Respiratory nitrate reductase from haloarchaeon Haloferax mediterranei: biochemical and genetic analysis

Biochimica et Biophysica Acta 1674 (2004) 50 – 59 www.bba-direct.com

Respiratory nitrate reductase from haloarchaeon Haloferax mediterranei: biochemical and genetic analysis B. Lledo´, R.M. Martı´nez-Espinosa, F.C. Marhuenda-Egea, M.J. Bonete * Divisio´n de Bioquı´mica y Biologı´a Molecular, Facultad de Ciencias, Universidad de Alicante, Ap. 99, E-03080, Alicante, Spain Received 6 February 2004; received in revised form 19 May 2004; accepted 19 May 2004 Available online 17 June 2004

Abstract The Haloferax mediterranei nar operon has been sequenced and its regulation has been characterized at transcriptional level. The nar operon encodes seven open reading frames(ORFs) (ORF1 narB, narC, ORF4, narG, narH, ORF7 and narJ). ORF1, ORF4 and ORF7 are open reading frames with no assigned function, however the rest of them encoded different proteins. narB codes for a 219-amino-acid-residue iron Rieske protein. narC encodes a protein of 486 amino acid residues identified by databases searches as cytochrome-b (narC). The narG gene encodes a protein with 983 amino acid residues and is identified as a respiratory nitrate reductase catalytic subunit (narG). NarH protein has been identified as an electron transfer respiratory nitrate reductase subunit (narH). The last ORF encodes a chaperonin-like protein (narJ) of 242 amino acid residues. The respiratory nitrate reductase was purified 21-fold from H. mediterranei membranes. Based on SDS-PAGE and gel-filtration chromatography under native conditions, the enzyme complex consists of two subunits of 112 and 61 kDa. The optimum temperature for activity was 70 jC at 3.4 M NaCl and the stability did not show a direct dependence on salt concentration. Respiratory nitrate reductase showed maximum activity at pH 7.9 and pH 8.2 when assays were carried out at 40 and 60 jC, respectively. The absorption spectrum indicated that Nar contains Fe – S clusters. Reverse transcriptase (RT-PCR) shows that regulation of nar genes occurs at transcriptional level induced by oxygen-limiting conditions and the presence of nitrate. D 2004 Elsevier B.V. All rights reserved. Keywords: Respiratory nitrate pathway; Cluster; Halophile; Archaea

1. Introduction Prokaryotic nitrate reduction can serve a number of physiological roles and can be catalysed by distinct nitrate reductases. Three classes of nitrate reductases can be identified in prokaryotes: one class of assimilatory nitrate reductases (Nas) and two non-assimilatory nitrate reductases. These reductases differ in their cellular location and function: a respiratory membrane-bound (Nar) enzyme plays a key role in the generation of metabolic energy by using nitrate as a terminal electron acceptor (nitrate respiration) and a periplasmic nitrate reductase (Nap), which participates in the dissipation of excess of reducing power for redox balancing (nitrate dissimilation) [1]. Abbreviations: Nar, respiratory nitrate reductase; DT, dithionite; MV, methyl viologen * Corresponding author. Tel.: +34-965-903-880; fax: +34-965-903880. E-mail address: [email protected] (M.J. Bonete). 0304-4165/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.bbagen.2004.05.007

Haloferax mediterranei is a halophilic archaeon that can use NO3
in an assimilatory process under aerobic conditions via Nas [2] and in a respiratory process via Nar [3], where NO3
is the electron acceptor when oxygen conditions are limited. The membrane-bound nitrate reductases are composed of three subunits: (i) a catalytic a subunit containing a molybdopterin cofactor, (ii) a soluble h subunit containing four [4Fe –4S] and (iii) the g subunit containing b-type haems [4]. The a and h subunits constitute the cytoplasmic domain and the g subunit is the membrane domain required for the attachment of the a and h subunits to the cytoplasmic side of the inner membrane. The g subunit accepts electrons from the quinol pool and transfers them via its b-haems to the h subunit [5]. The structural gene encoding the respiratory nitrate reductase was first sequenced from E. coli, Bacillus subtilis and Pseudomonas fluorescens [6– 8]. The nar genes are organized in an operon with another gene encoding a chaperone-like component required for the maturation of the ah complex and assembly of the ahg [9]. The operon is

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

often linked with at least one narK gene encoding proteins belonging to the transmembrane transporters [10]. In bacteria, the expression of the nar operon is induced by anaerobiosis and in the presence of nitrate or nitrite [11]. The transcriptional regulation protein Fnr is the anaerobic activator in E. coli and B. subtilis. Many bacteria, like Pseudomonas species, have been found to harbour an Fnr-like protein [12]. In addition, nar operon expression is mediated by the narXL genes encoding a two component regulatory system [13]. The narX gene encodes a sensor-transmitter which detects the presence of NO3
and NO2
in the periplasm. The narL gene encodes a response regulator which binds to the DNA. Previous studies of denitrifiers of halophilic archaea have been reported in the genus Haloferax [14 – 16] and Haloarcula [17]. In this study the complete H. mediterranei nar operon has been sequenced and its regulation has been characterized at the transcriptional level, thus allowing further questions about the transcriptional regulation mechanism. In Archaea, transcriptional regulation is carried out by a bacteria-like mechanism in contrast with transcription, which occurs in an eukaryotic-like fashion. The respiratory nitrate reductase has been purified and characterized reporting similarities with the bacterial counterparts at both the biochemical and sequence levels.

2. Materials and methods 2.1. Strains, media and growth conditions The strains used were: H. mediterranei (ATCC 33500T), Haloferax volcanii (DSM 3757), E. coli strain KW251 as the host for E EMBL3 vector (Promega). H. mediterranei cultures were grown in a 25% (wt/vol) mixture of salts and 0.5% yeast extract, as previously described by Rodrı´guez-Valera et al. [18]. The pH value 5 of the culture medium was adjusted to 7.3. H. mediterranei was grown aerobically at 37 jC in 1-l batch cultures, in 2l Erlenmeyer flasks on a rotary shaker at 200 rpm. Growth was monitored by measuring the optical density at 540 nm. Cells were grown under these conditions until just before the stationary phase of growth (OD540 = 1.8). The medium was then supplemented with 100 mM of nitrate and oxygen was eliminated by adding N2. After 11 days of incubation without shaking, cells were harvested by centrifugation at 30,000  g for 30 min at 4 jC in a Beckman J2-21 centrifuge. Growth conditions for H. volcanii were described previously [19]. 2.2. Preparation of H. volcanii probe H. volcanii genomic DNA was extracted from a stationary phase culture as described by Dyall-Smith et al. [19]. The nar probe was prepared from a PCR product obtained from H. volcanii genomic DNA with the primer

51

pair 5V-AAGTGTATCGGCTGTCACAC-3V (NARfor) and 5V-ACAGAAGATACACTTCTCCG-3V (NARrev), using optimised PCR conditions. One product of expected size (540 bp) was detected by electrophoresis and labelled with digoxigenin-dUTP, using a Non-Radioactive DNA labelling and Detection Kit (Roche Molecular Biochemicals). It was used as a probe in E plaque hybridisation experiments [20]. 2.3. H. mediterranei genomic library screening and DNA sequencing H. mediterranei genomic DNA was partially digested with SauIIIA, and size-fractioned by agarose gel electrophoresis. Using the Genomic Cloning Technical Manual kit (Promega), restriction fragments from 15 to 20 kb were ligated into the lambda vector. Ligated DNA was packaged and phages were plated on E. coli KW251. 6 For library screening, E plaques were blotted onto Hybond-N+ membranes (Amersham Pharmacia Biotech) and hybridised with the probe prepared above. Positive lambda phages were isolated after double screening and subsequent DNA extraction was carried out using the Lambda Maxi Kit (Qiagen). Nucleotide sequence was performed by the dideoxy chain termination method using 3100 DNA sequencer (Applied Biosystems). Synthesized specific oligonucleotides were used as primers, developing a primerwalking strategy. Analysis of the DNA and amino acid sequences was performed using the GCG package (Wisconsin Package Version 10.1, Genetics Computer Group (GCG), Madison, WI, USA). Multiple alignments were constructed using the CLUSTAL W program [21]. Searches in the SWISSPROT and TrEMBL databases for sequence similarities were carried out with the programs BLAST [22] and FASTA [23]. 2.4. Purification of respiratory nitrate reductase All the purification steps were carried out at 25 jC. Step 1: Preparation of crude extract. The freshly harvested cells were resuspended in 10 mM Tris – HCl buffer pH 8.0, containing 2.0 M NaCl. The cells were disrupted by sonication and the suspension was centrifuged at 106,000  g for 60 min at 4 jC. The precipitate (7 g wet weight) thus obtained was used as the membrane fraction for the purification of respiratory nitrate reductase. The membrane fraction was resuspended in 20 ml of 10 mM Tris – HCl buffer (pH 8.0) containing 2.0 M NaCl and Triton X-100 (Sigma) (20% w/v). The resulting suspension was gently stirred overnight and then centrifuged at 30,000  g for 30 min at 4 jC. The supernatant obtained was dialyzed against 10 mM Tris –HCl (pH 8.0) for 12 h. Step 2: DEAE-sepharose CL-6B chromatography. The supernatant from step 1 (30 ml) was applied to a DEAESepharose CL-6B column (2.5  4 cm) which had previ-

52

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

ously been equilibrated with 10 mM Tris – HCl (pH 8.0). The column was washed with 100 ml of 10 mM Tris –HCl (pH 8.0) containing 200 mM NaCl. Elution was carried out with a increasing linear gradient (200 ml) of 200 –500 mM NaCl in 10 mM Tris – HCl buffer pH 8.0, at a flow rate of 30 ml/h. Fractions containing Nar activity were pooled and dialyzed against 10 mM Tris – HCl buffer, pH 8.0. The resulting solution was applied to a small size DEAESepharose CL-6B column. Step 3: DEAE-sepharose CL-6B chromatography. A DEAE-Sepharose CL-6B column (2.5  1.5 cm) was equilibrated with 10 mM Tris –HCl buffer, pH 8.0. The column was washed using the same buffer, containing 200 mM NaCl, at a flow rate of 20 ml/h. The enzyme was eluted using an increasing linear gradient (200 ml) of 200 – 500 mM NaCl in 10 mM Tris –HCl buffer pH 8.0. Step 4: Sephacryl S-300 chromatography. Fractions containing Nar activity were loaded on a Sephacryl S-300 column (Pharmacia HiPrep 16/60), previously equilibrated with 10 mM Tris –HCl buffer (pH 8.0) containing 200 mM NaCl. The column was washed with the same buffer with a flow rate of 30 ml/h. 2.5. Characterization of respiratory nitrate reductase The protein content was determined by the Bradford method, with bovine serum albumin (fraction V) as a standard. Nar activity was measured by colorimetric determination of nitrite as previously described [2,24]. The assay mixture contained, in a final volume of 250 Al, 100 mM Tris – HCl pH 8, 3.6 M NaCl, 4 mM methyl viologen (MV), 35 mM KNO3, 17 mM Na2S2O4 (freshly prepared in 0.1 M NaHCO3) and 50 Al of enzyme preparations. The assay was developed at 40 jC for 20 min. Nar specific activity is expressed as micromoles of NO2
appearing per minute per milligram of protein. All the assays were carried out in triplicate and against a control assay without enzyme. The kinetic results were processed using the Michaelis– Menten equation. The values of Vmax and Km were determined by nonlinear regression analysis of the corresponding Michaelis –Menten curves (v vs. [NO3
]) using the algorithm of Marquartd – Levenberg with the SigmaPlot program (Jandel Scientific, v. 1.02). The molecular mass of the native enzyme was determined using a Sephacryl S-300 gel-filtration column equilibrated with 10 mM Tris – HCl (pH 8.0) and 200 mM NaCl. The enzyme from step 3 (2 ml) was loaded onto the column and the elution volume of Nar was determined from enzyme assays of eluted fractions. Subsequently, the Mr of Nar subunits were estimated by SDS-PAGE using molecular weight markers from PROMEGA. Native gel electrophoresis was carried out in 12% (w/v) polyacrylamide gels and the Nar activity in the gel was detected as described by Vega and Kamin [25].

2.6. Reverse transcriptase-PCR (RT-PCR) Total RNA from H. mediterranei grown in a complex nitrate-supplemented medium was harvested from cultures grown under both pre-anaerobic and postanaerobic conditions. In order to analyse the presence of nar operon mRNA, two oligonucleotides were designed at positions 6151 bp (RTfor 5V-AAGTGTATCGGCTGTCACAC-3V) and 6705 bp (RTrev 5V-TAACAGAAGATGCACTTCTCG-3V) amplifying the respiratory nitrate reductase electron transfer subunit (narH). Control reactions were performed amplifying a constitutive control (‘‘housekeeping’’) gene rRNA 16S expressed under all conditions.The presence of expected band indicates the integrity of the RNA and the lack of its degradation. cDNA synthesis was performed in a total volume of 20 Al with 1.5 Ag of RNA, 30 pmol of reverse oligonucleotide, 20 mM deoxynucleotide triphosphates, and 20 U of AMV Reverse Transcriptase (Sigma) in the commercial buffer and in the presence of 20 U Rnase inhibitor. The reaction was carried out for 1 h at 50 jC. Controls were performed in the absence of AMV retrotranscriptase. PCR (total volume of 50 Al) was carried out using 5-Al cDNA aliquots as template for the reaction, with both primers (0.5 AM each), 0.2 mM dNTPs and 1.5 U of Pfu DNA polymerase (Fermentas) in the commercial buffer. Initial denaturation was for 5 min at 95 jC, followed by 35 cycles of synthesis comprising 1 min of denaturation at 95 jC, 1 min of annealing at 50 jC, and elongation at 72 jC for 2 min. The amplification was ended with 10-min elongation at 72 jC. Ten-microliter aliquots of the various PCRs were analysed by agarose gel electrophoresis. 2.7. Nucleotide sequence accession number The EMBL accession numbers for the nucleotide sequence determined in this work are AJ621876, AJ621877, AJ621878, AJ621879, AJ621880, AJ621881, AJ621882 and AJ621883.

3. Results and discussion 3.1. Nucleotide sequence determination of the nitrate respiration genes cluster of H. mediterranei and gene organization The DNA fragment employed as a probe was sequenced to confirm that it was a fragment of the respiratory nitrate reductase (Nar). This probe was then used to determine the rest of the genes. Of the positive recombinant E phage clone, 9400 bp were sequenced. This region corresponds to seven open reading frames (ORF1, narB, narC, ORF4, narG, narH, ORF7 and narJ) of 654, 657, 1458, 280, 2949, 1056, 825 and 726 bp, respectively. ORF1, ORF4 and ORF7 are open reading frames (ORFs) without assigned function; however, the

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

53

Fig. 1. Schematic organisation of the nitrate respiration gene clusters of H. mediterranei. Vertical arrows indicate location and direction of putative transcription initiation element.

other ORFs encode different proteins. narB codes for a 219amino-acid-residue iron Rieske protein. The next ORF (with assigned function) encodes a protein of 486 amino acid residues identified by databases searches as cytochrome-b (narC), a fifth one of 983 amino acid residues has been identified as a respiratory nitrate reductase catalytic subunit (narG), the next one as an electron transfer respiratory nitrate reductase subunit (narH). The last ORF encodes a chaperonin-like protein (narJ) of 242 amino acid residues (Fig. 1). The molecular masses estimated from each amino acid sequences were 23, 53, 111, 41 and 26 kDa, respectively. Sequenced were 1.5 kb upstream and 2 kb downstream of nar operon, and other ORFs related to nitrate metabolism were not found upstream or downstream from ORF1 and narJ. Transcription is typically initiated in halophilic archaea 24 bp downstream from a TATAbox sequence with the consensus 5V-(T/C)T (T/A)(T/A)AN-3V [26]. At
33 bp from transcription initiation BRE sequence is located, the core of the consensus sequence, CGAYG, is the more conserved in halophilic BRE sequence [27]. One putative TATAbox was found located 167 bp upstream from the translation start codon for ORF1. Upstream of this consensus TATAbox, BRE sequences were located in adequate positions (Fig. 2). At least a four-base complementary sequence to the 3V end of 16S rRNA (Shine –Dalagarno) [28] is usually preceded at a 3 –10-nucleotide distance. All ORFs identified were preceded by S-D sequence. An interesting feature of the promoter region is the presence of sets of palindromic sequence, with a length position similar to other sequence involved in transcription regulation in Archaea [29]. The fact that all ORFs are overlapped, and the absence of another TATA-box like sequence upstream of them, suggests that the seven ORFs are co-transcribed in a polycistronic mRNA containing genes required for nitrate respiration. For this reason, all the ORFs are organized in an operon called narBCGHJ. Genes encoding the membrane-bound nitrate reductase are distributed among taxonomically diverse bacteria (from Proteobacteria to Gram-

positive bacteria) and archaea. The organization of the nar operon is conserved between bacteria. However, in the archaeal organisms the organization is not conserved. In the archaeon Pyrobaculum aerophilum, the narJ gene is located upstream from the narG gene and is transcribed in the opposite direction. In contrast, narJ is separated by approximately 2000 bp from narH in Aeropyrum pernix.In our studies a different organization could be identified and conserved between halophilic denitrifiers (Haloarcula), in which narJ is the last gene transcribed in the operon. NarK homologues have been found in the vicinity of the nar operon in all the bacterial genomes; however, in archaea this has not been found. The conservation of operon organization can provide valuable evolutionary and functional clues. The conservation of gene order is emerging as an informative property of the genome [3]. 3.2. Amino acid sequence comparisons During nitrate respiration, different donors transfer their electrons to naphtoquinones [30], which subsequently donate them to the acceptor site of nitrate reductase. Database searches identified narC as a cytochrome-b, being employed in nitrate respiration, and called g subunit of respiratory nitrate reductase. Sequence comparisons of narC indicate that H. mediterranei cytochrome-b is most closely related with cytochrome-b from halophilic organism, Haloarcula marismortui (70% identity). But high scores have also been obtained with Deinococcus radiodurans (54% similarity), and Streptomyces coelicolor (51% similarity). Cytochromeb is an integral membrane protein; from the sequence analysis we have identified eight transmembrane segments. Cytochrome-b binds noncovalently haems groups and conserved histidine residues are postulated to be the ligands of the iron atoms of these haems groups. A consensus pattern ([DENQ]-X-X-X-G-[FYWMQ]-X-[LIVMF]-R-X-X-H) (as PROSITE nomenclature) was identified in narC between amino acid residues 101 and 113 and implies a haem binding region. The best-conserved region in the sequence

Fig. 2. Promoter structure of the H. mediterranei nar operon. Double underline, putative TATAbox; grey letters, BRE sequence; single underline, putative ribosome binding site for ORF1; boxed letters, ATG codon; bold underline letter, putative transcriptional start site; ! p , palindromic sequences.

54

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

of cytochrome-b includes an invariant P-[DE]W-[FY][LFY] that lies in the loop that separates the fifth and the sixth transmembrane segments. It is important for electron transfer at the ubiquinone redox site located on the outer side of the membrane. We have identified this motif in H. mediterranei narC at amino acid residue 333. We have shown that narC constitutes a membrane protein, a fact that agrees with the predictions for transmembrane helixes. In addition, the presence of a putative signal peptide with a potential processing site (LA-Q) identical to that of the cytochrome c552 from Thermus thermophilus confirms that this is a membrane protein. Moreover, NarC should be required for the attachment of the a subunit to the membrane as in the extremophile T. thermophilus [31]. The reduction of nitrate could be used to substitute oxygen in anaerobic conditions by allowing cell respiration by respiratory nitrate reductase. It contains a catalytic subunit (a subunit) and electron transfer subunit (h subunit). Comparisons with the database revealed narG as the nitrate reductase a catalytic subunit. The ‘‘best 13 fits’’ of H. mediterranei nitrate reductase gene are with archaeal homologues, the best with a halophile one H. marismortui (89% similarity, 80% identity) and with A. pernix (32% identity); moreover, good scores were obtained with bacterial homo-

logues like Salmonella typhi (33% identity). Sequence alignments show amino acid residues of narG involved in the possible Fe – S cluster binding motif (Fig. 3A). In the Nterminal of the H. mediterranei narG a possible [4Fe – 4S] cluster binding motif [CH]-X-X-X-C-X-X-X-C-//C was identified. In addition, in the N-terminal region, we identified accurately at position 46 the typical signal peptide containing a twin arginine motif and the following hydrophobic stretch (-GLGVASLLGI-). Similar signal peptides of several other related enzymes are known to determine the localization of the enzymes relative to the cytoplasmic membrane and the catalytic site on the periplasmic side. However H. mediterranei does not have periplasmic side [32], indicating that its localization could be in the outer membrane. NarG could be classified into the molybdoenzyme family containing a motif implied in bis-MGD cofactor binding. In molybdoenzymes the molybdenum can be coordinated by – S, – O or – Se provided by cysteine, serine or selenocysteine residues in the polypeptide chain and a variable number of oxo ( –O), hydroxyl ( –OH) or water groups. Between Nars there are no conserved cysteines in the MGD binding subunit, but there are a number of conserved serine residues, which raises the possibility that Nar has a Mo-O-Ser ligand. These serine residues were

Fig. 3. Sequence alignments of the H. mediterranei narG (a, b) and narH (c). Asterisks indicate positions of identical amino acids; colons indicate positions with conserved replacements, and periods indicate semiconserved positions. (a) ClustalW alignment of the predicted H. mediterranei narG with nitrate reductase a subunit from H. marismortui (HmarismNarG), A. pernix (AeroperNarG), and S. typhi (SalmotiNarG). Boxed amino acid residues indicated residues implicated in 4Fe – 4S cluster formation. (b) Black boxed amino acid residues indicated residues implicated in Mo cofactor binding. (c) ClustalW alignment of the predicted H. mediterranei narG with nitrate reductase h subunit from H. marismortui (HmarismNarH), A. pernix (AeroperNarH), and S. coelicolor (StrpcoeNarH). Boxed amino acid residues showed residues implicated in Fe – S cluster formation. Black boxed indicates Lys residue that converts a 3Fe – 4S in homologues NarH in a 2Fe – 4S cluster in NarH H. mediterranei by the loss of a conserved Cys residue.

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

55

Table 1 Purification of Nar from H. mediterranei

Crude extract DEAE-Sepharose CL-6B (2.5  4 cm) DEAE-Sepharose CL-6B (2.5  1.5 cm) Sephacryl S-300

Volume (ml)

Total protein (mg)

Total activity (U)

Specific activity (U/mg protein)

30 21

9.00 6.00

7.37 6.64

0.082 0.11

1.0 1.3

100 90

14 1

0.65 0.35

0.90 0.60

0.15 1.71

1.8 20.8

12 8

identified at amino acid residues 191, 194, 197, 211 (Fig. 3B) of H. mediterranei narG. H. mediterranei narH encodes small h subunit from H. mediterranei respiratory nitrate reductase protein; it has the best scores with narH from H. marismortui (50% similarity), S. flexneri (28% identity) and S. coelicolor (28% identity). The deduced sequence clearly indicates the presence of one regular [2Fe – 4S] and three complete [4Fe – 4S] cluster binding motifs at the N-terminal of H. mediterranei narC (Fig. 3C), according to the possible arrangements of the four Fe –S clusters in the E. coli NarH protein [33]. Nitrate reductase belongs to the molybdoenzymes family, narJ codes for a putative chaperon-like protein involved in the Mo-pterin cofactor assembly. The sequence analysis comparison of H. mediterranei narJ shows a strong similarity with narJ from H. marismortui (63% similarity, 47% identity). The presence of specific chaperones for metalloenzymes seems to be widespread in Prokarya, although the operons of several molybdoenzymes do not encode NarJ type proteins. Other transcriptional units might also be involved in cofactor acquisition via processes similar to those involving NarJ in the maturation of these molybdoenzymes [34].

Purification (fold)

Yield (%)

for the native enzyme. SDS-PAGE of the purified enzyme in the presence of 23% of 2-mercaptoethanol showed two bands of Mr approximately 112 F 1.3 and 61.5 F 1.3 kDa (Fig. 4). These values correspond to estimated mass obtained from narG and narH gene products. The molecular mass observed is lower than that described for dissimilatory nitrate reductase from the same archaea [14], so we propose that the enzyme characterized in this study is a different enzyme from that purified by Alvarez-Ossorio et al. [14]. Respiratory nitrate reductases are complexes of two or three subunits depending on the method of isolation. This pattern has been described for several Nar purification processes [4,35]. Different respiratory nitrate reductases from Archaea have been purified: heterodimeric (116 and 60 kDa) or heterotrimeric (100, 61 and 31 kDa) structures have been reported as the subunits constructions of Haloferax denitrificans [16] and H. volcanii [15], respectively. The Nar from H. marismortui was first characterized as a homotrimer [35], but the sequence of the gene as well as SDS-PAGE in the presence of different concentrations of 2-

3.3. Respiratory nitrate reductase purification and characterization The respiratory nitrate reductase from H. mediterranei membranes has been purified and its properties analysed. Under the conditions described previously in Materials and methods, we observed the induction of Nar activity in H. mediterranei cultures. These results agree with other studies in which nitrate-reducing and denitrifying activities are induced under anaerobic growth conditions only in the presence of nitrate [4]. The purification scheme of Nar is summarised in Table 1, involving two DEAE-Sepharose CL-6B and Sephacryl S300 chromatographic steps. The enzyme was purified 21fold, and the specific activity of purified Nar was 1710 mU/ mg protein. Based on the purity of the preparation and using the enrichment factor of 21, it was calculated that the respiratory nitrate reductase constitutes approximately 4% of the membrane-bound protein. Gel-filtration chromatography under native conditions in a Sephacryl S-300 column indicated a mass of 155 F 6 kDa

Fig. 4. Gel electrophoresis of purified Nar from H. mediterranei. (a) SDSPAGE of the purified Nar: lane 1, standard proteins with their molecular masses indicated; lane 2, purified enzyme with 23% (w/v) 2-mercaptoethanol. (b) Native gel electrophoresis.

56

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

mercaptoethanol show that this enzyme is composed of two subunits with the apparent molecular masses of 117 and 47 kDa, respectively [17]. This significant difference among the archaeal respiratory nitrate reductases indicates the diversity in the structures. Nevertheless, the Nar archaeal enzymes follow the patterns of bacterial enzymes. It has been proposed that the properties of archaeal Nar are reminiscent of the bacterial enzymes with certain modifications [4]. Fractions containing Nar activity from the second DEAE-Sepharose CL-6B chromatography were combined and used for the characterization assays. The effect of electron donors and cofactors was tested. Reduced methyl viologen (MV) was the best electron donor (in vitro) for the respiratory nitrate reductase from H. mediterranei, as previously described for the assimilatory and dissimilatory nitrate reductases from the same archaea [2,14]. Nar from H. mediterranei did not use electrons from NADH (1 mM) (0.67% of the maximum activity) or NADPH (1 mM) (0.68% of the maximum activity). When NADH or NADPH was tested in the presence of MV, the Nar activity values were around 2.4% of maximum activity. Dithionite (DT) was not able to reduce nitrate in the absence of MV. Nar activity was not detected when reduced FAD (1 mM) was added to the assay solution as an electron donor. Several inhibitors of nitrate reductases were also tested. These compounds, dithiothreitol (1 mM) or EDTA, were not effective inhibitors for the nitrate reduction, decreasing partially the activity (75.45% and 77.8% of the maximum activity, respectively, at 1 mM). Azide and cyanide acted as potent inhibitors (31.6% and 24% of the maximum activity, respectively, at 1 mM). Like other non-halophilic and halophilic nitrate and nitrite reductases, cyanide and azide were strong inhibitors of the Nar from H. mediterranei [14,15,36,37]. These compounds are thought to inhibit the enzyme by metal chelating and the primary site of action is thought to be the molybdenum [38]. The pH dependence of enzyme activity in the range 5.9 to 8.9 was investigated in 100 mM Tris – HCl, 3.6 M NaCl, 4 mM MV and 35 mM KNO3. Assays were carried out at 40 and 60 jC for all pH values. Nar showed maximum activity at pH 7.9 and 8.2 when the nitrate reduction was carried out at 40 and 60 jC, respectively. These pH values are very similar to those that allow optimal Nar activity from H. denitrificans [16], H. volcanii [15] and P. aerophilum [39]. It has been described that dissimilatory nitrate reductase from H. mediterranei [14] showed maximum activity at pH 7.5 –7.7. This parameter also proves that the enzyme analysed in this study is not the same enzyme characterized previously from H. mediterranei [14]. Nar activity was also measured using the assay mixture with different NaCl concentrations at temperatures from 30 to 80 jC (Fig. 5). Nar activity does not exhibit a strong dependence on temperature at different NaCl concentrations. The maximum activity was found at 70 jC for all NaCl concentrations

Fig. 5. Effect of temperature on Nar activity from H. mediterranei at different NaCl concentrations. The activity was determined in the presence of (o) 3.8 M, (5) 2.8 M, (4) 1.4 M, and (5) 0.0 M NaCl.

assayed. The catalytic process was analysed according to the Arrhenius equation using a temperature range between 30– 60 jC. The activation energy was around 41 kJ/mol in both cases, absence NaCl or 3.8 M NaCl. Like halophilic nitrate reductase from genus Haloferax [2,14,15], Nar from H. mediterranei showed a remarkable thermophilicity and worked well up to 70 jC, but the Nar activity did not show a direct dependence on salt concentration (Fig. 5), as was expected. This pattern was also described for Nar from H. denitrificans and H. volcanii, in which optimum nitrate reductase activity could be found in the absence of NaCl. Not all nitrate reductases activities found in halophilic archaea exhibit similar dependence [14 – 16]. While most proteins from haloarchaea are stabilized by high concentrations of salts, there are some that are either active or stable in the absence of salt [16]. The origin of haloarchaeal enzymes that do not require salt is not clear; it has been proposed that Nar was acquired by the extreme halophiles from a eubacterial source [16]. In order to analyse Nar stability, the enzyme was incubated at 4 jC in the presence of 3.5, 2.5, 1.5, 0.5 M NaCl and in the absence of NaCl (in 100 mM phosphate buffer pH 8.0) and the activity was assayed after 170-h incubation. With 3.5 M NaCl, the enzyme was completely stable after 170 h. In the presence of 2.5, 1.5 and 0.5 M NaCl, the enzyme lost approximately 7% of the maximum activity, after the same incubation period. The activity decrease was 10% in the absence of NaCl. In general, the salts have a strong effect on the activity and stability of halophilic enzymes [14], i.e. the dissimilatory nitrate reductase from H. mediterranei is salt-dependent [14]; however, other

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

Fig. 6. (A) Absolute spectra of purified nitrate reductase (20 Ag of protein/ ml) in 10 mM Tris – HCl buffer (pH 8.0), containing 200 mM NaCl. The oxidised spectrum was obtained first and the reduced by re-running the same sample after addition of a few crystals of dithionite (B).

57

grown to stationary phase (complex aerobic medium) and transferred to anaerobic conditions with supplemented nitrate source. In aerobic conditions (pre-anaerobic), we could not detect mRNA for nar genes, but in post anaerobic conditions with nitrate, these could be amplified (Fig. 7), indicating that these are needed for nitrate respiration. In summary, regulation of nar genes occurs at transcriptional level induced by oxygen-limiting conditions and presence of nitrate. No regulatory protein could be identified in the vicinity of the nar operon, but the presence of a set of palindromic sequences in the promoter suggests that the transcriptional regulation occurs via protein binding. We were not able to identify any known regulatory sequences patterns, indicating that this transcriptional regulation could be novel in the denitrifier organisms and owed to Archaea. In conclusion, genes coding nar operon have been sequenced and sequence alignment shows strong similarities between extremophilic organisms. Nar operon expression is regulated at transcriptional level and sets of palindromic sequences in the promoter could be implied in its regulation. The Nar from H. mediterranei was induced during anaerobic incubation and it was obtained from membrane. These results suggest that nitrate reductase purified from H. mediterranei (strain R-4) in this

nitrate reductases from Haloferax and Haloarcula are not salt-dependent [16,35]. Kinetic parameters of Nar were determined using different concentrations of MV and nitrate, in the presence of 100 mM Tris –HCl buffer (pH 8.0) containing 3.6 M NaCl. The halophilic enzyme followed a Michaelis –Menten kinetic. Apparent Km values for nitrate and MV were 0.82 F 0.14 and 0.25 F 0.015 mM, respectively. The apparent Km for nitrate was in the range of the values obtained from another nitrate reductases (0.3 and 3.8 mM) [4]. The apparent Km obtained for nitrate by dissimilatory nitrate reductase from H. mediterranei was 6.7 mM [14]. The purified enzyme was brown in colour and had an absorption spectrum as shown in Fig. 6. In addition to the absorbance maximum due to the protein at 290 nm, there was an interesting broad band around 400 to 415 nm, which disappears on reduction with DT. Neither cytochrome nor flavin moiety was found in the purified preparation as judged from the absorption spectra [4]. The Fe –S clusters have a maximum between 400 and 460 nm. The shape of the absorption spectrum on the positions of the maximum indicated that Nar from H. mediterranei has Fe – S clusters (Fig. 6). This spectrum is similar to those obtained from different purified Nar from denitrifying microorganisms. 3.4. Regulation of nar genes In the present study, we used RT-PCR to determine the effect of anaerobic conditions and nitrate presence in the expression of nar operon. In order to analyse the gene expression, total RNA was isolated from H. mediterranei

Fig. 7. Analysis of the expression of nar operon by agarose gel electrophoresis of double-stranded DNA fragments generated in RT-PCR reactions. Reactions had been performed with total RNA isolated from complex medium in aerobic conditions (Lane 2), and in anaerobic conditions supplemented with nitrate (Lane 3). Lane 1, control reaction performed with H. mediterranei genomic DNA. Lane MW and 100 bp: molecular size markers.

58

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

study is a respiratory membrane nitrate reductase (Nar), which is different from nitrate reductase purified previously from the same microorganism. Some properties are different from those obtained from dissimilatory nitrate reductase from H. mediterranei [14], which probably allows the dissipation of excess of reducing power for redox balancing. This study sheds light on the role played by the nitrate respiratory pathway in Archaea, and opens the way to further research in this field.

Acknowledgements This work was supported by funds from CTIDIB/2002/ 154. Belen Lledo´ held a scholarship from the Spanish Ministerio Educacio´n, Cultura y Deporte during the development of this work. DNA sequencing was carried out in ‘‘Unidad de Biologı´a Molecular y Ana´lisis Gene´tico’’ of Servicios Te´cnicos de Investigacio´n (Universidad de Alicante).

[12]

[13]

[14]

[15]

[16]

[17]

[18]

References [1] D.J. Richardson, B.C. Berks, D.A. Russel, S. Spiro, C.J. Taylor, Functional Biochemical and genetic diversity of prokaryotic Nitrate Reductases, Cell. Mol. Life Sci. 58 (2001) 165 – 178. [2] R.M. Martı´nez-Espinosa, F.C. Marhuenda-Egea, M.J. Bonete, Assimilatory nitrate reductase from the haloarchaeon Haloferax mediterranei: purification and characterisation, FEMS Microbiol. Lett. 204 (2001) 381 – 385. [3] L. Philippot, Denitrifying genes in bacterial and Archaeal genomes, Biochim. Biophys. Acta 1577 (2002) 355 – 376. [4] W.G. Zumft, Cell biology and molecular basis of denitrification, Microbiol. Mol. Biol. Rev. 61 (1997) 533 – 616. [5] B.C. Berks, S.J. Ferguson, J.W. Moir, D.J. Richardson, Enzymes and associated electron transport systems that catalyse the respiratory reduction of nitrogen oxides and oxyanions, Biochim. Biophys. Acta 1232 (1995) 97 – 173. [6] F. Blasco, C. Iobbi, G. Giordano, M. Chippaux, V. Bonnefoy, Nitrate reductase of Escherichia coli: completion of the nucleotide sequence of the nar operon and reassessment of the role of the alpha and beta subunits in iron binding and electron transfer, Mol. Gen. Genet. 218 (1989) 249 – 256. [7] T. Hoffmann, B. Troup, A. Szabo, C. Hungerer, D. Jahn, The anaerobic life of Bacillus subtilis: cloning of the genes encoding the respiratory nitrate reductase system, FEMS Microbiol. Lett. 131 (1995) 219 – 225. [8] L. Philippot, A. Clays-Josserand, R. Lensi, I. Trinsoutreau, P. Normand, P. Potier, Purification of the dissimilative nitrate reductase of Pseudomonas fluorescens and the cloning and sequencing of its corresponding genes, Biochim. Biophys. Acta 1350 (3) (1997) 272 – 276. [9] F. Blasco, J.P. Dos Santos, A. Magalon, C. Frixon, B. Guigliarelli, C.L. Santini, G. Giordano, NarJ is a specific chaperone required for molybdenum cofactor assembly in nitrate reductase A of Escherichia coli, Mol. Microbiol. 28 (1998) 435 – 447. [10] M.D. Marger, M.H. Saier, A major superfamily of transmembrane facilitators that catalyse uniport, symport and antiport, Trends Biochem. Sci. 18 (1993) 13 – 20. [11] L. Philippot, P. Mirleau, S. Mazurier, S. Siblot, A. Hartmann, P. Lemanceau, J.C. Germon, Characterization and transcriptional anal-

[19]

[20]

[21]

[22] [23] [24]

[25]

[26] [27] [28] [29]

[30]

[31]

ysis of Pseudomonas fluorescens denitrifying clusters containing the nar, nir, nor and nos genes, Biochim. Biophys. Acta 1517 (2001) 436 – 440. G. Unden, S. Becker, J. Bongaerts, G. Holighaus, J. Schirawski, S Six, O2-sensing and O2-dependent gene regulation in facultatively anaerobic bacteria, Arch. Microbiol. 164 (1995) 81 – 90. E. Hartig, U. Schiek, K.U. Vollack, W.G. Zumft, Nitrate and nitrite control of respiratory nitrate reduction in denitrifying Pseudomonas stutzeri by a two-component regulatory system homologous to NarXL of Escherichia coli, J. Bacteriol. 181 (1999) 3658 – 3665. M.C. Alvarez-Ossorio, F.J.G. Muriana, F.F. de la Rosa, A.M. Relimpio, Purification and characterization of nitrate reductase from the halophile archaebacterium Haloferax mediterranei, Z. Naturforsch. 47C (1992) 670 – 676. S. Bickel-Sandko¨tter, M. Ufer, Properties of a dissimilatory nitrate reductase from the halophilic archaeon Halofertax volcanii, Z. Naturforsch. 50C (1995) 365 – 372. L.I. Hochstein, F. Lang, Purification and properties of a dissimilatory nitrate reductase from Haloferax denitrificans, Arch. Biochem. Biophys. 288 (1991) 380 – 385. K. Yoshimatsu, T. Iwasaki, T. Fujiwara, Sequence and electron paramagnetic resonance analyses of nitrate reductase NarGH from a denitrifying halophilic euryarchaeote Haloarcula marismortui, FEBS Lett. 516 (2002) 145 – 150. F. Rodrı´guez-Valera, F. Ruiz Berraquero, A. Ramos Cormenzana, Behaviour of mixed populations of halophilic bacteria in continuous cultures, Can. J. Microbiol. 26 (1980) 1259 – 1263. M. Dyall-Smith, M. Holmes, M. Kamekura, W. Lam, S. Nuttal, W. Woods, P. Jablonski, J. Serrano, K. Ngui, S.L. Tang, V. Tarasov, V. Athanasopoulos, C. Bath, D. Wendoloski, C. Ferrer, M. Pfeiffer, M. Pohlschroder, T. Allers, The Halohandbook: Protocols for Halobacterial Genetics. Version 4.5, Haloarchaeal Genetics Laboratory, Department of Microbiology and Inmunology, University of Melbourne, Australia, 2001. F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Sthruhl, (Eds.) Current Protocols in Molecular Biology. Greene Publishing Associates and Wiley-Interscience, New York, 1989. R. Chenna, H. Sugawara, T. Koike, R. Lopez, T.J. Gibson, D.G. Higgins, J.D. Thompson, Multiple sequence alignment with the Clustal series of programs, Nucleic Acids Res. 31 (2003) 3497 – 3500. S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment search tool, J. Mol. Biol. 215 (1990) 403 – 410. W.R. Pearson, D.J. Lipman, Improved tools for biological sequence comparison, Proc. Natl. Acad. Sci. U. S. A. 85 1988, pp. 2444 – 2448. R.M. Martı´nez-Espinosa, F.C. Marhuenda-Egea, M.J. Bonete, Purification and characterisation of a possible assimilatory nitrite reductase from the halophile archaeon Haloferax mediterranei, FEMS Microbiol. Lett. 196 (2001) 113 – 118. J.M. Vega, H. Kamin, Spinach nitrite reductase. Purification and properties of a sirohome-containing iron-sulfur enzyme, J. Biol. Chem. 252 (1977) 896 – 909. J.R. Palmer, C.J. Daniels, In vivo definition of an archaeal promoter, J. Bacteriol. 177 (1995) 1844 – 1849. J. Soppa, Basal and regulated transcription in Archaea, Mol. Microbiol. 31 (1999) 1295 – 1305. P.P. Dennis, Ancient ciphers: translation in Archaea, Cell 89 (1997) 1007 – 1010. G. Fiorentino, R. Cannio, M. Rossi, S. Bartolucci, Transcriptional regulation of the gene encoding an alcohol dehydrogenase in the archaeon Sulfolubus solfataricus involves multiple factors and control elements, J. Bacteriol. 185 (2003) 3926 – 3934. J.A. Fee, D. Kuila, M.W. Mather, T. Yoshida, Respiratory proteins from extremely thermophilic, aerobic bacteria, Biochim. Biophys. Acta 853 (1986) 153 – 185. O. Zafra, S. Ramirez, P. Castan, R. Moreno, F. Cava, C. Valles, E. Caro, J. Berenguer, A cytochrome c encoded by the nar operon is

B. Lledo´ et al. / Biochimica et Biophysica Acta 1674 (2004) 50–59

[32]

[33]

[34]

[35]

required for the synthesis of active respiratory nitrate reductase in Thermus thermophilus, FEBS Lett. 523 (2002) 99 – 102. F. Rodrı´guez-Valera, G. Juez, D.J. Kushner, Halobacterium mediteranei spec. nov., a new carbohydrate utilizing extreme halophile, Syst. Appl. Microbiol. 4 (1983) 369 – 381. B. Guigliarelli, A. Magalon, M. Asso, P. Bertrand, C. Frixon, G. Giordano, F. Blasco, Complete coordination of the four Fe – S centers of the beta subunit from Escherichia coli nitrate reductase. Physiological, biochemical, and EPR characterization of site-directed mutants lacking the highest or lowest potential [4Fe – 4S] clusters, Biochemistry 35 (1996) 4828 – 4836. F. Blasco, B. Guigliarelli, A. Magalon, M. Asso, G. Giordano, R.A. Rothery, The coordination and function of the redox centres of the membrane-bound nitrate reductases, Cell. Mol. Life Sci. 58 (2001) 179 – 193. K. Yoshimatsu, T. Sakurai, T. Fujiwara, Purification and charac-

[36]

[37]

[38]

[39]

59

terization of dissimilatory nitrate reductase from denitrifying halophilic archaeon, Haloarcula marismortui, FEBS Lett. 470 (2000) 216 – 220. I. Ken-Ichi, L.I. Hochstein, The purification and properties of a copper nitrite reductase from Haloferax denitrificans, Curr. Microbiol. 32 (1996) 72 – 76. C. Moreno-Vivia´n, P. Cabello, M. Martı´nez-Luque, R. Blasco, F. Castillo, Prokaryotic nitrate reduction: molecular properties and functional distinction among bacterial nitrate reductases, J. Bacteriol. 181 (1999) 6573 – 6584. D.W. McDonald, A. Coddington, Properties of assimilatory nitrate reductase from Aspergillus nidulans, Eur. J. Biochem. 46 (1974) 169 – 178. S. Afshar, E. Johnson, S. de Vries, I. Schro¨der, Properties of a thermostable nitrate reductase from the hyperthermophilic archaeon Pyrobaculum aerophilum, J. Bacteriol. 183 (2001) 5491 – 5495.