Analysis of genes involved in nitrate reduction in Clostridium perfringens

Katsuyo Fujinaga1, Yuki Taniguchi1, Yezhou Sun1, Seiichi Katayama1, Junzaburo Minami1, Osamu Matsushita1 and Akinobu Okabe1

Department of Microbiology, Faculty of Medicine, Kagawa Medical University, Miki-cho, Kita-gun, Kagawa 761-0793, Japan 1

Author for correspondence: Akinobu Okabe. Tel: +81 87 891 2129. Fax: +81 87 891 2129. e-mail: microbio{at}kms.ac.jp


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We have conducted the genetic analysis of fermentative nitrate reduction in Clostridium perfringens, a strict anaerobic bacterium. Nitrate reductase (NarA) was purified from the cytoplasmic fraction of the organism. Using a degenerate primer designed from its N- terminal amino acid sequence, a 9·5 kb fragment containing seven ORFs was cloned. The molecular mass and N-terminal amino acid sequence predicted from the nucleotide sequence of ORF 4 coincided with those determined for the purified NarA, indicating that ORF 4 corresponds to a narA gene. ORFs 5 and 6 encode a 15·4 kDa ferredoxin-like protein containing four iron–sulfur clusters and a 45 kDa protein homologous to NADH oxidase, respectively. Analyses involving primer extension and Northern blotting revealed that these three ORFs are transcribed as a polycistronic message. The ORF 5- and ORF 6-encoded proteins were shown by immunoblotting to be synthesized by cells grown in the presence of nitrate. Thus, these two proteins are likely to function as electron-transfer components in nitrate reduction in C. perfringens. The 9·5 kb fragment and a downstream region of 6·1 kb do not contain any genes involved in nitrate uptake or nitrite reduction. Instead, all 5 ORFs downstream of ORF 6 are homologous to genes reported for molybdopterin biosynthesis, unlike the genomic organization already determined for the respiratory and assimilatory nitrate-reduction systems. The evolutionary relationships between these two nitrate- reduction systems and the fermentative one based on the results of comparative genetic analysis are discussed.

Keywords: Clostridium perfringens, nitrate reductase, molybdopterin guanine dinucleotide, ferredoxin-like protein, NADH oxidase

Abbreviations: NaR, nitrate reductase; NarA, nitrate reductase of Clostridium perfringens; NiR, nitrite reductase; Mo-MGD, a molybdenum–molybdopterin guanine dinucleotide complex

The GenBank accession number for the sequence reported in this paper is AB017192.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Nitrate is produced from molecular nitrogen mainly by bacteria through nitrogen fixation and nitrification, and is universally present as the principal storage form of nitrogen in soil. It is utilized by many bacteria, fungi, algae and plants as a nitrogen source via nitrate assimilation. The physiology and genetics of nitrate assimilation have been well studied in plants and fungi (Campbell & Kinghorn, 1990 ; Crawford & Arst, 1993 ; Crawford, 1995 ; Campbell, 1996 ), and more recently in bacteria, such as Synechococcus, Bacillus, Klebsiella and Azotobacter (Lin & Stewart, 1998 ). Nitrate assimilation involves three pathway-specific steps: uptake of nitrate, reduction of nitrate to nitrite catalysed by nitrate reductase (NaR) and the six-electron reduction of nitrite to ammonium by nitrite reductase (NiR). Nitrate respiration is another metabolic process involving nitrate reduction; nitrate can be used as an electron acceptor under anaerobic conditions. Respiratory NaR couples substrate oxidation to nitrate reduction to generate the proton-motive force, and the resultant nitrite is finally converted to ammonium ions via ammonification or to gaseous nitrogen via denitrification. Respiratory NaR is widespread in both the Eubacteria (Berks et al., 1995 ) and Archaea (Zumft, 1992 ; Völkl et al., 1993 ) kingdoms. Its bioenergetics and genetics have been studied most extensively in Paracoccus denitrificans, Escherichia coli and Rodobacter capsulatus (Berks et al., 1995 ; Gennis & Stewart, 1996 ).

Both assimilatory and respiratory NaRs are molybdoenzymes containing a molybdopterin cofactor and an iron–sulfur cluster. They differ in that the former is soluble in the cytoplasm, while the latter is membrane bound. They also differ in the regulation of enzyme synthesis: the synthesis of the former is controlled by the availability of fixed nitrogen, e.g. glutamine and ammonium ions (Lin & Stewart, 1998 ), while the synthesis of the latter is independent of their availability. Another type of NaR, which is more similar to assimilatory NaR than to respiratory NaR (Berks et al., 1995 ), has been identified in the periplasm, although its physiological function has not yet been established.

Nitrate reduction in strict anaerobes has attracted much less interest and is not well understood mainly because of the lack of suitable bacteria for physiological and genetic studies. Clostridium perfringens, a strict anaerobic pathogen, lives in soil, water, and the intestines of humans and other animals. Some strains of the organism can reduce nitrate to ammonium ions (Hasan & Hall, 1975 ). NaR (Seki-Chiba & Ishimoto, 1977 ) and NiR (Sekiguchi et al., 1983 ) of C. perfringens, hereinafter designated NarA and NirA, respectively, were purified and their biochemical properties were characterized. NarA is a soluble molybdoenzyme whose molecular mass was estimated to be approximately 80 kDa (Seki-Chiba & Ishimoto, 1977 ; Seki et al., 1987 ). The enzyme can catalyse NAD(P)H-dependent nitrate reduction in the presence of rubredoxin (Seki et al., 1989 ) and a partially purified NAD(P)H-rubredoxin reductase (Seki et al., 1988 ). Its location and molecular size are more similar to those of assimilatory NaRs than to those of respiratory NaRs. However, induction of the enzyme activity by nitrate is independent of the presence of ammonium ions in the culture medium, as in the case of respiratory NaRs (Hasan & Hall, 1975 ). NirA does not contain siroheme, unlike assimilatory NiRs, or haem or copper, unlike respiratory NiRs. Its molecular mass, 54 kDa, is significantly less than those reported for assimilatory and respiratory NiRs (63 kDa for higher plant NiRs, and 88–140 kDa for fungal and bacterial NiRs) (Sekiguchi et al., 1983 ). The nitrite reduction by NirA generates hydroxylamine as the main product and possibly nitric oxide, but not ammonium ions (Sekiguchi et al., 1983 ). Thus, nitrate reduction in C. perfringens differs in many respects from assimilatory and respiratory nitrate reduction. This unique nitrate-reduction system has been considered not to be linked to nitrate assimilation but to energy production: this is called nitrate fermentation (Takahashi et al., 1963 ; Ishimoto et al., 1974 ; Hasan & Hall, 1975 ).

NarA and NirA were well studied until the late 1980s. Since then, however, no further work has been conducted and much of this system remains to be elucidated. In an attempt to determine the molecular characteristics of NarA and NirA, we have undertaken the cloning and sequencing of their genes (narA and nirA). Nitrate is reduced by bacteria to toxic compounds when it is ingested along with meals (Calmels et al., 1996 ; Vermeer et al. , 1998 ). Considering that C. perfringens inhabits the intestines of humans and other animals, and also that it may generate potentially harmful byproducts of nitrate reduction such as hydroxylamine and nitrous oxides (Sekiguchi et al., 1983 ), genetic analysis of nitrate reduction in C. perfringens is also of medical importance. Moreover, it would provide insights into the evolutionary relationships between the fermentative and other nitrate-reduction enzymes. In this study we have characterized the nar operon of C. perfringens and the downstream genes involved in molybdopterin synthesis, and compare them with already characterized homologous genes.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
The C. perfringens strain used in this study was type A PB6K (Soda et al., 1968 ). The plasmid vector, pT7Blue T-vector, and the host strain, E. coli NovaBlue (Novagen), were used for cloning DNA fragments obtained by PCR. To express the narB gene, we used pET-11a, a T7 RNA polymerase based expression vector (Novagen), and E. coli BL21(DE3) as the plasmid vector and host strain, respectively. To express the narC gene, we used pGEX-4T-2 and E. coli DH5{alpha} as the plasmid vector and host strain, respectively. The plasmid vector pUC19 (Messing, 1983 ) and the host strain E. coli DH5{alpha} (BRL) were used for all other recombinant DNA experiments.

Media and culture conditions.
C. perfringens PB6K was precultured in cooked-meat medium (Nissui Pharmaceuticals) at 37 °C for 12 h. The preculture was diluted 200-fold with fresh medium, GAM (Nissui Pharmaceuticals) or PYG broth (7·5 g polypeptone, 5 g yeast extract, 2·5 g glucose and 50 µl thioglycolic acid l -1; Seki-Chiba & Ishimoto, 1977 ) and incubated at 37 °C. KNO3 was added to the medium to a final concentration of 0·2% (w/v), when necessary. Transformed E. coli cells were selected on LB plates containing 20 g LB broth base (Gibco) and 15 g agar per litre, supplemented with 100 µg ampicillin ml-1, 0·2 mM IPTG (Wako Pure Chemicals) and 0·004% (w/v) X-Gal (Wako Pure Chemicals).

Assaying of NarA activity.
Cultures were collected by centrifugation at 12000 g for 10 min at 4 °C. The cells were washed three times with 10 mM Tris/HCl (pH 8·0) by centrifugation. The cell suspension was passed twice through a French pressure cell at 10000 p.s.i. (6·9 MPa), followed by centrifugation at 18000  g for 20 min. The supernatant was used as the crude extract. NarA activity was determined using methyl viologen as a substrate, as described by Seki-Chiba & Ishimoto (1977) . Enzyme activity was expressed as nmol nitrite formed min-1 . Protein concentrations were determined using the BCA protein assay reagent (Pierce) with BSA as a standard. All assays were carried out in triplicate.

SDS-PAGE and zymography.
SDS-PAGE on a 7·5% polyacrylamide gel and staining with Coomassie brilliant blue R were performed as described by Laemmli (1970) . To detect NarA activity by zymography, proteins were first separated on a gel under non-denaturing conditions and visualized by means of substrate-dependent oxidation of reduced methyl viologen, as described by Seki-Chiba & Ishimoto (1977) .

Purification of NarA.
C. perfringens PB6K was grown in 1 l PYG medium containing 0·2% (w/v) KNO3 at 37 °C for 9 h. A crude extract was prepared as described above. Purification was carried out by the method of Seki-Chiba & Ishimoto (1977) with the following modification. Briefly, ammonium sulfate was added to the crude extract to give 40% saturation, followed by centrifugation at 15000 g at 4 °C for 20 min. Ammonium sulfate was added to the supernatant to give 80% saturation and the precipitate was collected by centrifugation. The precipitate was dissolved in 15 ml 10 mM Tris/HCl (pH 8·0) and dialysed against the same buffer. The dialysed sample was applied to a DEAE Sephadex A-25 column (2·2x15 cm; Amersham Pharmacia Biotech). Proteins were eluted with a 400 ml linear gradient of 0–0·3 M NaCl in Tris/HCl (pH 8·0). Enzyme activity and the absorbance at 280 nm were monitored. The fractions eluted in the active peak were collected and the pooled fraction was dialysed against 10 mM Tris/HCl (pH 8·0). The dialysed sample was applied to an ion-exchange FPLC column (MonoQ, 1 ml bed volume; Amersham Pharmacia Biotech), that had been pre-equilibrated with 20 mM Tris/HCl (pH 8·0). Proteins were eluted with a 20 ml linear gradient of 0–0·3 M NaCl in the same buffer. The enzyme was eluted as a single peak at 170 mM NaCl. This step was repeated three times and the active fractions were combined.

Determination of the N-terminal amino acid sequence.
The purified NarA was subjected to SDS-PAGE, followed by transfer to a PVDF membrane (Trans-Blot transfer medium; Bio-Rad). Proteins were visualized by staining with Coomassie brilliant blue R and the corresponding protein band was cut out. The excised membrane was wetted with acetonitrile and incubated in 100 mM Tris/HCl (pH 8·3) containing 6 M guanidine hydrochloride and 2 mg DTT ml-1 at 37 °C for 2 h. Alkylation of cysteine residues was performed by incubation at 37 °C for 2 h after the addition of 4-vinylpyridine to a final concentration of 1% (w/v). After washing five times with 50% (v/v) methanol, the membrane was applied to a protein sequencer (model 492; Perkin Elmer) to determine the N-terminal amino acid sequence.

Cloning of a portion of the narA gene.
A pair of degenerate oligonucleotide primers was designed to amplify a portion of the narA gene by PCR. Their sequences, 5'-ACITGYAAYTAYTGYGCIYT-3' and 5'- CKYTTIATYTTICCRTCYTC-3', correspond to the amino acid sequences of Tyr7–Leu13 and Glu21–Arg 27, respectively, of NarA. C. perfringens chromosomal DNA was prepared as described previously (Matsushita et al., 1994 ) and used as the template. PCR was performed with a thermal cycler (model TC1, Perkin Elmer), and the PCR product was purified from an acrylamide gel and cloned into the pT7Blue T-vector (Novagen), as described elsewhere (Matsushita et al., 1999 ). The resulting plasmid, which was designated pNR1, was transformed into E. coli NovaBlue.

Cloning of the nar operon.
The insert DNA in pNR1 was amplified, labelled with digoxigenin-11- dUTP (Roche Diagnostics) by PCR as described previously (Matsushita et al., 1999 ) and used as a DNA probe. The chromosomal DNA was digested with PstI and then subjected to Southern hybridization at 55  °C. DNA fragments around the positive signal were recovered from an agarose gel and ligated into pUC19. E. coli DH5{alpha} was transformed with the ligation mixture. The colonies were screened by PCR as described previously (Matsushita et al., 1999 ). Of the plasmids recovered from positive clones, a plasmid named pNR4 produced a PCR amplification product of the expected size. This plasmid was confirmed by nucleotide sequencing to contain the insert DNA, of which the deduced amino acid sequence coincided with that determined for the purified NarA. This plasmid contained a 9·5 kb Pst I fragment. A further downstream region was detected on Southern hybridization using chromosomal DNA digested with HincII and a DNA probe prepared by PCR amplification of the region corresponding to nt 9286–9416 of the PstI fragment. The 6·1 kb hybridizing fragment was cloned into pUC19, as described above. The resulting plasmid was designated pNR7.

DNA manipulations and nucleotide sequencing.
Restriction endonucleases were purchased from Takara Shuzo, Toyobo and New England Biolabs. The DNA ligation kit was a product of Takara Shuzo. All recombinant DNA procedures were carried out as described by Sambrook et al. (1989) . The nucleotide sequence was determined by the dideoxy chain-termination method (Sanger et al., 1977 ), using an automated nucleotide sequencer (model ABI PRISM 377, Perkin Elmer). Plasmid template DNA was prepared using the Wizard plus miniprep DNA purification system (Promega). A Thermo Sequenase fluorescent-labelled primer cycle-sequencing kit with 7-deaza dGTP (Amersham Pharmacia Biotech) and M13 dye primers (Perkin Elmer) were used for sequencing. An ABI PRISM dye-terminator cycle- sequencing ready reaction kit with AmpliTaq DNA polymerase FS (Perkin Elmer) and various synthetic primers were also used to determine ambiguous nucleotides and to fill in sequence gaps.

Southern hybridization.
C. perfringens DNA (0·2 µg) was digested with appropriate restriction enzymes, applied to a 0·8% agarose mini-gel and electrophoresed at 100 V for 1 h. Treatment of the gel and transfer of DNA to a nylon membrane (Hybond-N; Amersham Pharmacia Biotech) were performed as described previously (Matsushita et al., 1994 ). After hybridization at 50 °C, the hybridized probes were detected with anti- digoxigenin-alkaline phosphatase Fab fragments (Roche Diagnostics) and a chemiluminescent dye (Lumi-Phos 530; Lumigen).

Northern hybridization and primer-extension analysis.
C. perfringens cells were grown in GAM broth with or without 0·2% KNO3 at 37 °C for 2 h until the culture reached an OD600 of 0·8. Total RNA was prepared by the SDS-phenol method (Aiba et al., 1981 ) and a sample (1 µg) was separated on a denaturing agarose gel (1%). Hybridization was carried out at 50 °C using DNA fragments within the narA (nt 4629–5605), narB (nt 6115–6528), narC (nt 6845–7119), mobB (nt 7830–8261) and moeA (nt 9286–9416) genes as DNA probes. RNA molecular weight marker II (Roche Diagnostics) was used as size markers. Primer-extension experiments were performed using the same RNA sample as described above. A 30 nt primer, 5'- GAGCACAATAGTTACAAGTAGATTGTATTC-3', which is complementary to nt +8 to +37 of the narA gene coding region, was 5' end labelled with [{gamma}-32P]dATP [4·5 kCi/mmol -1 (166·5 TBq mmol-1); ICN Biochemicals] and hybridized with total RNA from C. perfringens cells. The hybrids were extended with reverse transcriptase (Superscript RT; Life Technologies) and the extension products were electrophoresed on a sequencing gel as described by Ba-Thein et al. (1996) .

Amino acid sequence similarity search.
A search for similar protein sequences was carried out using the BLAST algorithm (Altschul et al., 1997 ) at the Center for Information Biology (National Institute of Genetics, Mishima, Japan). The default settings in the BLAST program were used without prefiltering low compositional complexity regions.

Expression of the narB and narC genes in E. coli.
The narB gene was expressed in E. coli as follows. The DNA fragment corresponding to NarB aa 1–137 was amplified using the two primers 5'-CATATGAAAAGAATAAAAATCAATAGAGAT-3' and 5'-GGATCC TTATTCTTCTACCTCCTCTAAAAGT-3', which contained NdeI and BamHI sites, respectively (underlined). A 423 bp Nde I–BamHI fragment containing the coding region of the narB gene was inserted into expression vector pET-11a. The resulting plasmid, named pET-NarB, was transformed into E. coli BL21(DL3), after the nucleotide sequence of the narB gene had been verified. The transformant was grown in LB broth containing 100 µg ampicillin ml-1 at 37 °C. When the cultures reached an OD600 of 0·6, IPTG was added to a final concentration of 1 mM. After incubation for 4 h, cells were harvested by centrifugation at 5000 g for 5 min at 4 °C. The cells were washed once with 10 mM phosphate buffer (pH 7·0) containing 1 mM EDTA, disrupted by two passages through a French pressure cell at 10000 p.s.i. (6·9 MPa) and centrifuged at 13000 g for 20 min. The precipitates containing large amounts of NarB inclusion bodies were washed three times with 10 mM phosphate buffer (pH 7·0) containing 4% Triton X-100 and 1 mM EDTA by centrifugation at 25000 g for 20 min to remove membrane proteins. To remove residual Triton X- 100, the precipitates were washed twice with distilled water. The NarB protein was solubilized with 50 mM Tris/HCl (pH 8·5) containing 8 M urea and 10 mM DTT. The NarB protein recovered in the supernatant fraction was dialysed against the same buffer containing 4 M urea and 10 mM DTT. The dialysed sample was applied to a Sephadex G75 column (Amersham Pharmacia Biotech) and the protein recovered in the void fraction was used as the purified recombinant NarB protein, since this was shown to be homogeneous on SDS-PAGE.

The narC gene was expressed in E. coli as follows. The DNA fragment corresponding to the coding region of the narC gene was amplified using two PCR primers 3'-GGATCC ATGAGATATATTGTTGTGGGGGCA-3' and 5'-CTCGAG TTAATATGTGTACTCTAAATTTTC-3', which contained BamHI and XhoI sites, respectively (underlined). A 1·2 kb BamHI–XhoI fragment was inserted into the pGEX- 4T-2 vector (Amersham Pharmacia Biotech) to produce the NarC protein as a glutathione S-transferase fusion protein. After the nucleotide sequence of the fusion gene had been verified, the resulting plasmid, named pGST-NarC, was transformed into E. coli DH5{alpha}. The production and purification of the fusion protein were performed as described by the manufacturer.

Immunological methods.
Polyclonal antibodies were raised in ddY mice by subcutaneous injection of 40 µg recombinant NarB or NarC protein in Freund’s complete adjuvant (Amersham Pharmacia Biotech). Booster injections were given after 3 weeks and bleeding was performed after 4 weeks. Prior to use for Western blots, the antisera were preadsorbed with a large excess of a cell extract prepared from E. coli BL21(DE3) cells harbouring pET-11a overnight at 4 °C. The preadsorbed antisera were used at 1:100 dilution. For Western-blot analysis, 1 ml cultures of C. perfringens were centrifuged in a microfuge. The cell pellets were resuspended in 100 µl SDS-containing sample buffer and boiled for 3 min. Twenty microlitres of a sample was subjected to SDS-PAGE on either a 12·5 or 15% polyacrylamide gel and the proteins were blotted onto a nitrocellulose membrane. The replicates were incubated with mouse antiserum against the NarB or NarC protein and then subjected to immunostaining with goat anti-mouse IgG–alkaline phosphatase conjugate (Bio-Rad), and nitro blue tetrazolium (Sigma) and BCIP (5- bromo-4-chloro-3-indolyl phosphate, p-toluidine salt; Wako Pure Chemicals) as chromogenic substrates.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Cloning and sequencing of narA and neighbouring genes
To clone the narA gene, we purified NarA from C. perfringens PB6K to homogeneity, as shown by SDS-PAGE (Fig. 1a ) and zymography (Fig. 1b, c ). The molecular mass of NarA was determined to be approximately 74 kDa. The N-terminal amino acid sequence of NarA was determined to be MKRIQSTCNYCALACNLDFYTEDGKIKRVVPT. PCR with the degenerate primers matching this sequence and the C. perfringens chromosome DNA amplified a DNA fragment of the expected size (62 bp). After this fragment had been cloned into the pT7Blue T- vector and its nucleotide sequence had been confirmed to coincide with the amino acid sequence, we performed Southern hybridization using it as a DNA probe. The 9·5 kb PstI fragment detected was cloned into pUC19. The entire sequence of the insert fragment was determined (data not shown; accession number AB017192). Seven ORFs are present in this 9·5 kb fragment, the two reading frames at the 5' and 3' extremities being truncated (Fig. 2 ). The N-terminal amino acid sequence determined for the purified NarA coincided with that deduced from the sequence of ORF 4. The molecular mass of the protein encoded by ORF 4 was calculated to be 78436 Da, this being in agreement with the value determined for the purified NarA. Thus, ORF 4 was identified as the narA gene.



View larger version (84K):
[in this window]
[in a new window]
 
Fig. 1. Analysis of purified NarA by SDS-PAGE and zymography. Purified NarA [1 µg (a) or 2·5 µg (b, c)] was electrophoresed on an SDS-containing 7·5% acrylamide gel (a) and a 10% native acrylamide gel (b, c). The gels were subjected to staining with Coomassie brilliant blue R (a, b) or zymography (c), as described in Methods. Lane 1, markers; lane 2, purified NarA.

 


View larger version (12K):
[in this window]
[in a new window]
 
Fig. 2. The physical/genetic map of the C. perfringens nar gene cluster and its upstream and downstream regions cloned into pNR4 and pNR7. The restriction sites are: P, PstI; E, EcoRI; X, XbaI; S, SpeI; H, Hin cII.

 
A similarity search of the protein database revealed that the proteins encoded by ORFs 4, 5 and 6 exhibit significant similarity to NasC (assimilatory NaR) of B. subtilis, DmsB (a ferredoxin- like electron-transfer subunit of dimethyl sulfoxide reductase) of E. coli and NasB (a NADH-oxidase subunit of assimilatory nitrate reductase) of B. subtilis, respectively (Table 1 ). This suggests that these three gene products form NaR and its electron-transfer carriers. Thus, ORFs 5 and 6 were tentatively designated the narB and narC genes, respectively. The protein encoded by ORF 7 shows homology to the mobB gene product of E. coli, which has been shown to be involved in the synthesis of the molybdenum–molybdopterin guanine dinucleotide complex (Mo-MGD), a cofactor of NaR of E. coli. There was no ORF with significant similarity to already known NiRs or proteins involved in nitrate transport. That the NiR and NaR/NiR transporter genes are not in the vicinity of the narA gene differs from the genetic organization in Synechococcus sp., B. subtilis and Klebsiella pneumoniae, in which assimilatory NaR and NiR genes are clustered (Lin & Stewart, 1998 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Genes and gene products of the nar operon and neighbouring genes

 
In an attempt to find a nirA gene in a further downstream region, we cloned and sequenced the 6·1 kb HincII fragment that overlapped the 3' region of the 9·6 kb PstI fragment (Fig. 2). A total of 14393 bp were sequenced by combining the nucleotide sequences of the two fragments (data not shown; accession number AB017192). A partial physical map, and the locations of narA, narB, narC and other ORFs are presented in Fig. 2. The protein sequences predicted from ORFs 7 to 11 showed significant homology to the products of the mob, moa and moe genes (Table 1), which are considered to be involved in the synthesis of Mo-MGD. Database searches did not reveal any sequences homologous to the other ORFs except for dacA, which exhibited significant similarity to DacB of B. subtilis (D-alanyl-D-alanine-carboxypeptidase, a penicillin-binding protein).

Identification of the nar operon
There are two putative rho-independent transcriptional terminators immediately upstream of the narA gene and downstream of the mobB gene, respectively, suggesting that the narA, narB, narC and mobB genes are transcribed as a polycistronic operon. Although a -10 consensus sequence, TATAAT, is present upstream of the narA gene, there is no sequence indicative of a -35 region. To determine the transcriptional-start site of this operon, primer-extension analysis was performed on cells grown in GAM broth with and without KNO3 . As shown in Fig. 3a , NarA activity increased only slightly in the medium without KNO3. On the contrary, the enzyme activity increased markedly in the presence of KNO3, indicating that the enzyme synthesis is induced by KNO3. RNA was prepared from the two cultures at an OD600 of 0·8. No positive band was detected for RNA from cells grown in the absence of KNO3. On the other hand, RNA from cells grown in the presence of KNO3 gave a positive signal, which clearly indicated the transcriptional-start site was A at position +1 (Fig. 3b). Northern-blot analysis was also performed on the same RNA samples using the DNA fragments from the narA, narB, narC and mobB genes as DNA probes. All blots had a band corresponding to approximately 5300 nt for the RNA from the cells grown in the presence of KNO3 but not for that in the absence of KNO3 (Fig. 3c). When Northern-blot analysis was performed using the DNA fragment within the moeA gene, a band was detected at the location corresponding to approximately 4800 nt, being distinct from that of narABCmobB transcripts (data not shown). These results indicate that the four genes are transcribed to give a polycistronic mRNA from the narA promoter.



View larger version (56K):
[in this window]
[in a new window]
 
Fig. 3. Induction of C. perfringens NarA activity, and analyses of the nar operon by primer extension and Northern blotting. (a) Changes in NarA activity during the growth of C. perfringens in the presence and absence of KNO3. Five millilitres of an overnight preculture in GAM broth without KNO 3 was inoculated into 1 l fresh GAM broth with (black symbols) or without (white symbols) 0·2% (w/v) KNO3. Samples (100 ml) were taken at the indicated times. After the cells had been washed, cell lysates were prepared as described in Methods. Growth was monitored by measuring the OD600 ({square}, {blacksquare}). NarA activity ({circ}, {bullet}) was determined as described under Methods. Error bars represent standard deviations. (b) Primer- extension mapping of the 5' end of the narA mRNA. RNA was extracted from C. perfringens PB6K grown in GAM broth with or without 0·2% KNO3 when cultures reached an OD 600 of 0·8. Sequencing ladders generated with the oligonucleotide used for the primer extension were loaded between the two reaction mixtures. Lane 1, RNA from cultures grown in the presence of KNO3; lane 2, RNA from cultures grown in the absence of KNO3. The DNA sequence of the sense strand around the first nucleotide in the transcript (designated +1) is shown and the -10 sequence is boxed. (c) Northern-blot analysis. Panels (i), (ii), (iii) and (iv) are blots with DNA probes specific to the narA, narB, narC and mobB genes, respectively. Lanes 1, markers; lanes 2, RNA from cultures grown in the presence of KNO 3; lanes 3, RNA from cultures grown in the absence of KNO3 .

 
Characterization of the narA, narB and narC gene products
NarA shows higher similarity to NasC (B. subtilis assimilatory NaR) (Poisson P value 3e-77) than NarG (B. subtilis respiratory NaR) (Poisson P value 3e-4). It also shows high similarity to FdhH, E. coli formate dehydrogenase H (Poisson P value 6e-77). A study on the three-dimensional structure of FdhH has well defined the amino acid residues ligating a [4Fe-4S] iron–sulfur cluster and Mo-MGD, a prosthetic group transferring an electron, as well as the catalytic site. The alignment of C. perfringens NarA with E. coli FdhH revealed that the following amino acid residues are conserved in the former: 15 of the 24 amino acid residues participating in the ligation of two Mo- MGD molecules; Lys44, a residue transferring an electron from Mo-MGD to the [4Fe-4S] iron–sulfur cluster; and a 4-cysteine- residue motif for binding the [4Fe-4S] iron–sulfur cluster, as shown in Fig. 4a.



View larger version (66K):
[in this window]
[in a new window]
 
Fig. 4. (a) Comparison of the deduced amino acid sequences of C. perfringens NarA and E. coli formate dehydrogenase H. The upper and lower sequences represent NarA and formate dehydrogenase H, respectively. Residue identities are indicated by asterisks and similarities by dots. The residues participating in the ligation of two Mo-MGD molecules are indicated by an open box. The lysine residue transferring an electron from the [4Fe-4S] iron–sulfur cluster to Mo-MGD, and the cysteine residues binding the [4Fe-4S] iron–sulfur cluster are indicated by an arrowhead and black backgrounds, respectively. (b) Alignment of the 4-cysteine motifs involved in ligation of the [4Fe-4S] cluster. The predicted amino acid sequences of NarB from C. perfringens (NarB, this work), DMSO reductase iron–sulfur subunit from E. coli (DmsB, P18776), nitrate reductase 1 beta chain from E. coli (NarH, P11349), fumarate reductase iron–sulfur protein from E. coli (FrdB, P00364) and succinate dehydrogenase iron–sulfur protein from E. coli (SdhB, P07014) are shown, with amino acid residues being numbered on the left. The conserved cysteine residues are indicated by black backgrounds. (c) Alignment of nucleotide-binding motifs. The predicted amino acid sequence of NarC from C. perfringens (NarC, this work) and the known amino acid sequences of NADH peroxidase from Streptococcus faecalis (Npr, P37062) and sulfide dehydrogenase from Chromatium vinosum (Sdh, Q06530) are shown, with amino acid residues being numbered on the left. The residues forming FAD- and NAD- binding domains are indicated by open boxes.

 
Alignment of the deduced amino acid sequences of NarB and homologous proteins showed a characteristic of NarB. Its deduced amino acid sequence includes four cysteine motifs, each of which belongs to a different [4Fe-4S] iron–sulfur cluster (Fig. 4b). This ferredoxin-like protein differs from ferredoxin in that it possesses four [4Fe-4S] iron–sulfur clusters, and also in that it consists of 178 aa residues, this being considerably longer than ferredoxin. It exhibits similarity to the electron-transfer subunits of many respiratory enzymes, which are referred to as ‘four-cluster ferredoxins’, since they possess motifs 1–4 differing in cysteine spacing (Gennis & Stewart, 1996 ), as shown in Fig. 4b.

The amino acid sequence deduced from the narC gene exhibits homology to those of NADH peroxidase (Poisson P value 1·2e-15) and sulfide dehydrogenase (Poisson P value 0·018). Their alignment revealed that NarC contains FAD- and NADH-binding domains, which were detected in the two enzymes on crystallographic analysis (Fig. 4c). These structural characteristics of NarB and NarC, and the linkage of the three genes suggest that they function to reduce nitrate in a cooperative manner such that an electron from NADH is first accepted by NarC, then transferred to NarB and finally to NarA.

Expression of the narB and narC genes
C. perfringens contains ferredoxin (Seki et al., 1979 ) and rubredoxin (Seki et al., 1989 ), both of which may function as an electron carrier for NarA (Seki et al., 1988 , 1989 ). Although it is clear that the narB gene is transcribed from the narA promoter upon induction by KNO3, the possibility that NarB is not synthesized cannot be ruled out. To determine whether or not NarB is produced upon induction, cells were grown in the presence and absence of KNO3, collected at an OD600 of 2·0 and subjected to Western-blot analysis (Fig. 5a ). A band with the same mobility as that of the NarB protein was detected for the cells grown in the presence of KNO3 but not for cells grown in its absence. However, other band(s) were also detected for both the induced and uninduced cells. They may be unknown protein(s) which are immunologically related to NarB or associated with the NarB protein. Synthesis of the NarC protein in induced and uninduced cells was also examined by Western-blot analysis (Fig. 5b). Antiserum was prepared from mice immunized with a GST-NarC fusion protein, because the fusion protein was insensitive to cleavage by thrombin. A band corresponding to a 45 kDa polypeptide, which coincided with the molecular mass predicted from the narC gene, was detectable for the induced cells. A band with the same mobility is weakly present in the uninduced cells. Since low but significant levels of nitrate reductase activity were detectable in the uninduced cultures grown at an OD600 of 2·0 (Fig. 3a), small amounts of NarB and NarC seem to be produced in the uninduced cells. The discrepancy that the NarC protein but not the NarB protein was detectable for the uninduced cells could be due to the difference in the immunoreactivity of the two antisera used. Thus, it seems very likely that the expression of the narB and narC genes at the protein level is coregulated with that of the narA gene.



View larger version (30K):
[in this window]
[in a new window]
 
Fig. 5. Immunoblot analysis of NarB and NarC from C. perfringens cells. Cultures were grown in GAM broth with or without 0·2% KNO3 and collected when the cultures reached an OD600 of 2. Cell lysates were prepared by boiling cell suspensions in SDS-sample buffer and electrophoresed through a 15% (a) or 12·5% (b) acrylamide gel. The proteins were blotted onto a nitrocellulose membrane and probed with anti-NarB (a) or anti-NarC (b) antisera. Lanes: 1, prestained size markers; 2, lysate prepared from KNO3 uninduced cells; 3, lysate prepared from KNO3 induced cells; 4 (a), purified NarB protein (b), purified GST- NarC fusion protein.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study we have cloned and characterized the nar operon of C. perfringens. The results presented here not only support the earlier finding that NarA has a molybdenum cofactor, four non-haem irons and acid-labile sulfides (Seki-Chiba & Ishimoto, 1977 ; Seki et al., 1987 ), but also provide additional information on NarA. The following are the characteristics of NarA suggested by the results of analysis of the nar operon: i) Mo-MGD is a likely cofactor of NarA, because it possesses the Mo-MGD-binding motif of E. coli FdhH and also because the mobB gene, which is involved in Mo-MGD synthesis, is present within the nar operon; ii) the iron–sulfur cluster of NarA seems to be the [4Fe-4S] type, as described above; iii) induction of NarA by KNO3 occurs at the transcriptional level; and iv) NarA is coregulated along with NarB (a ferredoxin-like protein) and NarC (an NADH oxidase). The location of the mobB gene within the nar operon may reflect a functional linkage between MobB and NarA. The mobA and mobB gene products of E. coli are involved in the final step of Mo-MGD biosynthesis, conversion of molybdopterin to Mo-MGD. In the case of NarG, a membrane- bound respiratory NaR, NarJ is also required for insertion of Mo-MGD to its apoenzyme. Recently, MobB was demonstrated to be a GTP-binding protein (Eaves et al., 1997 ) and NarJ was shown to be a specific chaperone allowing the apoenzyme to incorporate Mo-MGD (Blasco et al., 1998 ). In E. coli, the mobA and mobB genes are arranged as a single transcriptional unit, while a narJ gene is located within a narGHIJ operon. C. perfringens MobB seems to be involved in the final step of Mo-MGD biosynthesis like E. coli MobB. It may also play a role in the activation of NarA via incorporation of Mo-MGD like E. coli NarJ. This supposition requires further study on the activation of NarA in C. perfringens .

The organization of the nar operon suggests a functional relationship between NarA, NarB and NarC, i.e. the latter two proteins are linked to NarA. On the other hand, the in vitro reconstitution experiment performed by Seki et al. (1988) strongly suggested that rubredoxin, a 6 kDa iron–sulfur protein (Seki et al., 1989 ), serves as an intermediary electron carrier for NAD(P)H-linked nitrate reduction, and that ferredoxin reduced by pyruvate dehydrogenase is linked to NarA instead of hydrogenase in the presence of nitrate. Their results, however, do not rule out a possible functional relationship between NarA, NarB and NarC. Considering the results of both our genetic analysis and their in vitro study, we assume that an electron-transfer pathway from NADH to NarA via NarC and NarB could be the main route of nitrate reduction, at least upon induction by KNO 3, although ferredoxin and rubredoxin would also serve as intermediary electron carriers to NarA. To prove this, the NarA activity in the presence of NarB and NarC must be compared with that in their system.

Comparison of NaRs and their electron carriers between the three different nitrate-reduction systems would be informative for understanding their evolutionary relationships. B. subtilis possesses both respiratory and assimilatory NaRs, which are encoded by narG (Cruz Ramos et al., 1995 ; Hoffmann et al., 1995 ) and nasC (Ogawa et al. , 1995 ), respectively. The deduced amino acid sequence of C. perfringens NarA exhibits higher similarity to NasC than to NarG, indicating that C. perfringens NarA is more closely related to assimilatory NaRs. NasB, NADH dehydrogenase of B. subtilis, possesses three iron–sulfur clusters, while Klebsiella oxytoca NADH dehydrogenase, which is homologous to B. subtilis NasB, lacks such a cluster (Fig. 6 ). Instead, a region homologous to the clusters resides at the N-terminus of NaR of the latter organism (Fig. 6 ). Based on the homology between these iron–sulfur clusters and TodB, an electron-transfer subunit in the NADH-dependent toluene degradation system of Pseudomonas putida, the current K. oxytoca NaR and B. subtilis NADH-dependent dehydrogenase are postulated to have resulted from fusion with an intermediate electron-carrier protein containing an iron–sulfur cluster (Lin & Stewart, 1998 ). Our finding that NarB is the most likely candidate for an intermediate electron carrier for NarA strengthens the fusion hypothesis, which might be extended to the origin of eukaryotic assimilatory NaRs containing all three components corresponding to NarA, NarB and NarC of C. perfringens (Campbell & Kinghorn, 1990 ).



View larger version (19K):
[in this window]
[in a new window]
 
Fig. 6. Comparison between homologous components involved in assimilatory and fermentative nitrate reduction. Abbreviations: Cp, C. perfringens; Bs, B. subtilis; Ko, K. oxytoca ; NasC and NasB of B. subtilis, NaR and NAD(P) dehydrogenase, respectively; NasA and NasC of K. oxytoca, NaR and NAD(P) dehydrogenase, respectively. A domain or subunit predicted to function as an electron-transfer component is indicated by an arrowhead.

 
The similarity between NarA and assimilatory NaR does not necessarily exclude the notion of Hasan & Hall (1975) that fermentative nitrate reduction might have evolved as the most primitive form of nitrate respiration. Instead, the fact that the ferredoxin-like protein is homologous to the electron-transfer subunits of many respiratory molybdoenzymes, including respiratory NaR, supports their hypothesis. The location of the nirA gene is also of interest in terms of the evolutionary relationship between NaRs. Genes for nitrite reductase are linked to those for NaR in the assimilatory system, but not in the respiratory system. The nirA gene is not present in the 14·4 kb region sequenced in this study. Possibly the nirA gene is not located in the vicinity of the nar genes in C. perfringens, although it cannot be ruled out that the nirA gene may be further downstream, distal to the genes for molybdenum cofactor biosynthesis. Thus, fermentative nitrate reduction may have evolved into both assimilatory and respiratory nitrate-reduction systems. Further detailed study on the nar operon and analysis of the nir gene(s) will facilitate elucidation of the evolutionary path of the fermentative nitrate-reduction system.


   ACKNOWLEDGEMENTS
 
We are indebted to Shigehiko Taniguchi (Fujita Health University School Medicine, Toyoake-shi, Aichi, Japan) for encouraging us throughout this study. We also thank Nicholas J. Halewood for the invaluable assistance in preparation of the manuscript. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aiba, H. , Adhya, S. & de Crombrugghe, B. (1981). Evidence for two functional gal promoters in intact Escherichia coli cells. J Biol Chem 256, 11905-11910 .[Abstract/Free Full Text]

Altschul, S. F. , Madden, T. L. , Schaffer, A. A. , Zhang, J. , Zhang, Z. , Miller, W. & Lipman, D. J. (1997). Gapped BLAST and PSI- BLAST: a new generation of protein database search programs. Nucleic Acids Res 25, 3389-3402 .[Abstract/Free Full Text]

Ba-Thein, W. , Lyristis, M. , Ohtani, K. , Nisbet, I. T. , Hayashi, H. , Rood, J. I. & Shimizu, T. (1996). The virR/virS locus regulates the transcription of genes encoding extracellular toxin production in Clostridium perfringens. J Bacteriol 178, 2514-2520 .[Abstract]

Berks, B. C. , Ferguson, S. J. , Moir, J. W. & Richardson, D. J. (1995). Enzymes and associated electron transport systems that catalyse the respiratory reduction of nitrogen oxides and oxyanions. Biochim Biophys Acta 1232, 97-173.[Medline]

Blasco, F. , Dos Santos, J. P. , Magalon, A. , Frixon, C. , Guigliarelli, B. , Santini, C. L. & Giordano, G. (1998). NarJ is a specific chaperone required for molybdenum cofactor assembly in nitrate reductase A of Escherichia coli. Mol Microbiol 28, 435-447.[Medline]

Calmels, S. , Ohshima, H. , Henry, Y. & Bartsch, H. (1996). Characterization of bacterial cytochrome cd(1)-nitrite reductase as one enzyme responsible for catalysis of nitrosation of secondary amines. Carcinogenesis 17, 533-536.[Abstract]

Campbell, W. H. (1996). Nitrate reductase biochemistry comes of age. Plant Physiol 111, 355-361.[Free Full Text]

Campbell, W. H. & Kinghorn, J. R. (1990). Functional domains of assimilatory nitrate reductases and nitrite reductases. Trends Biochem Sci 15, 315-319.[Medline]

Crawford, N. M. (1995). Nitrate: nutrient and signal for plant growth. Plant Cell 7, 859-868.[Free Full Text]

Crawford, N. M. & Arst, H. N.Jr (1993). The molecular genetics of nitrate assimilation in fungi and plants. Annu Rev Genet 27, 115-146.[Medline]

Cruz Ramos, H. , Boursier, L. , Moszer, I. , Kunst, F. , Danchin, A. & Glaser, P. (1995). Anaerobic transcription activation in Bacillus subtilis: identification of distinct FNR-dependent and -independent regulatory mechanisms. EMBO J 14, 5984-5994 .[Abstract]

Eaves, D. J. , Palmer, T. & Boxer, D. H. (1997). The product of the molybdenum cofactor gene mobB of Escherichia coli is a GTP-binding protein. Eur J Biochem 246, 690-697.[Abstract]

Gennis, R. B. & Stewart, V. (1996). Respiration. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn, pp. 217–261. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Hasan, S. M. & Hall, J. B. (1975). The physiological function of nitrate reduction in Clostridium perfringens. J Gen Microbiol 87, 120-128.[Medline]

Hoffmann, T. , Troup, B. , Szabo, A. , Hungerer, C. & Jahn, D. (1995). The anaerobic life of Bacillus subtilis: cloning of the genes encoding the respiratory nitrate reductase system. FEMS Microbiol Lett 131, 219-225.[Medline]

Ishimoto, M. , Umeyama, M. & Chiba, S. (1974). Alteration of fermentation products from butyrate to acetate by nitrate reduction in Clostridium perfringens. Z Allg Mikrobiol 14, 115-121.[Medline]

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.[Medline]

Lin, J. T. & Stewart, V. (1998). Nitrate assimilation by bacteria. Adv Microb Physiol 39, 1-30.[Medline]

Matsushita, O. , Yoshihara, K. , Katayama, S.-I. , Minami, J. & Okabe, A. (1994). Purification and characterization of a Clostridium perfringens 120-kilodalton collagenase and nucleotide sequence of the corresponding gene. J Bacteriol 176, 149-156.[Abstract]

Matsushita, O. , Jung, C.-M. , Katayama, S. , Minami, J. , Takahashi, Y. & Okabe, A. (1999). Gene duplication and multiplicity of collagenases in Clostridium histolyticum. J Bacteriol 181, 923-933.[Abstract/Free Full Text]

Messing, J. (1983). New M13 vectors for cloning. Methods Enzymol 101, 20-78.[Medline]

Ogawa, K. , Akagawa, E. , Yamane, K. , Sun, Z. W. , LaCelle, M. , Zuber, P. & Nakano, M. M. (1995). The nasB operon and nasA gene are required for nitrate/nitrite assimilation in Bacillus subtilis. J Bacteriol 177, 1409-1413 .[Abstract]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sanger, F. , Nicklen, S. & Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74, 5463-5467 .[Abstract]

Seki, S. , Hagiwara, M. , Kudo, K. & Ishimoto, M. (1979). Studies on nitrate reductase of Clostridium perfringens. II. Purification and some properties of ferredoxin. J Biochem 85, 833-838.[Abstract]

Seki, S. , Hattori, Y. , Hasegawa, T. , Haraguchi, H. & Ishimoto, M. (1987). Studies on nitrate reductase of Clostridium perfringens. IV. Identification of metals, molybdenum cofactor, and iron–sulfur cluster. J Biochem 101, 503-509.[Abstract]

Seki, S. , Ikeda, A. & Ishimoto, M. (1988). Rubredoxin as an intermediary electron carrier for nitrate reduction by NAD(P)H in Clostridium perfringens. J Biochem 103, 583-584.[Abstract]

Seki, Y. , Seki, S. , Satoh, M. , Ikeda, A. & Ishimoto, M. (1989). Rubredoxin from Clostridium perfringens: complete amino acid sequence and participation in nitrate reduction. J Biochem 106, 336-341.[Abstract]

Seki-Chiba, S. & Ishimoto, M. (1977). Studies on nitrate reductase of Clostridium perfringens. I. Purification, some properties, and effect of tungstate on its formation. J Biochem 82, 1663-1671 .[Abstract]

Sekiguchi, S. , Seki, S. & Ishimoto, M. (1983). Purification and some properties of nitrite reductase from Clostridium perfringens. J Biochem 94, 1053-1059 .[Abstract]

Soda, S. , Yamamoto, A. , Ito, A. & Murata, R. (1968). On the maintenance of toxigenicity of seed cultures of Clostridium perfringens PB6K with special reference to alpha toxin. Jpn J Med Sci Biol 21, 91-94.[Medline]

Takahashi, H. , Taniguchi, S. & Egami, F. (1963). Inorganic nitrogen compounds: distribution and metabolism. In Comparative Biochemistry , pp. 91-202. Edited by M. Florkin & H. S. Mason. New York: Academic Press.

Vermeer, I. T. , Pachen, D. M. , Dallinga, J. W. , Kleinjans, J. C. & van Maanen, J. M. S. (1998). Volatile N-nitrosamine formation after intake of nitrate at the ADI level in combination with an amine-rich diet. Environ Health Perspect 106, 459-463.

Völkl, P. , Huber, R. , Drobner, E. , Rachel, R. , Burggraf, S. , Trincone, A. & Stetter, K. O. (1993). Pyrobaculum aerophilum sp. nov., a novel nitrate-reducing hyperthermophilic archaeum. Appl Environ Microbiol 59, 2918-2926 .[Abstract]

Zumft, W. G. (1992). The denitrifying prokaryotes. In The Prokaryotes, 2nd edn, pp. 554–582. Edited by A. Balows & others. Berlin: Springer-Verlag .

Received 8 April 1999; revised 13 August 1999; accepted 27 August 1999.