A multisubunit membrane-bound [NiFe] hydrogenase and an NADH-dependent Fe-only hydrogenase in the fermenting bacterium Thermoanaerobacter tengcongensis

Basem Soboh1, Dietmar Linder2 and Reiner Hedderich1

1 Max-Planck-Institut für terrestrische Mikrobiologie, Karl-von-Frisch-Straße, D-35043 Marburg, Germany
2 Biochemisches Institut, Fachbereich Humanmedizin, Justus-Liebig-Universität Giessen, Germany

Correspondence
Reiner Hedderich
hedderic{at}staff.uni-marburg.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Thermoanaerobacter tengcongensis is a thermophilic Gram-positive bacterium able to dispose of the reducing equivalents generated during the fermentation of glucose to acetate and CO2 by reducing H+ to H2. A unique combination of hydrogenases, a ferredoxin-dependent [NiFe] hydrogenase and an NADH-dependent Fe-only hydrogenase, were found to be responsible for H2 formation in this organism. Both enzymes were purified and characterized. The tightly membrane-bound [NiFe] hydrogenase belongs to a small group of complex-I-related [NiFe] hydrogenases and has highest sequence similarity to energy-converting [NiFe] hydrogenase (Ech) from Methanosarcina barkeri. A ferredoxin isolated from Ta. tengcongensis was identified as the physiological substrate of this enzyme. The heterotetrameric Fe-only hydrogenase was isolated from the soluble fraction. It contained FMN and multiple iron–sulfur clusters, and exhibited a typical H-cluster EPR signal after autooxidation. Sequence analysis predicted and kinetic studies confirmed that the enzyme is an NAD(H)-dependent Fe-only hydrogenase. When H2 was allowed to accumulate in the culture, the fermentation was partially shifted to ethanol production. In cells grown at high hydrogen partial pressure [p(H2)] the NADH-dependent hydrogenase activity was fourfold lower than in cells grown at low p(H2), whereas aldehyde dehydrogenase and alcohol dehydrogenase activities were higher in cells grown at elevated p(H2). These results indicate a regulation in response to the p(H2).


Abbreviations: ADH, alcohol dehydrogenase; ALDH, aldehyde dehydrogenase; BV, benzylviologen; Ech, energy-converting [NiFe] hydrogenase; Hyd, NAD(H)-dependent Fe-only hydrogenase from Thermotoga maritima and Thermoanaerobacter tengcongensis; MV, methylviologen, p(H2), hydrogen partial pressure

This work is dedicated to Professor Dr Rudolf K. Thauer on the occasion of his 65th birthday.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The decomposition of organic matter via fermentation is one of the major energy-yielding processes in anoxic habitats (Schmitz et al., 2003; Schwarz & Friedrich, 2003). Both obligate and facultative fermenting bacteria are able to reduce protons to H2, thereby releasing the reducing equivalents obtained from the anaerobic degradation of organic substrates. Depending on the organism, H2 production is either catalysed by an Fe-only hydrogenase or a [NiFe] hydrogenase (Schwarz & Friedrich, 2003).

The saccharolytic clostridia are well-studied members of the obligate fermenting bacteria (Gottschalk, 1986). H2 metabolism has in particular been studied in Clostridium pasteurianum. The organism ferments glucose via the Embden-Meyerhof pathway. Pyruvate is oxidized to acetyl-CoA and CO2 by pyruvate : ferredoxin oxidoreductase (POR). Reduced ferredoxin functions as electron donor for two soluble monomeric Fe-only hydrogenases (Adams, 1990). A portion of the NADH generated in the glyceraldehyde-3-phosphate dehydrogenase reaction serves to reduce ferredoxin, and thus electrons derived from NADH can also be used for H2 production (Jungermann et al., 1973). The reaction is catalysed by an NADH : ferredoxin oxidoreductase. Since the reduction of ferredoxin (E0' –420 mV) by NADH (E0' –320 mV) is energetically unfavourable, the NADH : ferredoxin oxidoreductase reaction is thought to be driven by reverse electron transport at high external hydrogen partial pressure [p(H2)]. A membrane-bound NADH : ferredoxin oxidoreductase has recently been purified from Clostridium tetanomorphum (W. Buckel, personal communication).

Another example of a micro-organism involved in fermentative H2 production is Thermotoga maritima, a thermophilic, strictly anaerobic bacterium (Schröder et al., 1994). H2 evolution in Tt. maritima is catalysed by a heterotrimeric cytoplasmic Fe-only hydrogenase (Verhagen et al., 1999). The purified enzyme accepts neither reduced ferredoxin nor NAD(P)H as electron donor. However, sequence analysis predicts that the enzyme is a flavoprotein containing a multiple iron–sulfur cluster that uses NAD(P)H as the electron donor (Verhagen et al., 1999). Another type of hydrogenase is responsible for fermentative H2 production in the hyperthermophilic archaeon Pyrococcus furiosus. The genome of P. furiosus contains the putative mbhA–N operon encoding a 14-subunit membrane-bound [NiFe] hydrogenase complex (Sapra et al., 2000; Silva et al., 2000). This enzyme complex is now thought to be responsible for fermentative H2 production in P. furiosus. An intact Mbh complex has not yet been purified. However, washed membranes of P. furiosus and partially purified Mbh preparations catalyse H2 production with reduced P. furiosus ferredoxin as electron donor (Silva et al., 2000). With inverted membrane vesicles of P. furiosus, it was shown that H2 evolution from reduced ferredoxin results in the generation of both a {Delta}pH and a {Delta}{psi}, which could be coupled to ATP synthesis. Hence, in addition to substrate-level phosphorylation, the organism gains energy by respiration. This also explains why P. furiosus has an unusual glycolytic pathway that uses ferredoxin in place of the expected NAD+ as electron acceptor for glyceraldehyde 3-phosphate oxidation (Sapra et al., 2003).

A classical example of fermentative H2 production in a facultative fermenting organism is the mixed acid fermentation of Escherichia coli (reviewed by Sawers, 1994). In this organism, pyruvate is converted to acetyl-CoA and formate by the pyruvate : formate lyase. Formate is subsequently converted to CO2 and H2. This latter reaction is catalysed by the formate : hydrogen lyase complex, which consists of a formate dehydrogenase, a polyferredoxin and a membrane-bound multisubunit [NiFe] hydrogenase (Böhm et al., 1990; Sauter et al., 1992).

In this study, the H2 metabolism in Ta. tengcongensis, a thermophilic Gram-positive bacterium recently isolated from a hot spring in China, was investigated. Batch cultures of this organism were previously shown to ferment starch or glucose via the Embden–Meyerhof pathway to acetate, ethanol, CO2 and H2 (Xue et al., 2001). We show here that the organism forms two distinct hydrogenases involved in H2 production during fermentation: a heterotetrameric NAD(H)-dependent Fe-only hydrogenase and a ferredoxin-dependent multisubunit membrane-bound [NiFe] hydrogenase.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Materials.
Dodecyl {beta}-D-maltoside was from Glycon Biochemicals. The chromatographic materials were from Amersham Pharmacia Biotech or Bio-Rad. All other chemicals were from Merck or Sigma.

Growth of the organism.
Ta. tengcongensis strain MB4T was from the Japan Collection of Microorganisms (JCM11007). The organism was cultivated at 75 °C and pH 7·5 in the medium described previously (Xue et al., 2001), with a few modifications. The medium was supplemented with tryptone (2 g l–1) and yeast extract (2 g l–1), and either soluble starch (10 g l–1) or glucose (25 mM) as energy substrate. The medium did not contain sodium thiosulfate or other external electron acceptors. Media were made anoxic by flushing with N2. Resazurin was used as redox indicator. The organism was cultivated either in 2 litre bottles (containing 1 litre of medium) tightly closed with rubber stoppers or in 12 litre fermenters (10 l medium), which were continuously flushed with N2 (~200 ml min–1) and stirred at 5000 r.p.m. Cultures at an OD578 of ~2 (fermenter cultures) were harvested under N2. Cells were stored at –20 °C.

Preparation of cell extracts and anaerobic procedures.
Cell extracts were routinely prepared from 40 g (wet wt) cells suspended in 120 ml 50 mM MOPS/KOH (pH 7·0) containing 2 mM dithiothreitol (DTT) (buffer A). Cells were disrupted by sonication at 4 °C with four bursts at 7 min intervals using an energy output of 200 W (Bandelin sonicator). Undisrupted cells and cell debris were removed by centrifugation at 10 000 g for 20 min. Cell extracts had to be kept strictly anoxic because hydrogenases were highly oxygen sensitive. All procedures for the preparation of cell fractions and enzyme purifications were carried out anoxically under an N2/H2 (95/5, v/v) atmosphere. All buffers were boiled, flushed with N2, supplied with 2 mM DTT, and maintained under a slight overpressure of N2.

Purification of the membrane-bound [NiFe] hydrogenase.
Crude membranes were isolated from cell extracts by ultracentrifugation at 160 000 g for 2 h. The 160 000 g supernatant was stored for the isolation of the soluble NADH-dependent hydrogenase. Crude membranes were resuspended in 130 ml buffer A using a Teflon Potter homogenizer. After a second ultracentrifugation at 160 000 g for 2 h, washed membranes were homogenized in ~300 ml buffer A (1·8 mg protein ml–1). Dodecyl {beta}-D-maltoside was added to a concentration of 16 mM [4·5 mg (mg protein)–1]. The suspension was incubated for 12 h at 4 °C with slight swirling. After centrifugation at 160 000 g for 40 min, the solubilized membrane proteins present in the supernatant were loaded onto a Q-Sepharose HiLoad column (2·6x15 cm) equilibrated with buffer A containing 2 mM dodecyl {beta}-D-maltoside (buffer A+detergent). The column was washed with 80 ml buffer A+detergent. Protein was eluted in a stepwise NaCl gradient at a flow rate of 5 ml min–1 (0·1, 0·2, 0·3, 0·4 and 0·5 M; 80 ml each in buffer A+detergent). The membrane-bound hydrogenase was recovered in the fractions eluting with 0·3 M NaCl. Hydrogenase-containing fractions were loaded on a ceramic hydroxyapatite column (1·6x20 cm) equilibrated with 0·03 M potassium phosphate buffer (pH 7·0) containing 2 mM DTT and 2 mM dodecyl {beta}-D-maltoside. Protein was eluted using a linear gradient from 0·03 to 1 M potassium phosphate (400 ml). The membrane-bound [NiFe] hydrogenase was recovered in the fractions eluting with 1 M potassium phosphate. Fractions containing hydrogenase activity were concentrated by ultrafiltration (YM100 ultrafiltration membranes, 100 kDa cut-off, Millipore) and applied to a Superdex 200 gel filtration column (2·6x60 cm) equilibrated with buffer A+detergent+0·1 M NaCl. The enzyme eluted after 225 ml (peak maximum). Protein was concentrated by ultrafiltration and stored in buffer A+detergent at a concentration of 3 mg protein ml–1 at 4 °C.

Purification of an NAD(H)-dependent Fe-only hydrogenase.
All buffers used throughout the purification contained 2 mM FMN. The 160 000 g supernatant (see above) was loaded onto a Q-Sepharose HiLoad column (2·6x15 cm) equilibrated with buffer A. The column was washed with 80 ml buffer A. Protein was eluted in a stepwise NaCl gradient at a flow rate of 5 ml min–1 (0·1, 0·2, 0·3, 0·4 and 0·5 M; 80 ml each in buffer A). The NAD+-reducing hydrogenase activity was recovered in the fractions eluting with 0·3 M NaCl. These fractions were brought to 0·6 M ammonium sulfate and loaded onto a hydrophobic interaction chromatography column (Phenyl-Sepharose HiLoad; 2·6x10 cm) equilibrated with 0·8 M ammonium sulfate in buffer A. Protein was eluted in a stepwise ammonium sulfate gradient at a flow rate of 5 ml min–1 (0·8, 0·6, 0·4, 0·2 and 0 M; 60 ml each in buffer A). NAD+-reducing hydrogenase activity was recovered in the fractions eluting with buffer A (0 M ammonium sulfate). These fractions were subjected to anion-exchange chromatography on Source 30 Q (5x10 cm). The column was washed with 100 ml buffer A. Protein was eluted in a stepwise NaCl gradient at a flow rate of 5 ml min–1 (0·1, 0·12, 0·14, 0·16, 0·18 and 0·2 M; 100 ml each in buffer A). The NAD+-reducing hydrogenase was recovered in the fractions eluting with 0·2 M NaCl. Fractions containing hydrogenase activity were concentrated and desalted by ultrafiltration and applied to a Superdex 200 gel filtration column (2·6x60 cm) equilibrated with buffer A+0·1 M NaCl. The enzyme eluted after 220 ml (peak maximum). The protein was concentrated by ultrafiltration and stored at a concentration of 8 mg protein ml–1 in buffer A at 4 °C.

Purification of ferredoxin.
Ta. tengcongensis ferredoxin was purified under anoxic conditions from the 160 000 g supernatant by chromatography on Q-Sepharose HiLoad (elution at 0·5 M NaCl), Phenyl Sepharose (elution at 0·4 M ammonium sulfate), and Superdex 75 (elution after 275 ml, peak maximum). The as-isolated oxidized ferredoxin showed absorption maxima at 280 and 390 nm, and an A390/A280 of 0·75. An {varepsilon}390 of 12·8 mM–1 cm–1 was used for the determination of ferredoxin concentrations. MALDI-TOF analysis resulted in a single peak corresponding to a molecular mass of 5724 Da.

Determination of enzyme activities.
Enzymes were routinely assayed at 70 °C either in 8 ml serum bottles or in 1·5 ml cuvettes (sealed with rubber stoppers) under anoxic conditions. In hydrogenase assays, 1 U enzyme activity corresponds to 1 µmol H2 formed or consumed min–1. Hydrogen-uptake activity with dyes as electron acceptors was determined by following the reduction of methylviologen (MV) or benzylviologen (BV) at 578 nm. The 0·8 ml assays contained buffer A and 4 mM BV or MV. Cuvettes were allowed to equilibrate with a 100 % H2 headspace (1·2x105 Pa). H2-formation activity with reduced MV as electron donor was measured by following the oxidation of reduced MV at 578 nm. The standard assay contained buffer A and 4 mM MV, which was reduced with sodium dithionite to a {Delta}A578 of 2. N2 (1·2x105 Pa) was the gas phase. The reaction was started by addition of enzyme (6 mU). Ferredoxin-dependent reduction of metronidazole by H2 was measured by following the decrease in A320 ({varepsilon}metronidazole=9·3 mM–1 cm–1, corresponding to a six-electron reduction). The 0·8 ml assay contained buffer A, 2 mM dodecyl {beta}-D-maltoside, 0·2 mM metronidazole, 20 µM ferredoxin, enzyme (0·5 mU) and 100 % H2 (1·2x105) Pa as the gas phase. To ensure completely anoxic conditions, 0·1 mM sodium dithionite was added. Ferredoxin-dependent H2 formation with sodium dithionite as electron donor was followed by determining the H2 concentration in the gas phase. The 1 ml assay in 8 ml serum bottles contained buffer A, 15 mM sodium dithionite, 20 µM ferredoxin (or as indicated) and enzyme (1·5 mU) under N2 as gas phase (1·2x105 Pa). The solution was stirred vigorously with a magnetic bar. At time intervals of 0·5 min, samples from the gas phase were withdrawn and H2 was quantified after separation by gas chromatography (see below). NAD(P)+-reducing hydrogenase activity was assayed by measuring the formation of NAD(P)H at 340 nm. The assay contained buffer A in a total volume of 0·8 ml. Cuvettes were allowed to equilibrate with 100 % H2 in the headspace (1·2x105 Pa). The reaction was initiated by the addition of NAD(P)+ (1·5 mM) or enzyme (0·1 U). NADH-dependent H2 formation was followed by determining the H2 concentration in the gas phase by gas chromatography (see below). The 1 ml assay in 24 ml serum bottles contained buffer A, 2 mM FMN, 1 mM Ti(III)citrate and 0·25 U enzyme. After a 2 min preincubation the reaction was initiated by the addition of NADH (50 mM). The solution was stirred vigorously with a magnetic bar. At 0·5 min intervals, samples from the gas phase were withdrawn and H2 was quantified after separation by gas chromatography. NADH : BV oxidoreductase activity was determined by following the NADH-dependent reduction of BV at 578 nm. The standard assay contained buffer A and 4 mM BV, which was reduced with sodium dithionite to an {Delta}A578 of 0·1. N2 (1·2x105 Pa) was the gas phase. The reaction was started by addition of NADH (0·2 mM) or enzyme (0·1 mU). Alcohol dehydrogenase (ADH) activity was assayed in both the forward and the reverse direction by measuring the oxidation or formation of NAD(P)H at 340 nm at 70 °C under a N2 gas phase. One unit of activity was defined as the amount of enzyme that forms or oxidizes 1 µmol NAD(P)H min–1. The reaction mixture (total volume, 1 ml) contained 50 mM Tris/HCl (pH 7·5 at 70 °C), 1·5 mM NAD(P)+, and 1·2 mM ethanol or 2-propanol (reverse direction); or 50 mM Tris/HCl (pH 7·5 at 70 °C), 0·2 mM NADH or NADPH, 1·2 mM acetaldehyde and enzyme (3 mU) (forward direction). Aldehyde dehydrogenase (ALDH) activity was determined by the acetaldehyde- and CoA-dependent reduction of NAD(P)+. One unit of ALDH activity was defined as the amount of enzyme that reduces 1 µmol NAD(P)+ min–1 at 70 °C under anoxic conditions. The reaction mixture (total volume: 1 ml) contained 50 mM Tris/HCl (pH 7·5), 1·5 mM NAD+ or NADP+, 1·2 mM acetaldehyde, 1·5 mM CoA and enzyme (1 mU).

Amino acid sequence analysis.
Sequence data of the Ta. tengcongensis genome were obtained from The Institute for Genomic Research website (http://www.tigr.org). For the prediction of transmembrane helices in proteins, programs at non-commercial servers were used (http://www.cbs.dtu.dk/services/TMHMM-2.0/). Multiple sequence alignments were made using the program at http://www.ebi.ac.uk/fasta3/.

Determination of amino acid sequences.
For determination of amino-terminal amino acid sequences, polypeptides were separated by SDS-PAGE and blotted on to PVDF membranes (Applied Biosystems) as described previously (Künkel et al., 1998). Sequences were determined using an Applied Biosystems 4774 protein/peptide sequencer and the protocol given by the manufacturer.

Analytical methods.
Protein concentration was routinely measured by the method of Bradford (1976) (Rotinanoquant; Roth) using BSA as a standard. Nickel was determined by atomic absorption spectroscopy on a 3030 Perkin Elmer atomic absorption spectrometer fitted with an HGA-600 graphite furnace assembly and an AS-60 autosampler. For identification of flavin and determination of the flavin content of the NADH-dependent hydrogenase, the enzyme was extensively washed by ultrafiltration with FMN-free buffer A. Protein (100 µl, 15 mg ml–1) was denatured by exposure to 10 % (w/v) trichloroacetic acid. Denatured protein was removed by centrifugation; the resulting supernatant was adjusted to pH 6 with 2 M K2HPO4 and analysed by chromatography using a reversed-phase HPLC column (LiChrospher 60 RP 18, 5 µm, 125x4 mm, Merck) equilibrated in 50 mM ammonium formate containing 25 % methanol. Flavins were eluted isocratically with the equilibration buffer. FAD and FMN standards were used to identify and quantify the flavin. H2 was determined by gas chromatography. The gas chromatograph (Carlo Erba GC Series 6000) was equipped with a thermal conductivity detector. Gases were separated by a molecular sieve (0·5 nm). The oven and injection port were at 110 °C; the detector was at 150 °C. The carrier gas was N2 at a flow rate of 30 ml min–1. Glucose was determined enzymically with glucose oxidase, peroxidase and 2,2-azino-di-(3-ethylbenzthiazoline)-6-sulfonate (ABTS) as chromogen (Kunst et al., 1981). Ethanol was determined enzymically with ADH and NAD+. Acetate was determined enzymically with acetyl-CoA synthetase, myokinase, pyruvate kinase and lactate dehydrogenase following the oxidation of NADH (Dorn et al., 1978). For the determination of ethanol in fermenter cultures, the outlet gas was led through 5 litres of cooled water (4 °C) to trap the ethanol.

EPR spectroscopy measurements.
EPR spectra at X-band (9·45 GHz) were obtained with a Bruker EMX spectrometer. All spectra were recorded with a field modulation frequency of 100 kHz and a modulation amplitude of 0·6 mT. The sample was cooled by an Oxford Instrument ESR 900 flow cryostat with an ITC4 temperature controller. EPR signals were simulated using non-commercial programs supplied by Dr S. P. J. Albracht Swammerdam Institute for Life Sciences, University of Amsterdam, Netherlands, and were based on formulas described previously (Beinert & Albracht, 1982).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Ta. tengcongensis when grown at low p(H2) ferments glucose to acetate, CO2 and H2, as will be described below. Cell extracts exhibited both NAD(H)- and ferredoxin-dependent hydrogenase activity. Two distinct hydrogenases, which were both purified and characterized, were found to be responsible for these activities.

A ferredoxin-dependent [NiFe] hydrogenase in the membrane fraction of Ta. tengcongensis
The membrane fraction of Ta. tengcongensis exhibited ferredoxin-dependent hydrogenase activity using Methanosarcina barkeri ferredoxin as substrate. This assay was used to identify and purify a ferredoxin from Ta. tengcongensis (see Methods), which was used in the studies described here. For routine activity measurements the H2-dependent-reduction of MV and the ferredoxin- and H2-dependent reduction of metronidazole were followed (Table 1). The enzyme could be solubilized by dodecyl {beta}-D-maltoside without loss of activity. Using the chromatographic steps summarized in Table 1, the enzyme was purified 200-fold to apparent homogeneity with a yield of 12 %. An analysis of the purified enzyme by SDS-PAGE revealed the presence of five polypeptides with apparent molecular masses of 69, 40, 32, 17 and 14 kDa (Fig. 1A). The enzyme exhibited an oxidized minus reduced UV/visible spectrum characteristic for iron–sulfur proteins (not shown). The enzyme contained 4·8 nmol Ni per mg protein, corresponding to 0·9 mol Ni per mol enzyme (187·7 kDa, see below).


View this table:
[in this window]
[in a new window]
 
Table 1. Purification of Ech hydrogenase from Ta. tengcongensis

The enzyme was purified from 40 g cells (3200 mg protein). Hydrogenase-uptake activity was measured after each chromatographic step by following the H2-dependent reduction of MV or the H2- and ferredoxin (Fd)-dependent reduction of metronidazole as described in Methods. One unit (U) of activity is defined as the amount of enzyme that catalysed the ferredoxin-dependent reduction of metronidazole with 1 µmol H2 min–1 or the reduction of 2 µmol of MV by H2 min–1.

 


View larger version (29K):
[in this window]
[in a new window]
 
Fig. 1. SDS-PAGE of purified hydrogenases from Ta. tengcongensis. (A) Ech hydrogenase (10 µg protein) was denatured for 30 min at room temperature in Laemmli buffer containing 5 mM DTT and 2 % SDS. (B) The purified NAD(H)-dependent hydrogenase (10 µg protein) was denatured for 5 min at 100 °C in Laemmli buffer containing 5 mM DTT and 2 % SDS. Proteins were separated in a 14 % polyacrylamide slab gel (8x7 cm), which was subsequently stained with Coomassie brilliant blue R250. The molecular masses (in kDa) of marker proteins (low-molecular-mass markers; Amersham Biosciences) are given on the right; the apparent molecular masses of the hydrogenase subunits and the subunit designations are given on the left. The polypeptide with an apparent molecular mass of 14 kDa was found to be a mixture of two polypeptides (Table 2).

 
The amino-terminal sequence of the 40 kDa and of the 14 kDa polypeptide was determined. The 14 kDa protein was found to be a mixture of two polypeptides with calculated molecular masses of 14·1 and 14·7 kDa (deduced from the nucleotide sequence, see below) (Table 2). The sequence information obtained allowed the encoding genes to be identified in the completely sequenced genome of Ta. tengcongensis (Bao et al., 2002). The genes were clustered and either were separated by only a few base pairs (less than 10) or overlapped, which indicated the formation of a transcription unit. The molecular masses of the deduced proteins were almost identical to those determined from the SDS-polyacrylamide gel (Fig. 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Amino-terminal sequences of the 40 kDa and the 14 kDa subunits of Ech hydrogenase from Ta. tengcongensis

Amino-terminal sequences derived by Edman degradation were used to identify the encoding genes in the published genome of Ta. tengcongensis (Bao et al., 2002).

 


View larger version (11K):
[in this window]
[in a new window]
 
Fig. 2. Genomic organization of the genes encoding the subunits of Ech hydrogenase. The designations of the genes encoding the Ech hydrogenase (ech) and the hydrogenase maturation proteins (hyp) are given above the gene map. The corresponding ORFs annotated by TIGR are TTE0123 (echA) to TTE0134 (hypE). The genes either have almost no intergenic regions (<10 bp) or overlap. echF and hypA are separated by only 2 bp.

 
The deduced amino acid sequences showed the highest level of identity to the amino acid sequences of six subunits of Ech hydrogenase from the methanogenic archaeon Ms. barkeri. The sequence identity for the different subunits varied between 63 % (EchB) and 38 % (EchD). Also, the gene order was the same in both organisms (Künkel et al., 1998; Meuer et al., 1999). The Ta. tengcongensis enzyme was therefore designated Ech hydrogenase. In both organisms, the echABCDEF genes encode two membrane proteins (EchA and EchB), a [NiFe]-carrying hydrogenase large subunit (EchE), a conserved hydrogenase small subunit with one [4Fe–4S] cluster (EchC), an additional iron–sulfur protein with two [4Fe–4S] cluster-binding motifs (EchF), and a small hydrophilic subunit with no detectable cofactor-binding site (EchD). In Ta. tengcongensis, the echF gene is directly followed by a second gene cluster, designated hypABFCDE, which encodes all proteins known to be essential for the biosynthesis of the [NiFe] centre of [NiFe] hydrogenases (Blokesch et al., 2002), with one exception (Fig. 2). This gene cluster does not encode a protease that could catalyse the C-terminal processing of the hydrogenase large subunit. The hydrogenase large subunit (EchE) of both Ta. tengcongensis and Ms. barkeri lacks the carboxy-terminal extension of the hydrogenase large subunit that is cleaved off after the correct insertion of the [NiFe] centre in other [NiFe] hydrogenases (Blokesch et al., 2002).

Purified Ech hydrogenase from Ta. tengcongensis catalysed H2 evolution with reduced ferredoxin as electron donor at a Vmax of 200 U (mg protein)–1 at 70 °C. The apparent Km for the reduced ferredoxin was 3 µM. The catalytic efficiency coefficient (kcat/Km) of 2x108 M–1 s–1 was calculated based on a molecular mass for Ech hydrogenase of 187·7 kDa. To determine the rate of ferredoxin reduction by H2, the metronidazole assay was used (Chen & Blanchard, 1979). Ech hydrogenase catalysed the reduction of metronidazole by H2 only in the presence of ferredoxin, which indicates that the ferredoxin is a direct electron acceptor of the enzyme. Reduced ferredoxin is oxidized in a fast chemical reaction by metronidazole. Ech hydrogenase catalysed the reduction of metronidazole and thus the reduction of the ferredoxin at a Vmax of 70 U (mg protein)–1. The enzyme exhibited an apparent Km for the oxidized ferredoxin of 15 µM. A catalytic efficiency coefficient (kcat/Km) of 1·5x107 M–1 s–1 was calculated. The enzyme was also assayed with MV as artificial electron donor or acceptor. The enzyme catalysed the reduction of MV by H2 at an apparent rate of 380 U (mg protein)–1 and H2 evolution from reduced MV at an apparent rate of 1260 U (mg protein)–1.

An NAD(H)-dependent Fe-only hydrogenase in the soluble fraction of Ta. tengcongensis
An NAD(H)-dependent Fe-only hydrogenase was purified from the soluble fraction of Ta. tengcongensis. The enzyme was enriched 20-fold with a yield of 27 %. For routine activity measurements, both the H2-dependent reduction of MV and the H2-dependent reduction of NAD+ were determined (Table 3). Activities with NAD(H) as substrate were significantly higher when FMN was added to all buffers used throughout the purification. SDS-PAGE of the purified enzyme revealed four polypeptides with apparent molecular masses of 65, 64, 20 and 14 kDa (Fig. 1B). The purified enzyme contained less than 0·06 mol Ni per mol enzyme. Upon autooxidation of the enzyme (for 20 min at 70 °C under N2), the EPR spectrum shown in Fig. 3 was obtained at 30 K. A rhombic signal with gxyz=2·106, 2·046 and 1·996 was identified in the spectrum. This signal could be simulated (Fig. 3, upper spectrum) and is characteristic for the oxidized H-cluster of Fe-only hydrogenases (Adams, 1990). The remaining features of the spectrum most probably resulted from the [4Fe–4S] clusters or [2Fe–2S] clusters of the enzyme (see below), which were not fully oxidized under the conditions employed. In addition the enzyme contained 6·6 nmol FMN per mg protein, which corresponded to 1 mol FMN per mol enzyme.


View this table:
[in this window]
[in a new window]
 
Table 3. Purification of NAD(H)-dependent hydrogenase from Ta. tengcongensis

The enzyme was purified from 40 g cells (3200 mg protein). Hydrogenase-uptake activity was measured after each chromatographic step by following the H2-dependent reduction of MV or the H2-dependent reduction of NAD+ as described in Methods. One unit (U) of activity is defined as the amount of enzyme that catalysed the reduction of NAD+ or MV with 1 µmol H2 min–1.

 


View larger version (18K):
[in this window]
[in a new window]
 
Fig. 3. EPR spectrum of NAD(H)-dependent hydrogenase from Ta. tengcongensis after autooxidation. NADH-oxidizing hydrogenase (4·8 mg protein ml–1) was autooxidized at 70 °C for 20 min under N2 (lower spectrum). The simulation of the H-cluster signal is shown in the upper spectrum. EPR conditions: temperature, 30 K; microwave power, 3 mW; microwave frequency, 9·45 GHz; modulation amplitude, 0·6 mT; modulation frequency, 100 kHz. Simulation parameters: g1,2,3=1·996, 2·046 and 2·106; W1,2,3=0·53, 0·67 and 0·65 mT.

 
The purified enzyme catalysed H2 evolution with NADH as electron donor at a Vmax of 10 U (mg protein)–1 at 70 °C. Based on a calculated molecular mass for the enzyme of 162·5 kDa and an apparent Km for NADH of 17 µM (see below), a catalytic efficiency coefficient (kcat/Km) of 1·6x106 M–1 s–1 was calculated. Reduction of NAD+ by H2 was catalysed at a Vmax of 5 U (mg protein)–1. The Km for NAD+ was 90 µM. A catalytic efficiency coefficient (kcat/Km) of 1·5x105 M–1 s–1 was calculated. The purified enzyme was highly specific for NAD(H); no activity was obtained with NADP(H). The enzyme also catalysed the reduction of BV by NADH at a rate of 1300 U (mg protein)–1. With this assay, an apparent Km for NADH of 17 µM was determined. The enzyme catalysed the reduction of MV by H2 at an apparent rate of 1700 U (mg protein)–1 and H2 evolution from reduced MV at an apparent rate of 1700 U (mg protein)–1. The data strongly indicate that electron transfer between the catalytic site for NADH-oxidation and the hydrogenase catalytic site is rate-limiting.

The determination of the amino-terminal sequences allowed the identification of the encoding genes in the completely sequenced genome of Ta. tengcongensis (Bao et al., 2002; Table 4). The genes are clustered and form a putative transcription unit (Fig. 4). The sequence of the 64 kDa subunit (HydA) (581 amino acids) could be perfectly aligned (42 % sequence identity) with the single-subunit hydrogenase from Cl. pasteurianum, for which the crystal structure is known (Fig. 5) (Peters et al., 1998). In addition to the H-cluster present in the catalytic centre, the Cl. pasteurianum hydrogenase harbours three [4Fe–4S] clusters and one [2Fe–2S] cluster. The cysteine residues coordinating these clusters are also conserved in the Ta. tengcongensis 64 kDa subunit, which identifies this subunit as the hydrogenase catalytic subunit.


View this table:
[in this window]
[in a new window]
 
Table 4. Amino-terminal sequences of the subunits of NADH-dependent hydrogenase from Ta. tengcongensis

Amino-terminal sequences derived by Edman degradation were used to identify the encoding genes in the published genome of Ta. tengcongensis (Bao et al., 2002). X, No clear assignment to an amino acid could be made.

 


View larger version (9K):
[in this window]
[in a new window]
 
Fig. 4. Genomic organization of the genes encoding the subunits of NAD(H)-dependent hydrogenase. The designations of the genes encoding the hydrogenase are given above the gene map. The corresponding ORFs annotated by TIGR are TTE0890 (hydC)–TTE0894 (hydA). The genes either have almost no intergenic regions or overlap. The protein encoded by the ORF TTE0891 was not present in the purified enzyme.

 


View larger version (28K):
[in this window]
[in a new window]
 
Fig. 5. Domain organization of multimeric Fe-only hydrogenases. Grey boxes represent hydrogenase subunits. The monomeric hydrogenase from Cl. pasteurianum is shown for comparison. H, H-cluster; [2Fe–2S], [2Fe–2S] cluster binding site; [4Fe–4S], [4Fe–4S] cluster binding site; F, FMN and NAD(P)+ binding site [modified from (Schwarz & Friedrich, 2003)]. Original data were from Malki et al. (1995) and Verhagen et al. (1999).

 
The 65 kDa subunit (HydB) showed sequence similarity to NAD(P)H-dependent dehydrogenases, e.g. subunit NuoF of NADH : quinone oxidoreductase (Weidner et al., 1993). The amino acid sequence of this subunit could be aligned over its full length with subunit HydB of the Fe-only hydrogenase from Tt. maritima (Verhagen et al., 1999). Both proteins had 65 % sequence identity. HydB from both organisms contains an NAD(P)+-binding site and an FMN-binding site and is predicted to ligate three [4Fe–4S] clusters and one [2Fe–2S] cluster, as described previously (Verhagen et al, 1999). The 20 kDa subunit (HydC) showed the highest level of sequence identity (47 %) to subunit HndA of the NADP+-reactive hydrogenase from Desulfovibrio fructosovorans (Malki et al., 1995) and to subunit HydC of the Fe-only hydrogenase from Tt. maritima (Verhagen et al., 1999). These proteins share four conserved cysteine residues, which in Dv. fructosovorans and Tt. maritima have been shown to ligate a [2Fe–2S] cluster (De Luca et al., 1998; Verhagen et al., 2001). Based on sequence similarity, the 14 kDa subunit (HydD) is closely related (36 % sequence identity) to subunit HndB of the NADP+-reducing hydrogenase from Dv. fructosovorans. Both proteins have three conserved cysteine residues.

As shown in Fig. 4, an additional ORF (TTE0891) is located between hydD and hydC in the Ta. tengcongensis genome. It encodes a 20·7 kDa protein, which was not part of the purified enzyme. The protein showed high sequence similarity to the ATP-binding domain of histidine kinases.

Fermentation products and enzymic activities in Ta. tengcongenis cultures grown at different p(H2)
It has previously been shown that Ta. tengcongensis, when grown in batch cultures, produces 1 mmol acetate, 0·7 mmol ethanol, 1·5 mmol CO2 and 0·3 mmol H2 per mmol glucose consumed. In that study, thiosulfate was added as external electron acceptor (Xue et al., 2001). In the present study, the fermentation balance was determined with cultures grown in the absence of any external electron acceptors. Since growth on glucose in a closed system in the absence of thiosulfate was extremely poor (growth rate, µ=0·1 h–1), starch was used as carbon and energy source. This resulted in a µ of 0·27 h–1. Acetate, ethanol and H2 were formed in a 1·4±0·2 : 0·6±0·2 : 2·8±0·2 ratio. The maximal p(H2) attained in batch cultures was 10–1 atm. Growth of Ta. tengcongensis was most probably inhibited at higher p(H2), as has previously been described for other fermenting organisms (Schröder et al., 1994; van Niel et al., 2003).

Growth of Ta. tengcongensis in a fermenter continuously flushed with N2 to keep the p(H2) low resulted in higher growth rates (µ=0·24 h–1 with glucose as substrate). Glucose or starch could be used as substrate. Per mmol glucose consumed, 2±0·2 mmol acetate was formed. Only trace amounts of ethanol were detected (0·02 mmol per mmol glucose consumed). H2 could not be quantified accurately in these experiments owing to the high N2 flushing rates; however, since glucose was almost quantitatively converted to acetate and since no external electron acceptors were added, reduction of H+ to H2 is the only way to release the reducing equivalents generated during glucose oxidation to acetate. Thus, 4 mmol H2 must have been formed per mmol glucose consumed, to account for an even [H] balance.

To study if the enzymes involved in H2 and ethanol formation are regulated in response to p(H2), activities of the key enzymes in cell extracts were measured. For these experiments starch was used as substrate for the closed-bottle cultures and glucose was used as substrate for the fermenter cultures; under these conditions, the two cultures exhibited similar growth rates (µ=0·24 h–1 or 0·27 h–1). Cell extracts of fermenter-grown cells catalysed H2 production from NADH with specific activities of 0·6 U (mg protein)–1 and from reduced ferredoxin with specific activities of 1·8 U (mg protein)–1. While the specific activity of the ferredoxin-dependent hydrogenase activity was not influenced by the growth conditions, the NAD(H)-dependent hydrogenase activity was about fourfold lower in cell extracts from closed-bottle cultures. Also the activities of the two key enzymes involved in ethanol formation, alcohol dehydrogenase (ADH) and aldehyde dehydrogenase (ALDH), were determined (Table 5). As will be described in more detail in the Discussion, ethanol formation has been extensively studied in Thermoanaerobacter ethanolicus. In this organism the ADH responsible for fermentative ethanol formation was NADP(H) specific and, in addition to ethanol, used secondary alcohols as substrate. Cell extracts of closed-bottle cultures catalysed the NADPH-dependent reduction of acetaldehyde to ethanol with specific activities of 15 U (mg protein)–1. Cell extracts of fermenter cultures catalysed this reaction at fivefold lower rates [3 U (mg protein)–1]. When NADPH as electron donor was replaced by NADH, the rates of acetaldehyde reduction were approximately 15-fold lower. The reverse reaction, the NADP+- or NAD+-dependent oxidation of ethanol, was catalysed at significantly lower rates (Table 5). 2-Propanol was also tested as substrate of T. tengcongensis. Cell extracts from closed-bottle cultures catalysed the NADP+-dependent oxidation of 2-propanol at a rate of 11 U (mg protein)–1; cell extracts from fermenter cultures catalysed the oxidation at a rate of 1·5 U (mg protein)–1. ALDH activity was determined by following the acetaldehyde-dependent reduction of NADP+. Cell extracts from closed-bottle cultures catalysed this reaction with a specific activity of 1·8 U (mg protein)–1 and those from fermenter cultures with a specific activity of 0·2 U (mg protein)–1. The reaction was strictly CoA-dependent, which indicated the formation of acetyl-CoA. No activity was observed when NADP+ was replaced by NAD+ as substrate.


View this table:
[in this window]
[in a new window]
 
Table 5. Rates of reactions specific for enzymes involved in H2 and ethanol formation in Ta. tengcongensis cell extracts

Cells were grown in closed bottles or fermenter cultures. ND, Not determined.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The results presented here show that the thermophilic Gram-positive bacterium Ta. tengcongensis has a unique set of hydrogenases, an NADH-dependent Fe-only hydrogenase and a ferredoxin-dependent [NiFe] hydrogenase, for the release of excess reducing equivalents generated during fermentation as H2. H2 metabolism in Ta. tengcongensis thus clearly differs from that in other fermentative micro-organisms, which employ either a ferredoxin-dependent or an NADH-dependent system for H2 evolution. Some clostridia have been shown to have an NADH : ferredoxin oxidoreductase that allows H2 formation from NADH via reduced ferredoxin (Jungermann et al., 1973).

The [NiFe] hydrogenase of Ta. tengcongensis belongs to a small family of multisubunit membrane-bound enzymes that have been identified in a few bacteria and archaea in recent years. These enzymes form a distinct group within the large family of [NiFe] hydrogenases (Vignais et al., 2001). This list of enzymes includes hydrogenases, three from Escherichia coli (Böhm et al., 1990; Sauter et al., 1992), CO-induced hydrogenase from Rhodospirillum rubrum and Carboxydothermus hydrogenoformans (Fox et al., 1996b; Soboh et al., 2002) and Ech hydrogenase from Ms. barkeri (Künkel et al., 1998; Meuer et al., 1999). The large and small subunits of these hydrogenases show surprisingly little sequence similarity to other (standard) [NiFe] hydrogenases, except for the conserved residues coordinating the active site and the proximal [Fe–S] cluster (Albracht & Hedderich, 2000). In addition to the hydrogenase large and small subunits, these enzymes contain four other subunits: two hydrophilic proteins and two integral membrane proteins. These six subunits are conserved and form the basic module of these hydrogenases. To date, only two members of this hydrogenase family have been purified: Ech hydrogenase from Ms. barkeri (Meuer et al., 1999) and the hydrogenase from C. hydrogenoformans. The latter enzyme forms a tight complex with a CO dehydrogenase and a polyferredoxin (Soboh et al., 2002). This enzyme complex catalyses the conversion of CO to CO2 and H2. Ech from Ms. barkeri can use a ferredoxin isolated from Ms. barkeri as substrate. Kinetic studies have revealed that ferredoxin reduction by H2 is slightly favoured over H2 evolution using reduced ferredoxin as electron donor (Meuer et al., 1999). Genetic studies have shown that both reactions are of physiological importance (Meuer et al., 2002). A few other members of this hydrogenase family having a much more complex subunit architecture, with up to 20 predicted subunits, have been identified. These include the Mbh hydrogenase from Pyrococcus furiosus (Sapra et al., 2000; Silva et al., 2000) and the Eha and Ehb hydrogenases from Methanothermobacter species (Tersteegen & Hedderich, 1999); these enzymes have not yet been purified. The subunits conserved in Ech-type hydrogenases show a striking amino acid sequence similarity to six subunits of the energy-conserving NADH : quinone oxidoreductase, also known as complex I (Albracht & Hedderich, 2000; Friedrich & Scheide, 2000; Friedrich & Weiss, 1997; Hedderich, 2004; Yano & Ohnishi, 2001). From growth experiments with R. rubrum and C. hydrogenoformans (Kerby et al., 1995; Svetlichny et al., 1991), from cell-suspension experiments and genetic studies with Ms. barkeri (Bott & Thauer, 1989; Meuer et al., 2002), and from experiments with inverted membrane vesicles of P. furiosus (Sapra et al., 2003), it can be inferred that the [NiFe] hydrogenases in these organisms probably pump protons or sodium ions. These enzymes have therefore been designated energy-converting [NiFe] hydrogenases (Ech) (Vignais et al., 2001).

Here we report the first purification of a member of this hydrogenase family from a fermentative organism. The kinetic data obtained strongly indicate that Ta. tengcongensis ferredoxin is the physiological substrate. Ech from Ta. tengcongensis favours hydrogen evolution over H2 uptake, which is in agreement with its role in H2 production during fermentative growth. The same function has recently been proposed for the Mbh hydrogenase from P. furiosus. The membrane fraction of P. furiosus catalyses H2 production from reduced ferredoxin at high rates (Silva et al., 2000). Addition of reduced ferredoxin to inverted membrane vesicles of P. furiosus results in the generation of both a {Delta}pH and a {Delta}{psi}, which can be coupled to ATP synthesis (Sapra et al., 2003). This ‘proton respiration’ is proposed to also operate in Ta. tengcongensis.

The multimeric NADH-dependent Fe-only hydrogenase from Ta. tengcongensis is most closely related to the HndABCD hydrogenase from Dv. fructosovorans and to the HydABC hydrogenase from Tt. maritima (Malki et al., 1995; Verhagen et al., 1999). The subunit architecture is conserved in the three enzymes, with only a few differences (Fig. 5). The Ta. tengcongensis enzyme and the Dv. fructosovorans enzyme contain a fourth subunit (HydD and HndB) not present in the Tt. maritima hydrogenase. The function of this subunit, which contains three conserved cysteine residues, is unknown. Subunit HydB of the Tt. maritima and the Ta. tengcongensis enzyme contains an amino-terminal extension of 106 amino acids, which is lacking in the corresponding subunit (HndC) of the Dv. fructosovorans enzyme. This extension of HydB is predicted to ligate a [2Fe–2S] cluster. The Tt. maritima HydA subunit contains a carboxy-terminal extension of 65 amino acids with one [2Fe–2S] binding motif. This extension is lacking in subunit HydA of the Ta. tengcongensis enzyme and in subunit HndA of the Dv. fructosovorans enzyme.

The low amount of the enzyme from Dv. fructosovorans relative to the other cell proteins has prevented its purification. However, the characterization of an hndC deletion mutant has shown that the enzyme is responsible for the reduction of NADP+ by H2 (Malki et al., 1995). H2-dependent NADP+-reductase activity is 10-fold higher in H2-grown cells. NAD+ is not reduced by cell extracts of Dv. fructosovorans. Therefore, the enzyme from Dv. fructosovorans is thought to function as an NADP+-reducing H2-uptake hydrogenase (Malki et al., 1995). The HydABC hydrogenase from Tt. maritima has been purified as a heterotrimeric enzyme. Sequence analysis predicts a flavoprotein containing multiple iron–sulfur clusters that uses NADH or NADPH as the electron donor. However, the purified enzyme neither contains a flavin nor does it catalyse NAD(P)H-dependent H2 evolution or reduction of NAD(P)+ by H2 (Verhagen et al., 1999).

In contrast, the enzyme from Ta. tengcongensis contained equimolar amounts of FMN and could utilize NAD(H) as substrate. Analysis of the deduced amino acid sequence of the enzyme predicts the presence of two active sites: the hydrogenase active site containing the H-cluster and the NAD(H) oxidation/reduction active site containing the flavin cofactor. Both active sites are interconnected by a large number of iron–sulfur clusters. The activities of the two active sites could be determined individually using viologen dyes as one redox partner. The enzyme catalysed the NADH-dependent reduction of BV and H2 evolution from reduced MV at very high rates. Compared to these activities, the total activity, the NADH-dependent H2 evolution and the reduction of NAD+ by H2 were approximately 130- to 170-fold lower. Hence, the electron transfer between the two catalytic centres must be the rate-limiting step. One possible reason for the low electron transfer rate could be that one of the electron-transferring metal centres became damaged during enzyme purification. However, even in cell extracts, the rate of NADH-dependent H2 evolution was approximately 140-fold lower than the individual activities determined with viologen dyes (not shown). Hence, the low electron transfer rate might be an intrinsic property of the enzyme. On the other hand, the rates of NADH-dependent H2 evolution [0·5 U (mg protein)–1] determined in cell extracts should be sufficient to account for the growth of the organism since the rates of glucose consumption in growing cultures were in the same range [0·3 µmol min–1 (mg protein)–1; data not shown].

In Ta. tengcongensis the Fe-only hydrogenase gene-cluster contains an additional ORF(TTE0891), which does not encode a hydrogenase subunit (Fig. 4). TTE0891 encodes a 21 kDa protein. Closely related proteins (40 % sequence identity) with a similar molecular mass are encoded by the genomes of Tt. maritima (TM1665) and Clostridium thermocellum. While in Tt. maritima the encoding gene is not linked to a hydrogenase gene cluster, in Cl. thermocellum, this gene is part of a gene cluster encoding a putative NAD(H)-dependent Fe-only hydrogenase. This was deduced from the preliminary genome sequence of Cl. thermocellum (http://genome.jgi-psf.org/draft_microbes/cloth/cloth.home.html). A sequence analysis revealed a high level of sequence identity between the amino-terminal domain of the protein from these three organisms (first 90 amino acids) and the ATP-binding kinase domain of histidine kinases (West & Stock, 2001). The proteins, however, lack the dimerization domain of histidine kinases with the conserved histidine residue. The carboxy-terminal sequence of the protein (90 amino acids) does not show significant sequence similarity to known proteins. Since the ATP-binding domain of histidine kinases is also conserved in the ATPase domain of gyrase B and the chaperone Hsp90, the function of the protein encoded by TTE0891 remains elusive. This gene product could also be involved in hydrogenase assembly. Thus far, accessory genes possibly required for Fe-hydrogenase assembly have not been identified. In addition to the Ta. tengcongensis hydrogenase, the Fe-only hydrogenase from Tt. maritima is the only example studied where non-structural genes (TM1420, TM1421, TM1422, TM1423 and TM1427) are part of the hydrogenase gene-cluster. Three of the five accessory genes associated with the hydrogenase genes in Tt. maritima encode iron–sulfur proteins but their function is not yet known (Pan et al., 2003). In the Ta. tengcongensis genome these genes are also conserved but are not in the proximity of the hydrogenase genes.

Interestingly, both hydrogenases of Ta. tengcongensis are related to complex I. The Fe-only hydrogenase is related to the NADH-dehydrogenase fragment of complex I. The subunits of the enzyme were incorrectly annotated as complex I subunits NuoG (75 kDa subunit), NuoE (24 kDa subunit) and NuoF (51 kDa subunit) (Bao et al., 2002). The results presented here show that these subunits form an Fe-only hydrogenase and are not part of complex I. The [NiFe] hydrogenase is related to the central energy-conserving module of complex I as described above.

Ta. tengcongensis uses the enzymes of the Embden–Meyerhof pathway to convert glucose to pyruvate. The pathway generates 2 ATP and 2 NADH. Subsequently, pyruvate is oxidized to acetyl-CoA and CO2 via pyruvate : ferredoxin oxidoreductase (POR), thus generating reduced ferredoxin. Acetyl-CoA is converted to acetate via phosphate transacetylase and acetate kinase. The reaction results in the formation of 1 mol ATP per mol acetate. The formation of the enzymes catalysing the different steps has recently been shown in proteomic studies (Wang et al., 2004). Formation of POR was also shown in this study via activity measurements (data not shown). When the p(H2) in the culture was kept low (approx. 10–4 atm), glucose was completely converted to acetate, CO2 and H2. Thus, both the oxidation of NADH and the oxidation of reduced ferredoxin are coupled to H2 formation. At a p(H2) of 10–4 atm and glucose, acetate and , each at a concentration of ~10 mM, the {Delta}G' for the conversion of glucose to 2 acetate, 2 CO2 and 4 H2 is –329 kJ (mol glucose)–1 at 25 °C (Tewes & Thauer, 1980). At 70 °C, the {Delta}G' increases to approximately –352 kJ mol–1 (Amend & Plyasunov, 2001; Amend & Shock, 2001). This would allow the formation of 5 mol ATP, assuming that +70 kJ are required to drive the synthesis of ATP from ADP and Pi at physiological concentrations and under irreversible conditions (Thauer et al., 1977). This is significantly more than the 4 mol ATP generated by substrate-level phosphorylation. Hence, the thermodynamic data would allow the conservation of extra energy in the Ech hydrogenase reaction as has been proposed above. In cultures grown in closed bottles, where a p(H2) of up to 10–1 atm was obtained, the reaction becomes less exergonic ({Delta}G'=–284 kJ mol–1 at 70 °C). Therefore, the metabolism is shifted towards ethanol formation, which results in the formation of less ATP. Under these conditions reduced ferredoxin is the major source for H2 production; more than 50 % of the NADH is oxidized via the reduction of acetyl-CoA to ethanol.

The metabolic shift by Ta. tengcongensis in response to the p(H2) is consistent with the observation that in cells grown in closed bottles, the NADH-dependent hydrogenase activity was lower than in cells grown in the fermenter, whereas ALDH and ADH activities were higher in cells grown in closed bottles. The activity of the ferredoxin-dependent hydrogenase was almost identical under the two culture conditions (Table 5). The changes in the activities of NADH-dependent hydrogenase, ADH and ALDH indicate that either the formation of these enzymes at the transcriptional or translational level is regulated, or these enzymes are directly (in)activated at the protein level.

In this study, the enzymes involved in ethanol production by Ta. tengcongensis were investigated only in cell extracts. In earlier studies, ethanol fermentation has been studied in detail in Ta. ethanolicus, which produces ethanol as the major fermentation product. From this organism, a primary and a so-called secondary ADH have been purified and characterized (Burdette et al., 1996; Burdette & Zeikus, 1994). Both enzymes utilize a broad range of alcohols as substrates, but the enzymes can be clearly distinguished by their cofactor and substrate preferences. The secondary ADH is specific for NADP(H), while the primary ADH uses both pyridine nucleotides. The secondary ADH also reduces 2-propanol or butan-2-ol, which are not substrates of the primary ADH. The secondary ADH is a bifunctional ADH-acetyl-CoA thioesterase that catalyses the reduction of acetyl-CoA to ethanol. It has been proposed that this is the physiological reaction catalysed by the enzyme in vivo. A Tt. ethanolicus mutant strain that lacks the primary ADH is still able to carry out ethanol fermentation, which indicates that the secondary ADH is responsible for fermentative ethanol production (Burdette et al., 2002). The Ta. tengcongensis genome contains an ORF (TTE0695) encoding a protein with 96 % sequence identity to the secondary ADH from Tt. ethanolicus. It also contains an ORF (TTE0696) encoding a protein with 91 % sequence identity to the primary ADH of Tt. ethanolicus. The activity profile determined in cell extracts of Ta. tengcongensis (Table 5) indicated that both ADHs are formed also in Ta. tengcongensis. Although cell extracts of Ta. tengcongensis exhibited high ALDH activity, the genome does not encode an ALDH related to one of the characterized ALDHs. The secondary ADH of Ta. tengcongensis has high sequence identity to the corresponding enzyme from Tt. ethanolicus and therefore is proposed to be bifunctional as well, exhibiting both ADH and ALDH activities. A point to be clarified in a future study is how the NADH generated in the glyceraldehyde-3-phosphate dehydrogenase reaction is converted to NADPH, which is the substrate of the bifunctional ALDH/ADH. Glyceraldehyde-3-phosphate dehydrogenase of Ta. tengcongensis was found to be NAD+-specific (data not shown).


   ACKNOWLEDGEMENTS
 
This work was supported by the Max-Planck-Gesellschaft, by the Deutsche Forschungsgemeinschaft and by the Fonds der Chemischen Industrie. Antonio Pierik is acknowledged for help with EPR spectroscopy and valuable discussions. We thank Karen Brune for editing the manuscript.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Adams, M. W. (1990). The structure and mechanism of iron-hydrogenases. Biochim Biophys Acta 1020, 115–145.[Medline]

Albracht, S. P. & Hedderich, R. (2000). Learning from hydrogenases: location of a proton pump and of a second FMN in bovine NADH-ubiquinone oxidoreductase (Complex I). FEBS Lett 485, 1–6.[CrossRef][Medline]

Amend, J. P. & Plyasunov, A. V. (2001). Carbohydrates in thermophile metabolism: calculation of the standard molal thermodynamic properties of aqueous pentoses and hexoses at elevated temperatures and pressures. Geochim Cosmochim Acta 65, 3901–3917.[CrossRef]

Amend, J. P. & Shock, E. L. (2001). Energetics of overall metabolic reactions of thermophilic and hyperthermophilic archaea and bacteria. FEMS Microbiol Rev 25, 175–243.[CrossRef][Medline]

Bao, Q., Tian, Y., Li, W. & 18 other authors (2002). A complete sequence of the T. tengcongensis genome. Genome Res 12, 689–700.[Abstract/Free Full Text]

Beinert, H. & Albracht, S. P. J. (1982). New insights, ideas and unanswered questions concerning iron-sulfur clusters in mitochondria. Biochim Biophys Acta 683, 245–277.[Medline]

Blokesch, M., Paschos, A., Theodoratou, E., Bauer, A., Hube, M., Huth, S. & Bock, A. (2002). Metal insertion into NiFe-hydrogenases. Biochem Soc Trans 30, 674–680.[Medline]

Böhm, R., Sauter, M. & Böck, A. (1990). Nucleotide sequence and expression of an operon in Escherichia coli coding for formate hydrogenlyase components. Mol Microbiol 4, 231–243.[Medline]

Bott, M. & Thauer, R. K. (1989). Proton translocation coupled to the oxidation of carbon monoxide to CO2 and H2 in Methanosarcina barkeri. Eur J Biochem 179, 469–472.[Abstract]

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.[CrossRef][Medline]

Burdette, D. S. & Zeikus, J. G. (1994). Purification of acetaldehyde dehydrogenase and alcohol dehydrogenases from Thermoanaerobacter ethanolicus 39e and characterisation of the secondary-alcohol dehydrogenase as a bifunctional alcohol dehydrogenase-acetyl-CoA reductive thioesterase. Biochem J 302, 163–170.[Medline]

Burdette, D. S., Vieille, C. & Zeikus, J. G. (1996). Cloning and expression of the gene encoding the Thermoanaerobacter ethanolicus 39E secondary-alcohol dehydrogenase and biochemical characterization of the enzyme. Biochem J 316, 115–122.[Medline]

Burdette, D. S., Jung, S. H., Shen, G. J., Hollingsworth, R. I. & Zeikus, J. G. (2002). Physiological function of alcohol dehydrogenases and long-chain (C-30) fatty acids in alcohol tolerance of Thermoanaerobacter ethanolicus. Appl Environ Microbiol 68, 1914–1918.[Abstract/Free Full Text]

Chen, J.-S. & Blanchard, D. K. (1979). A simple hydrogenase-linked assay for ferredoxin and flavodoxin. Anal Biochem 93, 216–222.[Medline]

De Luca, G., Asso, M., Belaich, J. P. & Dermoun, Z. (1998). Purification and characterization of the HndA subunit of NADP-reducing hydrogenase from Desulfovibrio fructosovorans overproduced in Escherichia coli. Biochemistry 37, 2660–2665.[CrossRef][Medline]

Dorn, M., Andreesen, J. R. & Gottschalk, G. (1978). Fermentation of fumarate and L-malate by Clostridium formicoaceticum. J Bacteriol 133, 26–32.[Medline]

Fox, J. D., He, Y., Shelver, D., Roberts, G. P. & Ludden, P. W. (1996b). Characterization of the region encoding the CO-induced hydrogenase of Rhodospirillum rubrum. J Bacteriol 178, 6200–6208.[Abstract]

Friedrich, T. & Scheide, D. (2000). The respiratory complex I of bacteria, archaea and eukarya and its module common with membrane-bound multisubunit hydrogenases. FEBS Lett 479, 1–5.[CrossRef][Medline]

Friedrich, T. & Weiss, H. (1997). Modular evolution of the respiratory NADH:ubiquinone oxidoreductase and the origin of its modules. J Theor Biol 187, 529–540.[CrossRef][Medline]

Gottschalk, G. (1986). Bacterial Metabolism. New York: Springer.

Hedderich, R. (2004). Energy-converting [NiFe] hydrogenases from archaea and extremophiles: ancestors of complex I. J Bioenerg Biomembr 36, 65–75.[CrossRef][Medline]

Jungermann, K., Thauer, R. K., Leimenstoll, G. & Decker, K. (1973). Function of reduced pyridine nucleotide-ferredoxin oxidoreductases in saccharolytic Clostridia. Biochim Biophys Acta 305, 268–280.[Medline]

Kerby, R. L., Ludden, P. W. & Roberts, G. P. (1995). Carbon monoxide-dependent growth of Rhodospirillum rubrum. J Bacteriol 177, 2241–2244.[Abstract]

Künkel, A., Vorholt, J. A., Thauer, R. K. & Hedderich, R. (1998). An Escherichia coli hydrogenase-3-type hydrogenase in methanogenic archaea. Eur J Biochem 252, 467–476.[Abstract]

Kunst, A., Draeger, B. & Ziegenhorn, J. (1981). Colorimetric methods with glucose oxidase and peroxidase. In Methods of Enzymatic Analysis, pp. 178–185. Edited by H. U. Bergmeyer. Weinheim: Verlag Chemie.

Malki, S., Saimmaime, I., De Luca, G., Rousset, M., Dermoun, Z. & Belaich, J. P. (1995). Characterization of an operon encoding an NADP-reducing hydrogenase in Desulfovibrio fructosovorans. J Bacteriol 177, 2628–2636.[Abstract]

Meuer, J., Bartoschek, S., Koch, J., Künkel, A. & Hedderich, R. (1999). Purification and catalytic properties of Ech hydrogenase from Methanosarcina barkeri. Eur J Biochem 265, 325–335.[Abstract/Free Full Text]

Meuer, J., Kuettner, H. C., Zhang, J. K., Hedderich, R. & Metcalf, W. W. (2002). Genetic analysis of the archaeon Methanosarcina barkeri Fusaro reveals a central role for Ech hydrogenase and ferredoxin in methanogenesis and carbon fixation. Proc Natl Acad Sci U S A 99, 5632–5637.[Abstract/Free Full Text]

Pan, G., Menon, A. L. & Adams, M. W. (2003). Characterization of a [2Fe-2S] protein encoded in the iron-hydrogenase operon of Thermotoga maritima. J Biol Inorg Chem 8, 469–474.[Medline]

Peters, J. W., Lanzilotta, W. N., Lemon, B. J. & Seefeldt, L. C. (1998). X-ray crystal structure of the Fe-only hydrogenase (CpI) from Clostridium pasteurianum to 1·8 angstrom resolution. Science 282, 1853–1858.[Abstract/Free Full Text]

Sapra, R., Verhagen, M. & Adams, M. W. W. (2000). Purification and characterization of a membrane-bound hydrogenase from the hyperthermophilic archaeon Pyrococcus furiosus. J Bacteriol 182, 3423–3428.[Abstract/Free Full Text]

Sapra, R., Bagramyan, K. & Adams, M. W. (2003). A simple energy-conserving system: proton reduction coupled to proton translocation. Proc Natl Acad Sci U S A 100, 7545–7550.[Abstract/Free Full Text]

Sauter, M., Böhm, R. & Böck, A. (1992). Mutational analysis of the operon (hyc) determining hydrogenase 3 formation in Escherichia coli. Mol Microbiol 6, 1523–1532.[Medline]

Sawers, G. (1994). The hydrogenases and formate dehydrogenases of Escherichia coli. Antonie Van Leeuwenhoek 66, 57–88.[Medline]

Schmitz, R. A., Daniel, R., Deppenmeier, U. & Gottschalk, G. (2003). The anaerobic way of life. In The Prokaryotes: an Evolving Electronic Resource for the Microbiological Community (http://141.150.157.117 : 8080/prokPUB/index.htm). Edited by M. Dworkin and others. New York: Springer.

Schröder, C., Selig, M. & Schönheit, P. (1994). Glucose fermentation to acetate, CO2 and H2 in the anaerobic hyperthermophilic eubacterium Thermotoga maritima: involvement of the Emden-Meyerhof pathway. Arch Microbiol 161, 460–470.[CrossRef]

Schwarz, E. & Friedrich, B. (2003). The H2-metabolizing prokaryotes. In The Prokaryotes: an Evolving Electronic Resource for the Microbiological Community (http://141.150.157.117 : 8080/prokPUB/index.htm). Edited by M. Dworkin and others. New York: Springer.

Silva, P. J., van den Ban, E. C., Wassink, H., Haaker, H., de Castro, B., Robb, F. T. & Hagen, W. R. (2000). Enzymes of hydrogen metabolism in Pyrococcus furiosus. Eur J Biochem 267, 6541–6551.[Abstract/Free Full Text]

Soboh, B., Linder, D. & Hedderich, R. (2002). Purification and catalytic properties of a CO-oxidizing:H2-evolving enzyme complex from Carboxydothermus hydrogenoformans. Eur J Biochem 269, 5712–5721.[Abstract/Free Full Text]

Svetlichny, V. A., Sokolova, T. G., Gerhardt, M., Ringpfeil, M., Kostrikina, N. A. & Zavarzin, G. A. (1991). Carboxydothermus hydrogenoformans gen. nov., sp. nov., a CO-utilizing thermophilic anaerobic bacterium from hypothermal environments of Kunashir island. Syst Appl Microbiol 14, 254–260.

Tersteegen, A. & Hedderich, R. (1999). Methanobacterium thermoautotrophicum encodes two multi-subunit membrane-bound [NiFe] hydrogenases. Transcription of the operons and sequence analysis of the deduced proteins. Eur J Biochem 264, 930–943.[Abstract/Free Full Text]

Tewes, F. J. & Thauer, R. K. (1980). Regulation of ATP-synthesis in glucose fermenting bacteria involved in interspecies hydrogen transfer. In Anaerobes and Anaerobic Infections, pp. 97–104. Edited by G. Gottschalk. Stuttgart & New York: Gustav Fischer Verlag.

Thauer, R. K., Jungermann, K. & Decker, K. (1977). Energy conservation in chemotrophic anaerobic bacteria. Bacteriol Rev 41, 100–180.[Medline]

van Niel, E. W., Claassen, P. A. & Stams, A. J. (2003). Substrate and product inhibition of hydrogen production by the extreme thermophile, Caldicellulosiruptor saccharolyticus. Biotechnol Bioeng 81, 255–262.[CrossRef][Medline]

Verhagen, M. F., O'Rourke, T. & Adams, M. W. (1999). The hyperthermophilic bacterium, Thermotoga maritima, contains an unusually complex iron-hydrogenase: amino acid sequence analyses versus biochemical characterization. Biochim Biophys Acta 1412, 212–229.[Medline]

Verhagen, M. F., O'Rourke, T. W., Menon, A. L. & Adams, M. W. (2001). Heterologous expression and properties of the gamma-subunit of the Fe-only hydrogenase from Thermotoga maritima. Biochim Biophys Acta 1505, 209–219.[Medline]

Vignais, P. M., Billoud, B. & Meyer, J. (2001). Classification and phylogeny of hydrogenases. FEMS Microbiol Rev 25, 455–501.[CrossRef][Medline]

Wang, J., Xue, Y., Feng, X. & 11 other authors (2004). An analysis of the proteomic profile for Thermoanaerobacter tengcongensis under optimal culture conditions. Proteomics 4, 136–150.[CrossRef][Medline]

Weidner, U., Geier, S., Ptock, A., Friedrich, T., Leif, H. & Weiss, H. (1993). The gene locus of the proton-translocating NADH: ubiquinone oxidoreductase in Escherichia coli. Organization of the 14 genes and relationship between the derived proteins and subunits of mitochondrial complex I. J Mol Biol 233, 109–122.[CrossRef][Medline]

West, A. H. & Stock, A. M. (2001). Histidine kinases and response regulator proteins in two-component signaling systems. Trends Biochem Sci 26, 369–376.[CrossRef][Medline]

Xue, Y., Xu, Y., Liu, Y., Ma, Y. & Zhou, P. (2001). Thermoanaerobacter tengcongensis sp. nov., a novel anaerobic, saccharolytic, thermophilic bacterium isolated from a hot spring in Tengcong, China. Int J Syst Evol Microbiol 51, 1335–1341.[Abstract/Free Full Text]

Yano, T. & Ohnishi, T. (2001). The origin of cluster N2 of the energy-transducing NADH-quinone oxidoreductase: comparisons of phylogenetically related enzymes. J Bioenerg Biomembr 33, 213–222.[CrossRef][Medline]

Received 11 March 2004; revised 14 April 2004; accepted 16 April 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Soboh, B.
Articles by Hedderich, R.
Articles citing this Article
PubMed
PubMed Citation
Articles by Soboh, B.
Articles by Hedderich, R.
Agricola
Articles by Soboh, B.
Articles by Hedderich, R.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.