, tungsten, nickel, vanadium and cobalt. Modulated by the protein environment, the specific properties of these metals and of special metal-binding cofactors such as siroheme and topa quinone are used to catalyze a vast array of fascinating reactions.

Crystallization, preliminary X-ray study and crystal activity of the hydrogenase from Desulfovibrio gigas

Hydrogenase (EC 1.12) from Desulfovibrio gigas is a dimeric enzyme (26 and 62 (X 10(3) Mr) that catalyzes the reversible oxidation of molecular hydrogen. Single crystals of hydrogenase have been produced using the hanging drop method, with either PEG (polyethylene glycol) 6000 or ammonium sulfate as precipitants at pH 6.5. X-ray examination of the crystals indicates that those obtained with ammonium sulfate are suitable for structure determination to at least 3.0 A resolution when synchrotron radiation Sources are used (1 A = 0.1 nm). The crystals are monoclinic, with space group C2, and cell dimensions a = 257.0 A, b = 184.7 A, c = 148.3 A and beta = 101.3 degrees, and contain between four and ten molecules per asymmetric unit. The enzyme can be reactivated within the crystals under reducing conditions without crystal damage.