A novel membrane-bound Ech [NiFe] hydrogenase inDesulfovibrio gigasq

Rute Rodrigues,a Filipa M.A. Valente,a Inees A.C. Pereira,a

Solange Oliveira,a,b and Claudina Rodrigues-Pousadaa,*

a Instituto de Tecnologia Qu�ıımica e Biol�oogica, Universidade Nova de Lisboa, Apartado 127, 2780-901 Oeiras, Portugalb Departamento de Biologia, Universidade de �EEvora, �EEvora, Portugal

Received 3 May 2003

Abstract

In the present study, we report the identification of an operon with six coding regions for a multisubunit membrane-bound [NiFe]

hydrogenase in the genome of Desulfovibrio gigas. Sequence analysis of the deduced polypeptides reveals a high similarity to su-

bunits of proteins belonging to the family of Ech hydrogenases. The operon is organised similarly to the operon coding for the Ech

hydrogenase from Methanosarcina barkeri, suggesting that both encode very similar hydrogenases. Expression of the operon was

detected by Northern blot and RT-PCR analyses, and the presence of the encoded proteins was examined by Western blotting. The

possible role of this hydrogenase is discussed, relating it with a potential function in the H2 cycling as a mechanism for energy

conservation in D. gigas. The present study provides therefore valuable insights into the open question of the energy conserving

mechanism in D. gigas.

� 2003 Elsevier Science (USA). All rights reserved.

Keywords: Desulfovibrio gigas; Ech hydrogenase; H2 cycling; Energy conservation

Desulfovibrio species are sulfate-reducing bacteria

rich in metalloproteins, which act as redox enzymes and

electron carriers in a respiratory metabolism involving

very low redox potentials. Many studies have been

carried out on the metabolism of these bacteria, as well

as on the characterisation of their protein components.

However, several fundamental questions remain to beanswered, namely the process by which a proton motive

force is achieved in these organisms.

Hydrogen cycling was proposed by Odom and Peck

[1], as the general mechanism for energy conservation in

Desulfovibrio. According to this hypothesis, the reducing

power produced in the oxidation of the carbon sub-

strates is used by a cytoplasmic hydrogenase to form

hydrogen, which then diffuses across the membrane to

the periplasm, where it is oxidised by a periplasmic hy-

drogenase. The electrons generated by this process re-

turn to the cytoplasm for the reduction of sulfate, via a

membrane-bound electron transfer chain, thus creating

a proton gradient that is used to activate the ATP syn-

thase. Some arguments have been put forward againstthis mechanism, namely the lack of data supporting the

presence of a cytoplasmic hydrogenase in Desulfovibrio

species, with the exception of the NADP-reducing hy-

drogenase in Desulfovibrio fructosovorans [2]. However,

several Desulfovibrio species, including Desulfovibrio gi-

gas, can grow using hydrogen as the sole energy source

[3], indicating that the oxidation of hydrogen linked to

the reduction of sulfate is an energy conserving process.It is also widely accepted that hydrogen plays a central

role in the metabolism of these bacteria, behaving as an

intermediate in the respiratory electron transfer chain or

as a means to regulate the redox status of the cell.

Hydrogenases are among the most abundant proteins

in these organisms. Several types of hydrogenases

Biochemical and Biophysical Research Communications 306 (2003) 366–375

www.elsevier.com/locate/ybbrc

BBRC

qThis paper is dedicated to the memory of Professor Jean LeGall,

deceased on March 1st of 2003. He played a crucial role in the

development of our research in Desulfovibrio gigas contributing

through his exchange of ideas and fruitful discussions.* Corresponding author. Fax: +351-21-443-36-44.

E-mail address: claudina@itqb.unl.pt (C. Rodrigues-Pousada).

0006-291X/03/$ - see front matter � 2003 Elsevier Science (USA). All rights reserved.

doi:10.1016/S0006-291X(03)00975-6

have been reported in Desulfovibrio ([Fe], [NiFe], and[NiFeSe]) and usually more than one type is present [4].

However, the exact physiological function and condi-

tions of expression of each hydrogenase type have not

been elucidated. So far, only a periplasmic [NiFe]

hydrogenase was detected in D. gigas [5].

A small group of multisubunit membrane-bound

[NiFe] hydrogenases has recently been identified in sev-

eral organisms, including Escherichia coli (hydrogenases3 and 4) [6,7], Methanosarcina barkeri [8], and Methan-

obacterium thermoautotroficum [9]. Sequence analysis of

the proteins belonging to this family reveals that they are

more closely related to subunits of the proton-pumping

NADH:ubiquinone oxireductase (complex I) from vari-

ous organisms than to subunits of other [NiFe] hydrog-

enases [8,10,11]. The only enzyme in this group that was

isolated and characterised is the Ech hydrogenase fromM. barkeri [12,13]. The designation Ech, initially stand-

ing for E. coli hydrogenase 3-type, was recently proposed

to correspond to the more general meaning of energy

converting hydrogenase, due to the probable involve-

ment of these hydrogenases in proton pumping and en-

ergy coupling [11,14]. The physiological electron donor/

acceptor of the M. barkeri Ech hydrogenase was shown

to be ferredoxin [12]. This hydrogenase has two subunitspredicted to be integral membrane-spanning proteins,

namely EchA and EchB, while the other four subunits are

expected to extrude into the cytoplasm. Subunits EchE

and EchC are related to the large and small subunits of

[NiFe] hydrogenases and are thus involved in H2 pro-

duction. Subunits EchA and EchB are supposed to be

involved in the transfer of protons across the membrane.

Subunit EchF, closely related to the TYKY subunit ofbovine complex I, was suggested to be involved in the

electron transfer coupled to proton translocation [13].

Here, we report the characterisation of a group of six

genes, encoding a multisubunit membrane-bound [NiFe]

hydrogenase of the Ech-type in the genome of D. gigas

[15]. The expression of this operon is analysed through

mRNA detection using RT-PCR and Northern blot

analyses. Additionally, Western blot analysis indicatedthe presence of an Ech-like hydrogenase subunit (EchE)

in partially purified extracts of D. gigas membranes.

Materials and methods

Bacterial strains and plasmids. E. coli strains P2 392 and LE 392

were used to screen a D. gigas genomic library, constructed in the

vector k-DashII (Stratagene) and in the purification of the re-

combinant positive phages [16]. Competent E. coli XL2-Blue cells

(Stratagene) were prepared according to standard protocols [17] and

used to transform the DNA fragments subcloned into the plasmid

pZErO-1 (Invitrogen).

Preparation of DNA. Phage DNA was extracted using the Qiagen

Lambda Maxi Kit. Plasmid DNA was extracted using the Plasmid

Purification Kit from SIGMA.

Cloning and sequencing analysis. A DNA probe for the echA gene

was designed based on the previously identified sequence of echA. This

probe was used to screen a library from D. gigas constructed in the

vector k-DashII (Stratagene). The DNA of positive phages was sub-

cloned in pZErO. Sequencing was performed with ABI Prism 373A

Automatic Sequencer (Perkin–Elmer) using ABI Prism DyeDeoxy

Terminator Cycle Sequencing Kit. Protein sequence identification and

analysis were performed with BLAST 2.0 [18], ExPASy server analysis

tools [19], Clustal W [20], and SMART [21].

Molecular phylogeny was reconstructed with Molecular Evolu-

tionary Genetics Analysis (MEGA) software (version 2.1) [22] using

the Neighbour-Joining method [23] with bootstrap support (1000

replicates).

Desulfovibrio growth. Cells were grown anaerobically at 37 �C in a

basal medium modified from [24] containing (per litre): 2 g NH4Cl,

1.65 g MgCl2 � 6H2O, 0.2 g CaCl2, 0.007 g FeCl2 � 4H2O, 10ml of trace

mineral solution, 0.2 g cysteine–HCl, 0.5 g K2HPO4, and 10ml of vi-

tamin solution, yeast extract (0.1 g/L) (resazurin was omitted). Cys-

teine–HCl was added as the reducing agent and sodium bicarbonate

was used to adjust the pH to 7.2–7.6. This basal medium was sup-

plemented with lactate (40mM) as carbon and energy source, and

sulfate (40mM) as terminal electron acceptor. Other electron donor

and acceptors were also used, namely fumarate (40mM), pyruvate

(40mM), formate (40mM), and H2. For growth with hydrogen, the

cultures were supplemented with acetate (20mM) and a gas phase of

H2/CO2 80:20 at 1 bar. The media were inoculated with 10% (v/v) of a

fresh culture in the same medium.

RNA isolation and Northern blot analysis. Total RNA was extracted

from early-log phase D. gigas cells grown anaerobically in lactate/

sulfate medium, according to [16]. RNA was separated on 1.2% aga-

rose in MOPS buffer and 6% formaldehyde (v/v) gel using a 0.24–

9.5 kb RNA Ladder (Gibco-BRL, Life Technologies) as a size marker.

Northern blotting, prehybridisation, and hybridisation were carried

out as previously described [25], using a a32P-labelled probe for D.

gigas echA gene.

RT-PCR analysis. RT-PCR experiments were performed using

Superscript First-Strand Synthesis System for RT-PCR (Gibco-BRL),

as described by the manufacturer. Primers specific for a 531-bp region

in echA were used in PCR assays with cDNA as a template.

The total RNA used was treated with DNase RNase-free (Gibco-

BRL). To rule out the presence of DNA in the RNA preparation, a

control experiment was performed, in which the RNA sample was

incubated under the conditions used for reverse transcriptase, but

without reverse transcriptase, and then used as a template in PCR with

the echA specific primers.

Preparation of cell extracts. The cells were suspended in 50mM

Mops, pH 7, buffer with 2mM DTT and ruptured by passing twice

through a French press. The resulting extract was centrifuged at

10,000g for 15min to remove cell debris and the supernatant was then

centrifuged at 100,000g for 15min.

The soluble extract was obtained in the supernatant fraction. The

membrane pellet obtained was resuspended in 50mM Mops, pH 7,

buffer with 2mM DTT. All procedures were carried out under

anaerobic conditions.

Partial protein purification. All purification steps were performed

aerobically at pH 7.6 and 4 �C. The membrane extract of D. gigas cells

grown in lactate/sulfate medium was solubilised with Triton X-100 in a

final concentration of 2% (w/v). The suspension was stirred for 1 h and

then centrifuged at 100,000g for 1 h. Two extraction steps were per-

formed. The solubilised membrane extract was loaded on a DEAE–

Sepharose fast flow column (5� 40 cm; Pharmacia), equilibrated with

buffer A (30mM Tris–HCl and 0.2% Triton X-100 (w/v)). The column

was washed with buffer A (400ml) and a linear gradient from 0% to

40% buffer B (100mM Tris–HCl, 0.2% Triton X-100, and 1M NaCl)

was applied. The fraction that eluted from DEAE column before the

start of the gradient was pooled, concentrated, and then passed on a

Q-Sepharose high-performance column (Hiload 26/10, flow rate

R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375 367

5ml/min) equilibrated with buffer A. After washing with buffer A, a

stepwise gradient to 40% buffer B was applied.

Western blot analysis. Protein extracts obtained from whole cells as

well as from membranes and from the cytoplasmic fraction were de-

natured in a sample buffer and separated in 15% SDS–PAGE [26]. The

separated polypeptides were then transferred to nitrocellulose mem-

branes by electroblotting, according to the manufacturer�s instructions.Immunodetection was performed using the protocol for the ECL-

Plus Kit (Amersham–Pharmacia) with a 1:5000 dilution of a rabbit

anti-Ech serum obtained for M. barkeri Ech hydrogenase [12] and a

1:100,000 dilution of alkaline phosphatase-conjugated anti-(rabbit

IgG) antibody.

Results and discussion

Organisation of the operon genes

Sequence analysis of the cloned D. gigas DNA frag-

ment revealed a group of six genes, encoding a multi-

subunit membrane-bound [NiFe] hydrogenase of the

Ech-type. The six genes are organised in one operon

(echABCDEF), as shown in the diagram of Fig. 1.The ATG start codons of all the ech genes are im-

mediately preceded by a sequence corresponding to a

hypothetical ribosome-binding site. Distances between

contiguous ech genes vary from 5 to 27 bp. The se-

quences TTGACA and TAACTT are located 103 bp

upstream the start codon of echA, separated by 16 bp,

constituting most probably the promotor sites for re-

gions )35 and )10, respectively. A predicted 62 bp stem-loop structure acting as a hypothetical transcription

terminator signal is detected 182 nucleotides down-

stream the echF stop codon.

The predicted polypeptides revealed a high similarity

to some of the predicted subunits encoded by an operon

from Thermoanaerobacter tengcongensis identified as a

NADH:ubiquinone oxireductase, to all subunits of Ech

hydrogenase from M. barkeri, and to the E. coli Hy-drogenase 3 (Table 1). A detailed analysis of the ho-

mologous subunits from T. tengcongensis [18] revealed

that they are part of an ech-like operon, and were

probably misannotated as subunits from a NADH:ubi-

quinone oxidoreductase. The Ech operons from M.

barkeri and T. tengcongensis share the same organisation

with D. gigas, suggesting that they encode very similar

Ech hydrogenases. This is also the case of an ech-likeoperon identified in the genome of Desulfovibrio vulgaris

Hildenborough (from TIGR, unfinished sequence, Ac-

cession No. NC_002937).

Analysis of the predicted polypeptides

EchA

Desulfovibrio gigas EchA is predicted to be a large

hydrophobic polypeptide of 647 aminoacids (69.04 kDa)

and a pI of 8.9. Its hydropathic profile indicates that it

has 13 transmembrane segments.

EchB

Desulfovibrio gigas EchB is also predicted to be alarge hydrophobic polypeptide of 284 aminoacids

(30.92 kDa) and a pI of 7.83. Its hydropathic profile

indicates that it has six transmembrane segments.

EchC

Desulfovibrio gigas EchC is a polypeptide of 147

aminoacids (15.75 kDa) with a pI of 8.24. This poly-peptide shows a low similarity (20.5%) with the small

subunit of the D. gigas periplasmic [NiFe] hydrogenase.

As in the case of its equivalent subunits from M. barkeri

and Methanosarcina mazei, D. gigas EchC shows only

one of the conserved iron–sulphur-binding motifs typi-

cally present in the small subunit of [NiFe] hydrogenases

(Fig. 2). It also lacks the twin-arginine RRxFxK motif

found in the signal peptide of some [NiFe] hydrogenasessmall subunits that directs translocation of the hydro-

genase across the membrane [11]. This suggests that the

Ech hydrogenase is facing the cytoplasm.

EchD

Desulfovibrio gigas EchD, although a small hydro-

philic polypeptide of 125 aminoacids (14 kDa) with a pI

of 4.75, contains also one transmembrane domain.

EchE

Desulfovibrio gigas EchE is a hydrophilic polypeptide

with a length of 358 aminoacids, with a calculated mo-

lecular mass of 40.24 kDa, and a pI of 6.22. Sequence

analysis indicates a low similarity (24.3%) with the large

subunit of other [NiFe] hydrogenases, namely the

D. gigas periplasmic [NiFe] hydrogenase.

The two conserved motifs in the amino-terminal re-

gion (R–x–C–x(2)–C–x(3)–H) and near the carboxyterminus (D–P–C–x(2)–Cx(2)–H) that contain the cys-

teine residues involved in binding of the Ni–Fe binuclear

site are conserved in the D. gigas EchE subunit, as well

as the substitution of the histidine residue in the car-

boxy-terminal motif by an arginine residue as it is typ-

ical in this group of enzymes (Fig. 2) [8].

EchF

Desulfovibrio gigas EchF is a hydrophilic polypeptideof 105 aminoacids (11.79 kDa) and a pI of 8.23. As its

orthologues (Fig. 2), it has two conserved binding motifs

for bacterial type [4Fe–4S] clusters (C–x(2)–C–x(2)–C–

x(3)–C–[PEG]).Fig. 1. Genetic organisation of the ech operon in D. gigas.

368 R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375

Overall, the six subunits predicted from the D. gigas

ech operon show sequence features similar to the

M. barkeri Ech hydrogenase (Fig. 2).

Phylogenetic analysis based on EchC amino acid se-

quence revealed that the Desulfovibrio EchC subunits areclosely related to the archaea Methanosarcina ones (Fig.

3), which appears to contradict the universal phyloge-

netic tree. This might be due to lateral gene transfer for

the Ech-type hydrogenases during the evolutionary pro-

cess, as already suggested by Vignais et al. [11].

Expression of the operon coding units

Northern blot analysis of D. gigas total RNA ex-tracted from cells grown in lactate/sulfate medium using

an echA homologous probe revealed a smear, deriving

most probably from the degradation of a band of ap-

proximately 6 kb (data not shown). This is the expected

size of the mRNA, predicted from the operon sequence.

A possible cause for this result could be the presence of a

small amount of mRNA, due to a low level of expres-

sion. It is also plausible that this mRNA is highly sus-

ceptible to degradation.

As the results obtained by Northern blot analysis

were not clear and suggested a low concentration of thetranscripts, RT-PCR analysis was performed.

RT-PCR analysis showed that echA is expressed un-

der standard growth conditions (lactate/sulfate me-

dium). A PCR product of expected size of about 530 bp

was obtained. In the control reaction, performed to

confirm the absence of contaminating DNA, no PCR

product was detected (Fig. 4). These data show therefore

that the gene encoding EchA is expressed, suggestingthat the operon is transcribed.

Analysis of the protein

In order to study the expression conditions of the

Ech hydrogenase operon in D. gigas, this bacterium

was grown under different media and its expression

Table 1

Identity and similarity percentages between the predicted Ech polypeptides of D. gigas and their homologues

Organism Protein Similarity Identity

echA D. vulgaris Hildenborough EchA 66.8 50.6

T. tengcogensis NADH:Ubiq. Oxired. Sub. 5 (chain L) 60.2 43.8

M. barkeri EchA 53.1 35.2

M. mazei (GOE1) Ech hydrogenase 52 35.6

echB D. vulgaris Hildenborough EchB 76.1 62.8

T. tengcogensis NADH:Ubiq. Oxired. Sub. 1 (Chain H) 68.5 47.6

M. mazei (GOE1) Ech hydrogenase 63.8 43.1

M. barkeri EchB 63.2 42.7

echC D. vulgaris Hildenborough EchC 66 55.1

T. tengcogensis NADH:Ubiq. Oxired. 20 kDa Sub. 72.4 55.1

M. barkeri EchC 62 42

M. mazei (GOE1) Ech hydrogenase 61 46

echD D. vulgaris Hildenborough EchD 54.9 45.1

T. tengcogensis ECHD 55 36

M. mazei (GOE1) Ech hydrogenase 51 32

M. barkeri EchD 50 32

echE D. vulgaris Hildenborough EchE 83.8 74

T. tengcogensis NADH:Ubiq. Oxired. 49 kDa Sub. 7 71.3 47.6

M. barkeri EchE 68 42

M. mazei (GOE1) Ech hydrogenase 69 43

echF T. tengcogensis NADH:Ubiq. Oxired. 23 kDa Sub. (chain I) 48 38

D. vulgaris Hildenborough EchF 40.4 36

M. barkeri EchF 40 29

M. mazei (GOE1) Ech hydrogenase 41 27

Desulfovibrio gigas predicted subunits of Ech hydrogenase (Accession No. AY282786), D. vulgaris Hildenborough predicted subunits of a

putative Ech hydrogenase (from TIGR, unfinished sequence, Accession No. NC_002937), T. tengcogensis NADH:Ubiq. Oxired. Sub. 5 (chain L)

(Accession No. NP_621823), M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634344), M. barkeri EchA (Accession No. CAA76117), T.

tengcogensis NADH:Ubiq. Oxired. Sub. 1 (Chain H) (Accession No. NP_621824.), M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634345),

M. barkeri EchB (Accession No. CAA76118), T. tengcogensis NADH:Ubiq. Oxired. 20 kDa Sub. (Accession No. NP_621825), M. barkeri EchC

(Accession No. CAA76119), M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634346), T. tengcogensis ECHD (Accession No. NP_621826),

M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634347), M. barkeri EchD (Accession No. CAA76120), T. tengcogensis NADH:Ubiq.

Oxired. 49 kDa Sub. 7 (Accession No. NP_621827), M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634348), M. barkeri EchE (Accession

No. CAA76121), T. tengcogensis NADH:Ubiq. Oxired. 23 kDa Sub. (chain I) (Accession No. NP_621828), M. barkeri EchF (Accession No.

CAA76122), and M. mazei (GOE1) Ech hydrogenase (Accession No. NP_634349).

R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375 369

Fig. 2. Sequence alignment of the Ech subunits from D. gigas with similar proteins. (A) DgEchA (D. gigas EchA), DvEchA (D. vulgaris Hilden-

borough EchA), TtNUO5 (T. tengcogensis NADH:Ubiq. Oxired. Sub. 5 chain L), MmEch (M. mazei (GOE1) Ech hydrogenase), and MbEchA (M.

barkeri EchA); (B) DgEchB (D. gigas EchB), DvEchB (D. vulgaris Hildenborough EchB), TtNUO1 (T. tengcogensis NADH:Ubiq. Oxired. Sub. 1

Chain H), and MmEch (M. mazei (GOE1) Ech hydrogenase), and MbEchB (M. barkeri EchB); (C) DgEchC (D. gigas EchC), and DvEchC (D.

vulgaris Hildenborough EchC), TtNUO20kDA (T. tengcogensis NADH:Ubiq. Oxired. 20 kDa Sub.), MbEchC (M. barkeri EchC), MmEch (M.

mazei (GOE1) Ech hydrogenase); (D) DgEch (D. gigas EchD), TtECHD (T. tengcogensis ECHD), MmEch (M. mazei (GOE1) Ech hydrogenase),

MmEchD (M. barkeri EchD), DvEchD (D. vulgaris Hildenborough EchD); (E) DgEchE (D. gigas EchE), DvEchE (D. vulgaris Hildenborouh EchE),

TtNUO7 (T. tengcogensis NADH:Ubiq. Oxired. 49 kDa Sub. 7), CtHyp (Clostridium thermoacellum ATCC27405 hypthetical protein), MmEch (M.

mazei (GOE1) Ech hydrogenase), and MbEchE (M. barkeri EchE); (F) DgEchF (D. gigas EchF), DvEchF (D. vulgaris Hildenborough EchF),

TtNUO23kDa (T. tengcogensis NADH:Ubiq. Oxired. 23 kDa Sub. chain I), MbEchF (M. barkeri EchF), and MmEch (M. mazei (GOE1) Ech

hydrogenase). The cysteine residues possibly involved in the ligation of an iron–sulphur cluster in EchC are marked with #. The characteristic nickel-

binding motifs found in the amino-terminus (R–x–C–x(2)–C–x(3)–H) and near the carboxy terminus (D–P–C–x(2)–C–x(2)–H) of EchE are marked

with D1 and D2, respectively. The two * marked C�s are nickel ligands. Amino acids thought to be involved in electron transfer in the large subunit

are marked with ); + histidine residue in the carboxy-terminal motif of other large subunits is substituted by an arginine residue [8]. Two iron–

sulphur binding region signatures in EchF are marked with D3 and D4.

370 R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375

was monitored by Western blotting using an anti-se-

rum produced against the M. barkeri Ech hydroge-

nase. Lactate, pyruvate, fumarate, formate, andhydrogen were tested as electron donors using sulfate

as electron acceptor. Fumarate and lactate were also

tested as electron acceptor and energy source, re-

spectively.

Subunits EchA and EchE from M. barkeri are the

ones that give the strongest reaction in immunoblots

[12], so signals corresponding to polypeptides presenting

the expected molecular weight for those subunits in

D. gigas were searched. Under standard growth condi-tions (lactate/sulfate medium), faint signals were ob-

served in partially purified membrane extracts of

D. gigas corresponding to the molecular weight of the

EchE subunit. However, none of the conditions tested

resulted in an evident increase of the expression of this

Ech subunit.

Fig. 2. (continued)

R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375 371

These results suggest that expression of the D. gigas

Ech hydrogenase is low and/or that the anti-Ech serum

produced for M. barkeri Ech is not specific for the D.

gigas Ech subunits.

In order to detect Ech in the membrane fraction of

D. gigas grown in lactate/sulfate, a partial purification

was carried out and fractions were screened using the

same procedure. The anti-Ech serum recognised a pro-tein with the expected molecular weight of subunit EchE

in one of the fractions (Fig. 5). As already observed with

mRNA analysis, the band detected by Western blot

analysis suggests that the protein is present in low

amounts. An attempt to determine the N-terminal se-

quence of this protein was unsuccessful probably due to

the blockage of the N-terminus.

Contrary to other Desulfovibrio species, D. gigas was

thought to contain only a periplasmic [NiFe] hydroge-

nase [5] and therefore, it was proposed that in this

bacterium a bioenergetic mechanism different from the

hydrogen cycling hypothesis could be operating, or a

completely distinct hydrogenase remained to be found

[27]. The present study reveals the presence of a multi-

subunit membrane-bound [NiFe] hydrogenase belong-ing to the Ech hydrogenase family in D. gigas. Sequence

analysis of the predicted subunits suggests that this hy-

drogenase is bound to the cytoplasmic side of the

membrane, in a situation ideal to constitute the missing

link in the hydrogen cycling mechanism that is respon-

sible for hydrogen formation in the cytoplasm. D. gigas

Ech hydrogenase could reduce protons coming from the

Fig. 2. (continued)

372 R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375

oxidation of organic compounds to H2, which could

then pass through the membrane to the periplasm, being

oxidised by the periplasmic hydrogenase, thus creating a

proton gradient and leading to energy conservation.

It was suggested in D. vulgaris Hildenborough that

the putative hydrogenase encoded by the ech-like operon

is involved in energy conservation, functioning as a

hydrogen-producing enzyme in lactate metabolism, withferredoxin as a redox partner [28]. This may also be the

case in D. gigas, but in order to clarify this mechanism,

further genetic, biochemical, and structural studies have

to be performed.

For D. fructosovorans a mutant strain was con-

structed devoid of the three characterised hydrogenases

(the periplasmic [Fe] and [NiFe] hydrogenases and the

cytoplasmic NADP-reducing hydrogenase) [29,30]. This

triple mutant showed no significant growth differences

relative to the wild-type strain and could still grow with

H2 as sole energy source, pointing to the existence of a

fourth uncharacterised hydrogenase in D. fructosovo-

rans, which may be an Ech hydrogenase similar to the

one found in D. gigas and D. vulgaris Hildenborough.Interestingly, genes encoding homologues of the Ech

hydrogenase subunits could not be found in the genome

of D. desulfuricans G20, which reveals diversity within

the genus Desulfovibrio.

Fig. 3. Phylogenetic tree of D. gigas EchC (Ech hydrogenase small subunit) and homologous sequences. The tree was obtained with Molecular

Evolutionary Genetics Analysis (MEGA) software (version 2.1) [22] using the Neighbour-Joining method [23] with bootstrap support (1000 repli-

cates). Archaeal subunits are shaded in light grey and bacterial subunits in darker grey.

Fig. 4. RT-PCR analysis of echA. Lane 1, molecular marker (1 kb Plus,

Invitrogen); lane 2, RT-PCR; and lane 3, negative control.

Fig. 5. Western blot detection of Ech-like hydrogenase in a partially

purified fraction from D. gigas membranes. Lane 1, purified Ech hy-

drogenase from M. barkeri and lane 2, partially purified fraction from

D. gigas membranes.

R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375 373

Various members of the Ech family of hydrogenaseshave been proposed to function in energy-conserving

processes [11], and recently Ech from M. barkeri was

shown to play a central and diverse role in its metab-

olism, including hydrogen formation from reduced fer-

redoxin with energy conservation, as well as reduction

of ferredoxin by hydrogen via reverse electron transport

[14]. The newly found D. gigas ech operon has the same

organisation as the homologous operons in M. barkeri

and D. vulgaris Hildenborough, suggesting that they

encode very similar hydrogenases that possibly play

similar roles. Due to the sequence similarities to Com-

plex I, it is also possible that Ech may be functioning as

a proton pump in D. gigas.

Ech hydrogenase operon from M. barkeri is consti-

tutively and abundantly expressed [8], whereas its or-

thologue in D. gigas is expressed in low levels, at leastunder standard growth conditions. This suggests that

Ech hydrogenase from D. gigas may have a less prom-

inent role in its metabolism than its counterpart in

M. barkeri.

In summary, this work presents evidence for the

presence of a new hydrogenase in a sulfate-reducing

bacterium, adding new insights into the possible meta-

bolic pathways involved in the still poorly understoodprocess of the anaerobic respiration of sulfate.

Acknowledgments

This work was supported by PRAXIS XXI (11074/98, to C.R.P.,

SFRH/BD 5219/2001 fellowship to R.R.) and POCTI/FEDER FCT

Grant (POCTI/36562/ESP/2000 to I.A.C.P. and SFRH/BD/9187/2002

fellowship to F.M.A.V.). We thank Dr. Reiner Hedderich, Max-

Planck-Institut fuer Terrestrische Mikrobiologie for kindly providing

us the rabbit anti-Ech serum [12] and purified Ech from M. barkeri.

STAB Gen�oomica is also acknowledged for running the sequencing

reactions.

References

[1] J. Odom, Peck, Hydrogen cycling as a general mechanism for

energy coupling in the sulfate-reducing bacteria, Desulfovibrio sp,

FEMS Microbiol. Lett. 12 (1984) 47–50.

[2] S. Malki, I. Saimmaime, G. De Luca, M. Rousset, Z. Dermoun,

J.P. Belaich, Characterization of an operon encoding an NADP-

reducing hydrogenase in Desulfovibrio fructosovorans, J. Bacteriol.

177 (10) (1995) 2628–2636.

[3] A. Brandis, R.K. Thauer, Growth of Desulfovibrio species on

hydrogen and sulfate as sole energy source, J. Gen. Microbiol. 126

(1981) 249–252.

[4] G. Fauque, H.D. Peck Jr., J.J. Moura, B.H. Huynh, Y. Berlier,

D.V. DerVartanian, M. Teixeira, A.E. Przybyla, P.A. Lespinat, I.

Moura, The three classes of hydrogenases from sulfate-reducing

bacteria of the genus Desulfovibrio, FEMS Microbiol. Rev. 4

(1988) 299–344.

[5] G. Voordouw, V. Niviere, F.G. Ferris, P.M. Fedorak, D.W.S.

Westlake, Distribution of hydrogenase genes in Desulfovibrio spp.,

their use in identification of species from the oil-field environment,

Appl. Environ. Microbiol. 56 (1990) 3748–3754.

[6] R. B€oohm, M. Sauter, A. B€oock, Nucleotide sequence and expres-

sion of an operon in Escherichia coli coding for formate

hydrogenlyase components, Mol. Microbiol. 4 (1990) 231–243.

[7] S.C. Andrews, B.C. Berks, J. McClay, A. Ambler, M.A. Quail, P.

Golby, J.R. Guest, A 12-cistron Escherichia coli operon (hyf)

encoding a putative proton-translocating formate hydrogenlyase

system, Microbiology 143 (1997) 3633–3647.

[8] J. Kunkel, J. Vorholt, R. Thauer, R. Hedderich, An Escherichia

coli hydrogenase-3-type hydrogenase in methanogenic archaea,

Eur. J. Biochem. 252 (1998) 467–476.

[9] A. Tergsteen, R. Hedderich, Methanobacterium thermoautotrofi-

cum encodes two multisubunit membrane-bound [NiFe] hydrog-

enases. Transcription of the operons and sequence analysis of the

deduced proteins, Eur. J. Biochem. 264 (1999) 930–943.

[10] T. Friedrich, D. Scheide, The respiratory complex I of bacteria,

archaea and eukarya and its module common with membrane-

bound multisubunit hydrogenases, FEBS Lett. 479 (2000) 1–5.

[11] P.M. Vignais, B. Billoud, J. Meyer, Classification and phylogeny

of hydrogenases, FEMS Microbiol. Rev. 25 (2001) 455–501.

[12] J. Meuer, Bartoschek, J. Koch, A. Kunkel, R. Hedderich,

Purification and catalytic properties of Ech hydrogenase from

Methanosarcina barkeri, Eur. J. Biochem. 265 (1999) 325–335.

[13] S. Kurkin, J. Meuer, J. Koch, R. Hedderich, S.P.J. Albracht, The

membrane-bound [NiFe] hydrogenase (Ech) from Methanosarcina

barkeri: unusual properties of the iron–sulphur clusters, Eur. J.

Biochem. 269 (2002) 6101–6111.

[14] J. Meuer, H. Kuettner, J. Zhang, R. Hedderich, W. Metcalf,

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. USA

99 (2002) 5632–5637.

[15] R. Rodrigues, S. Oliveira, C. Rodrigues-Pousada, One hypothet-

ical multisubunit membrane-bound [NiFe] hydrogenase encoded

in Desulfovibrio gigas genome, Eur. J. Biochem. 268 (1) (2001) 49.

[16] F.M. Ausubel, R. Brent, R. Kingston, D. More, Seidman, J.

Smith, Struhl, Current Protocols in Molecular Biology, Greene

Publishing Associates and Wiley, Interscience, NY, 1995.

[17] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning. A

Laboratory Manual, second ed., Cold Spring Harbor, New York,

1989.

[18] Altschul, F. Stephen, T.L. Madden, A.A. Schaffer, J. Zhang, Z.

Zhang, W. Miller, D.J. Lipman, Gapped BLAST and PSI-

BLAST: a new generation of protein database search programs,

Nucleic Acids Res. 25 (1997) 3389–3402.

[19] R.D. Appel, A. Bairoch, D.F. Hochstrasser, A new generation of

information retrieval tools for biologists: the example of the

ExPASy WWW server, Trends Biochem. Sci. 19 (1994) 258–260.

[20] D. Higgins, J. Thompson, T. Gibson, J.D. Thompson, D.G.

Higgins, T.J. Gibson, Clustal W: improving the sensitivity of

progressive multiple sequence alignment through sequence weight-

ing, position specific gap penalties and weight matrix, Nucleic

Acids Res. 22 (1994) 4673–4680.

[21] I. Letunic, L. Goodstadt, N.J. Dickens, T. Doerks, J. Schultz, R.

Mott, F. Ciccarelli, R.R. Copley, C.P. Ponting, P. Bork, Recent

improvements to the SMART domain-based sequence annotation

resource, Nucleic Acids Res. 30 (1) (2002) 242–244.

[22] S. Kumar, K. Tamura, I.B. Jakobsen, M. Nei, MEGA2: molec-

ular evolutionary genetics analysis software, Bioinformatics 17

(2001) 1244–1245.

[23] N. Saitou, M. Nei, The Neighbor-Joining method: a new method

for constructing phylogenetic trees, Mol. Biol. Evol. 4 (1987) 406–

425.

[24] R. Devereux, C.L. Doyle, S. Orkland, D.A. Stahl, J. LeGall, W.B.

Whitman, Diversity and origin of Desulfovibrio species: phyloge-

netic definition of a family, J. Bacteriol. 172 (1990) 3609–3619.

374 R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375

[25] G. Silva, S. Oliveira, J. LeGall, A. Xavier, C. Rodrigues-Pousada,

Analysis of the Desulfovibrio gigas transcriptional unit containing

rubredoxin (Rd) and rubredoxin-oxygen oxireductase (roo) genes

and upstream ORFs, Biochem. Biophys. Res. Commun. 280

(2001) 491–502.

[26] U.K. Laemmli, Cleavage of structural proteins during the assem-

bly of the head of bacteriophage T4, Nature 227 (1970) 680–685.

[27] I.A.C. Pereira, C.V. Rom~aao, A.V. Xavier, J. LeGall, M. Teixeira,

Electron transfer between hydrogenases and mono- and multi-

heme cytochromes in Desulfovibrio ssp, JBIC 3 (1998) 494–

498.

[28] B.K. Pohorelic, J.K. Voordouw, E. Lojou, A. Dolla, J. Harder, G.

Voordouw, Effects of deletion of genes encoding Fe-only hydrog-

enase of Desulfovibrio vulgaris Hildenborough on hydrogen and

lactate metabolism, J. Bacteriol. 184 (3) (2002) 679–686.

[29] L. Casalot, O. Valette, G. De Luca, Z. Dermoun, M. Rousset, P.

de Philip, Construction and physiological studies of hydrogenase

depleted mutants of Desulfovibrio fructosovorans, FEMS Micro-

biol. Lett. 214 (1) (2002) 107–112.

[30] L. Casalot, G. De Luca, Z. Dermoun, M. Rousset, P. de Philip,

Evidence for a fourth hydrogenase in Desulfovibrio fructosovorans,

J. Bacteriol. 184 (3) (2002) 853–856.

R. Rodrigues et al. / Biochemical and Biophysical Research Communications 306 (2003) 366–375 375