Orientation-selected ENDOR of the active center in Chromatium vinosum [NiFe] hydrogenase in the oxidized "ready" state

 Electron nuclear double resonance (ENDOR) was applied to study the active site of the oxidized "ready" state, Ni_r, in the [NiFe] hydrogenase of Chromatium vinosum. The magnetic field dependence of the EPR was used to select specific subsets of molecules contributing to the ENDOR response by stepping through the EPR envelope. Three hyperfine couplings could be clearly followed over the complete field range. Two protons, H1 and H2, display a very similar large isotropic coupling of 12.5 and 12.6 MHz, respectively. Their dipolar coupling is small (2.1 and 1.4 MHz, respectively). A third proton, H3, exhibits a small isotropic coupling of 0.5 MHz and a larger anisotropic contribution of 3.5 MHz. Based on a comparison with structural data obtained from X-ray crystallography of single crystals of hydrogenases from Desulfovibrio gigas and D. vulgaris and the known g-tensor orientation of Ni_r, an assignment of the ¹H hyperfine couplings could be achieved. H1 and H2 were assigned to the β-CH₂ protons of the bridging cysteine Cys533 and H3 could belong to a β-CH₂ proton of Cys68 or to a protonated cysteine (-SH) of Cys68 or Cys530. Received: 26 November 1998 / Accepted: 1 April 1999     

Duplication of hyp genes involved in maturation of [NiFe] hydrogenases in Alcaligenes eutrophus H16

  Alcaligenes eutrophus H16 harbors seven hyp genes (hypA, B, F, C, D, E, and X) as part of the hydrogenase gene cluster on megaplasmid pHG1. Here we demonstrate that three of the hyp genes (hypA, B, and F) are duplicated in A. eutrophus, which explains the lack of a phenotypic change in single-site mutants impaired in one of the two copies. Mutants with lesions in both copies showed clear alterations in hydrogenase activities. Deletions in hypF1 and hypF2 completely abolished activities of the soluble hydrogenase and of the membrane-bound hydrogenase, mutations in hypA1 and hypA2 totally blocked the membrane-bound hydrogenase activity, while residual soluble hydrogenase activity accounted for the extremely slow growth of the strain on H₂. Both hydrogenase activities of mutants defective in hypB1 and hypB2 were partially restored by elevating the concentration of nickel chloride in the medium. Reduction of hydrogenase activities in the double mutants correlated with varying degrees of maturation deficiency based upon the amount of unprocessed nickel-free hydrogenase precursor. Despite a high identity between the two copies of hyp gene products, substantial structural differences were identified between the two copies of hypF genes. HypF1, although functionally active, is a truncated version of HypF2, whose structure resembles HypF proteins of other organisms. Interestingly, the N-terminus of HypF2, which is missing in the HypF1 counterpart, contains a putative acylphosphatase domain in addition to a potential metal binding site. Received: 15 June 1998 / Accepted: 5 August 1998     

Energy-Converting [NiFe] Hydrogenases from Archaea and Extremophiles: Ancestors of Complex I

[NiFe] hydrogenases are well-characterized enzymes that have a key function in the H2 metabolism of various microorganisms. In the recent years a subfamily of [NiFe] hydrogenases with unique properties has been identified. The members of this family form multisubunit membrane-bound enzyme complexes composed of at least four hydrophilic and two integral membrane proteins. These six conserved subunits, which built the core of these hydrogenases, have closely related counterparts in energy-conserving NADH:quinone oxidoreductases (complex I). However, the reaction catalyzed by these hydrogenases differs significantly from the reaction catalyzed by complex I. For some of these hydrogenases the physiological role is to catalyze the reduction of H+ with electrons derived from reduced ferredoxins or poly-ferredoxins. This exergonic reaction is coupled to energy conservation by means of electron-transport phosphorylation. Other members of this hydrogenase family mainly function to provide the cell with reduced ferredoxin with H2 as electron donor in a reaction driven by reverse electron transport. As complex I these hydrogenases function as ion pumps and have therefore been designated as energy-converting [NiFe] hydrogenases.     

Characterization of the active site of catalytically inactive forms of [NiFe] hydrogenases by density functional theory

The inactive forms, unready (Ni-A, Ni-SU) and ready (Ni-B), of NiFe hydrogenases are modeled by examining the possibility of hydroxo, oxo, hydroperoxo, peroxo, and sulfenate groups in active-site models and comparing predicted IR frequencies and g tensors with those of the enzyme. The best models for Ni-A and Ni-SU have hydroxo (μ-OH) bridges between Fe and Ni and a terminal sulfenate [Ni–S(=O)Cys] group, although a hydroperoxo model for Ni-A is also quite viable, whereas the best model for Ni-B has only a μ-OH bridge. In addition, a mechanism for the activation of unready hydrogenase is proposed on the basis of the relative stabilities of sulfenate models versus peroxide models.     

Sequential and structural analysis of [NiFe]-hydrogenase-maturation proteins from Desulfovibrio vulgaris Miyazaki F

The complete primary structure of the hyn-region in the genome of Desulfovibrio vulgaris Miyazaki F (DvMF), encoding the [NiFe]-hydrogenase and two maturation proteins has been identified. Besides the formerly reported genes for the large and small subunits, this region comprises genes encoding an endopeptidase (HynC) and a putative chaperone (HynD). The complete genomic region covers 4086 nucleotides including the previously published upstream located promoter region and the sequences of the structural genes. A phylogenetic tree for both maturation proteins shows strongest sequential relationship to the orthologous proteins of Desulfovibrio vulgaris Hildenborough (DvH). Secondary structure prediction for HynC (168 aa, corresponding to a molecular weight of 17.9 kDa) revealed a practically identical arrangement of α-helical and β-strand elements between the orthologous protein HybD from E. coli and allowed a three-dimensional modelling of HynC on the basis of the formerly published structure of HybD. The putative chaperone HynD consists of 83 aa (molecular weight of 9 kDa) and shows 76% homology to DvH HynD. Preliminary experiments demonstrate that the operon is expressed under the control of its own promoter in Escherichia coli, although no further processing could be observed, providing evidence that additional proteins have to be involved in the maturation process. Accession numbers: DQ072852, HynC protein ID AAY90127, HynD protein ID AAY90128.     

Properties of the [NiFe]-hydrogenase maturation protein HypD

A mutational screen of amino acid residues of hydrogenase maturation protein HypD from Escherichia coli disclosed that seven conserved cysteine residues located in three different motifs in HypD are essential. Evidence is presented for potential functions of these motifs in the maturation process.     

Crystal structures of hydrogenase maturation protein HypE in the Apo and ATP-bound forms

The hydrogenase maturation protein HypE serves an essential function in the biosynthesis of the nitrile group, which is subsequently coordinated to Fe as CN(-) ligands in [Ni-Fe] hydrogenase. Here, we present the crystal structures of HypE from Desulfovibrio vulgaris Hildenborough in the presence and in the absence of ATP at a resolution of 2.0 A and 2.6 A, respectively. Comparison of the apo structure with the ATP-bound structure reveals that binding ATP causes an induced-fit movement of the N-terminal portion, but does not entail an overall structural change. The residue Cys341 at the C terminus, whose thiol group is supposed to be carbamoylated before the nitrile group synthesis, is completely buried within the protein and is located in the vicinity of the gamma-phosphate group of the bound ATP. This suggests that the catalytic reaction occurs in this configuration but that a conformational change is required for the carbamoylation of Cys341. A glutamate residue is found close to the thiol group as well, which is suggestive of deprotonation of the carbamoyl group at the beginning of the reactions.     

The role of the active site-coordinating cysteine residues in the maturation of the H2-sensing [NiFe] hydrogenase from Ralstonia eutropha H16

The H₂-splitting active site of [NiFe] hydrogenases is tightly bound to the protein matrix via four conserved cysteine residues. In this study, the nickel-binding cysteine residues of HoxC, the large subunit of the H₂-sensing regulatory hydrogenase (RH) from Ralstonia eutropha, were replaced by serine. All four mutant proteins, C60S, C63S, C479S, and C482S, were inactive both in H₂ sensing and H₂ oxidation and did not adopt the native oligomeric structure of the RH. Nickel was bound only to the C482S derivative. The assembly of the [NiFe] active site is a complex process that requires the function of at least six accessory proteins. Among these proteins, HypC has been shown to act as a chaperone for the large subunit during the maturation process. Immunoblot analysis revealed the presence of a strong RH-dependent HypC-specific complex in extracts containing the C60S, C63S, and C482S derivatives, pointing to a block in maturation for these mutant proteins. The lack of this complex in the extract containing C479S indicates that this specific cysteine residue might be crucial for the interaction between HoxC and HypC.This work is dedicated to Prof. H.G. Schlegel on the occasion of his 80th birthday.     

A theoretical study of spin states in Ni-S<Subscript>4</Subscript> complexes and models of the [NiFe] hydrogenase active site

We have applied density functional theory, using both pure (BP86) and hybrid (B3LYP and B3LYP*) functionals, to investigate structural parameters and reaction energies for nickel(II)-sulfur coordination compounds, as well as for small cluster models of the Ni-SI and Ni-R redox state of [NiFe] hydrogenases. Results obtained investigating experimentally well-characterized complexes show that BP86 is well suited to describe the structural features of this class of compounds. However, the singlet-triplet energy splitting and even the computed ground state are strongly dependent on the applied functional. Results for the cluster models of [NiFe] hydrogenases lead to the conclusion that in the reduced protein structures characterized by X-ray diffraction a hydride bridges the two metal centres. The energy splitting of the singlet and triplet states in Ni-R and Ni-SI models is calculated to be very small and may be overcome at room temperature to allow a spin crossover. Moreover, the relative stability of the Ni-SI and Ni-R structures adopted in the present investigation is fully compatible with their involvement in the reversible heterolytic cleavage of H(2).     

Activation process of [NiFe] hydrogenase elucidated by high-resolution X-ray analyses: conversion of the ready to the unready state

Hydrogenases catalyze oxidoreduction of molecular hydrogen and have potential applications for utilizing dihydrogen as an energy source. [NiFe] hydrogenase has two different oxidized states, Ni-A (unready, exhibits a lag phase in reductive activation) and Ni-B (ready). We have succeeded in converting Ni-B to Ni-A with the use of Na2S and O2 and determining the high-resolution crystal structures of both states. Ni-B possesses a monatomic nonprotein bridging ligand at the Ni-Fe active site, whereas Ni-A has a diatomic species. The terminal atom of the bridging species of Ni-A occupies a similar position as C of the exogenous CO in the CO complex (inhibited state). The common features of the enzyme structures at the unready (Ni-A) and inhibited (CO complex) states are proposed. These findings provide useful information on the design of new systems of biomimetic dihydrogen production and fuel cell devices.     

Inhibition of the [NiFe] hydrogenase from Desulfovibrio vulgaris Miyazaki F by carbon monoxide: An FTIR and EPR spectroscopic study

X-ray crystallographic studies [Ogata et al., J. Am. Chem. Soc. 124 (2002) 11628-11635] have shown that carbon monoxide binds to the nickel ion at the active site of the [NiFe] hydrogenase from Desulfovibriovulgaris Miyazaki F and inhibits its catalytic function. In the present work spectroscopic aspects of the CO inhibition for this bacterial organism are reported for the first time and enable a direct comparison with the existing crystallographic data. The binding affinity of each specific redox state for CO is probed by FTIR spectro-electrochemistry. It is shown that only the physiological state Ni-SI_a reacts with CO. The CO-inhibited product state is EPR-silent (Ni²⁺) and exists in two forms, Ni-SCO and Ni-SCO_red. At very negative potentials, the exogenous CO is electrochemically detached from the active site and the active Ni-R states are obtained. At temperatures below 100 K, photodissociation of the extrinsic CO from the Ni-SCO state results in Ni-SI_a that is identified to be the only light-induced state. In the dark, rebinding of CO takes place; the recombination rate constants are of biexponential character and the activation barrier is determined to be approximately 9 kJ mol⁻¹. In addition, formation of a paramagnetic CO-inhibited state (Ni-CO) was observed that results from the interaction of carbon monoxide with the Ni-L state. It is proposed that the nickel in Ni-CO is in a formal monovalent state (Ni¹⁺).     

Structural features of [NiFeSe] and [NiFe] hydrogenases determining their different properties: a computational approach

Hydrogenases are metalloenzymes that catalyze the reversible reaction \textH₂ \leftrightarrows 2\textH ⁺ + 2\texte ^- {\text{H}}_{2} \leftrightarrows 2{\text{H}}^{ + } + 2{\text{e}}^{ - } , being potentially useful in H₂ production or oxidation. [NiFeSe] hydrogenases are a particularly interesting subgroup of the [NiFe] class that exhibit tolerance to O₂ inhibition and produce more H₂ than standard [NiFe] hydrogenases. However, the molecular determinants responsible for these properties remain unknown. Hydrophobic pathways for H₂ diffusion have been identified in [NiFe] hydrogenases, as have proton transfer pathways, but they have never been studied in [NiFeSe] hydrogenases. Our aim was, for the first time, to characterize the H₂ and proton pathways in a [NiFeSe] hydrogenase and compare them with those in a standard [NiFe] hydrogenase. We performed molecular dynamics simulations of H₂ diffusion in the [NiFeSe] hydrogenase from Desulfomicrobium baculatum and extended previous simulations of the [NiFe] hydrogenase from Desulfovibrio gigas (Teixeira et al. in Biophys J 91:2035–2045, 2006). The comparison showed that H₂ density near the active site is much higher in [NiFeSe] hydrogenase, which appears to have an alternative route for the access of H₂ to the active site. We have also determined a possible proton transfer pathway in the [NiFeSe] hydrogenase from D. baculatum using continuum electrostatics and Monte Carlo simulation and compared it with the proton pathway we found in the [NiFe] hydrogenase from D. gigas (Teixeira et al. in Proteins 70:1010–1022, 2008). The residues constituting both proton transfer pathways are considerably different, although in the same region of the protein. These results support the hypothesis that some of the special properties of [NiFeSe] hydrogenases could be related to differences in the H₂ and proton pathways.     

Maturation of [NiFe]-hydrogenases in Escherichia coli: the HypC cycle

Carbamoyl phosphate (CP) has been implicated as an educt for the synthesis of the CO and CN ligands of the metal centre of [NiFe]-hydrogenases in Escherichia coli, since CP synthetase mutants (carAB) are unable to generate active hydrogenases due to a block in enzyme maturation. Citrulline, when added to the growth medium in high concentrations, compensated for the phenotype of the mutants. It is now shown that overexpression of the argI gene lowered the effective concentration of citrulline, thus proving that the amino acid serves as a source for CP. The DeltaCarAB mutant accumulated a complex consisting of the hydrogenase maturation proteins HypC and HypD. This complex was resolved upon citrulline addition and followed-up by the appearance of a complex between HypC and the precursor of the large subunit of hydrogenase 3, preHycE. In the absence of the hycE gene, the HypC-HypD complex did not disappear upon addition of citrulline but developed into a form migrating slower in a non-denaturing polyacrylamide gel, providing strong evidence for the notion that the HypC-HypD complex is the intermediate in hydrogenase maturation where CP or its products are added to the iron atom of the metal centre. This step precedes nickel insertion, since extracts of carAB cells that had been cultivated in the absence of citrulline are unable to process preHycE after the addition of nickel. Complex formation between HypC and HypD, and between HypC and preHycE display dependence on identical primary structure elements of HypC. On the basis of the results, a cycle of HypC activity is proposed whose function is to transfer the iron atom that has been liganded at the HypC-HypD complex to the precursor of the large hydrogenase subunit.     

The activation of the [NiFe]-hydrogenase from<Emphasis Type="Italic"> Allochromatium vinosum</Emphasis>. An infrared spectro-electrochemical study

The membrane-bound [NiFe]-hydrogenase from Allochromatium vinosum can occur in several inactive or active states. This study presents the first systematic infrared characterisation of the A. vinosum enzyme, with emphasis on the spectro-electrochemical properties of the inactive/active transition. This transition involves an energy barrier, which can be overcome at elevated temperatures. The reduced Ready enzyme can exist in two different inactive states, which are in an apparent acid–base equilibrium. It is proposed that a hydroxyl ligand in a bridging position in the Ni-Fe site is protonated and that the formed water molecule is subsequently removed. This enables the active site to bind hydrogen in a bridging position, allowing the formation of the fully active state of the enzyme. It is further shown that the active site in enzyme reduced by 1 bar H₂ can occur in three different electron paramagnetic resonance (EPR)-silent states with a different degree of protonation.Abbreviations BV benzyl viologen - MB methylene blue - MBH membrane-bound hydrogenase - SHE standard hydrogen electrode     

Single crystal EPR studies of the oxidized active site of [NiFe] hydrogenase from Desulfovibrio vulgaris Miyazaki F

The Ni-A and the Ni-B forms of the [NiFe] hydrogenase from Desulfovibrio vulgaris Miyazaki F have been studied in single crystals by continuous wave and pulsed EPR spectroscopy at different temperatures (280?K, 80?K, and 10?K). For the first time, the orientation of the g-tensor axes with respect to the recently published atomic structure of the active site at 1.8?Å resolution was elucidated for Ni-A and Ni-B. The determined g-tensors have a similar orientation. The configuration of the electronic ground state is proposed to be Ni(III) 3d ¹ _z2 for Ni-A and Ni-B. The g _z principal axis is close to the Ni-S(Cys549) direction; the g _x and the g _y axes are approximately along the Ni-S(Cys546) and Ni-S(Cys81) bonds, respectively. It is proposed that the difference between the Ni-A and Ni-B states lies in a protonation of the bridging ligand between the Ni and the Fe.     

Structural differences in the inner part of Photosystem II between higher plants and cyanobacteria

A detailed comparison of key components in the Photosystem II complexes of higher plants and cyanobacteria was carried out. While the two complexes are overall very similar, significant differences exist in the relative orientation of individual components relative to one another. We compared a three-dimensional map of the inner part of plant PS II at 8 Å resolution, and a 5.5 Å projection map of the same complex determined by electron crystallography, to the recent 3.5–3.8 Å X-ray structures of cyanobacterial complexes. The largest differences were found in the rotational alignment of the cyt b_^559 subcomplex, and of the CP47 core antenna with respect to the D1/D2 reaction centre. Within the D1/D2 proteins, there are clear differences between plants and cyanobacteria at the stromal ends of membrane-spanning helices, even though these proteins are highly homologous. Notwithstanding these differences in the protein scaffold, the distances between the critical photosynthetic pigment cofactors seem to be precisely conserved. The different protein arrangements in the two complexes may reflect an adaptation to the two very different antenna systems, membrane-extrinsic phycobilisomes for cyanobacteria, and membrane-embedded chlorophyll a/b proteins in plants.