Oxidative stress response in Clostridium perfringens

Delphine Jean, Valérie Briolat and Gilles Reysset

Unité de Génétique Moléculaire Bactérienne, Institut Pasteur, 28 rue du docteur Roux, 75724 Paris Cedex 15, France

Correspondence
Gilles Reysset
greysset{at}pasteur.fr


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Clostridium perfringens, a strictly anaerobic bacterium, is able to survive when exposed to oxygen for short periods of time and exhibits a complex adaptive response to reactive oxygen species, both under aerobic and anaerobic conditions. However, this adaptive response is not completely understood. C. perfringens possesses specialized genes that might be involved in this adaptive process, such as those encoding superoxide dismutase (SOD), superoxide reductase and alkyl hydroperoxide reductase, but their contribution to the oxidative stress response and their control mechanisms are unknown. By a combination of functional complementation of Escherichia coli strains impaired in either SOD, alkyl hydroperoxide reductase (AhpC) or catalase activity (Cat), transcription analysis and characterization of mutants impaired in regulatory genes, it was concluded that: (i) the product of the sod gene is certainly essential to scavenge superoxide radicals, (ii) the ahpC gene, which is fully induced in all oxidative stress conditions, is probably involved in the scavenging of all intracellular peroxides, (iii) the three rubrerythrin (rbr) genes of C. perfringens do not encode proteins with in vivo H2O2 reductase activity, and (iv) the two rubredoxin (rub) genes do not contribute to the hypothetical superoxide reductase activity, but are likely to belong to an electron transfer chain involved in energy metabolism.


Abbreviations: Ap, ampicillin; Cm, chloramphenicol; Em, erythromycin; Tc, tetracycline; CRP, cAMP-binding protein; Pl, plumbagin; PQ, paraquat; SOD, superoxide dismutase; t-BT, ter-butylhydroperoxide


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Clostridium perfringens is a ubiquitous, Gram-positive, oligosporogenous pathogen, present in the air, soil and water and in the intestinal tracts of humans and animals. Although C. perfringens is strictly anaerobic, it is considered to be an aerotolerant anaerobe. Indeed, vegetative and stationary-phase cells can survive in a growth-arrested state in the presence of oxygen and/or low concentrations of superoxide- or hydroxyl-radical-generating compounds (Trinh et al., 2000; Briolat & Reysset, 2002). Like many other bacteria, C. perfringens exhibits a typical adaptive response to oxidative stresses: cells treated with a sublethal dose of a stress agent can subsequently withstand higher doses of the same agent or products generating identical reactive oxygen species (for a review see Sies, 1986). These properties may explain why C. perfringens can be easily found in and recovered from environmental samples in a wide variety of biotopes. This aerotolerance and the fact that vegetative cells are not rapidly killed by the main reactive oxygen species might also explain how C. perfringens survives during the initial stage of the infectious process. O'Brien & Melville (2000) have shown that although human polymorphonuclear leukocytes and monocytes are able to phagocytose and kill C. perfringens cells during the early stages of infection, some bacteria are phagocytosed by macrophages, but can survive in the cytoplasm. This suggests that C. perfringens cells are protected from the oxidative burst of the macrophages. According to Voyich et al. (2003), upregulation of oxidative stress genes, virulence genes, cell-wall biosynthesis genes and related regulatory genes is observed when group A streptococci evade polymorphonuclear leukocyte phagocytosis. The genes encoding the glutathione peroxidase, alkyl hydroperoxide reductase (etc.) proteins, which detoxify cell-damaging reactive oxygen species, and the genes encoding proteins involved in DNA repair are fully induced 180 min after phagocytosis. By analogy, C. perfringens within macrophages probably elicit such protective responses.

The genes involved in the oxidative stress response of anaerobic Gram-positive bacteria have not all been identified. However, rubrerythrin (rbr) and superoxide dismutase (sod) genes have been identified in C. perfringens strain NCIMB8875 (Lehmann et al., 1996; Geissmann et al., 1999), but it was shown that neither sod nor rbr transcription was influenced by oxidative stress, and these results did not lead to any conclusions about the role of the two genes in the defence against oxygen in C. perfringens. In parallel, studies of sulfate-reducing Desulfovibrio species have recently given new insight in the area of stress resistance in anaerobic bacteria (Pianzzola et al., 1996; Lumppio et al., 1997; Voordouw & Voordouw, 1998; Coulter et al., 1999; Jenney et al., 1999; Lumppio et al., 2001; Emerson et al., 2002). Homologues of some of the oxidative stress response genes found in Desulfovibrio vulgaris have thus been sought in C. perfringens. Analysis of the genome sequence of strain 13 shows that C. perfringens possesses genes potentially encoding (i) a classical Mn-SOD (sod), (ii) an alkyl hydroperoxide reductase (ahpC), (iii) two rubredoxins (rub) suspected to belong to two superoxide reductase systems and (iv) three non-haem iron rubrerythrins (rbr), with hypothetical H2O2 reductase activity.

We used various strategies to evaluate the role of these genes in the oxidative stress response of C. perfringens. First, we carried out functional complementation experiments in Escherichia coli strains impaired in either SOD, alkyl hydroperoxide reductase or catalase activity to evaluate the catalytic properties of these proteins. Second, we compared the level of transcription of these genes in stressed and unstressed cells of C. perfringens. Finally, we studied the physiological properties of regulatory mutants constructed by gene inactivation, to confirm whether inactivated genes were indeed involved in the oxidative stress response.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
The bacterial strains and plasmids used in this study are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids used in this study

 
Media and growth conditions.
E. coli strains were routinely grown aerobically at 37 °C in Luria–Bertani (LB) broth containing ampicillin (Ap, 100 µg ml–1), chloramphenicol (Cm, 25 µg ml–1) or erythromycin (Em, 300 µg ml–1) when needed. Physiological assays were carried out in M63 minimal medium (Miller, 1972) or in this medium supplemented with 5 g Casamino acids l–1 (M63casa medium). C. perfringens strains were routinely grown and stored under anaerobic conditions in TGY medium (Privitera et al., 1979), and in MCP or MMP media (Trinh et al., 2000) for oxidative stress challenges. The media were supplemented with Cm (25 µg ml–1) or Em (25 µg ml–1) when required. Anaerobic cultures were incubated at 37 °C in an anaerobic chamber (Jacomex) in a 90 % N2, 5 % CO2 and 5 % H2 atmosphere. Growth was assessed by measuring the optical density at 440 or 520 nm. For cell counts, aliquots from liquid cultures were serially diluted in anaerobically prepared 90 mM potassium phosphate buffer (PB, pH 7·0) and plated onto pre-reduced MCP (containing 0·5 % cysteine/HCl). Colony-forming-units (c.f.u.) were counted after 24–48 h at 37 °C. The number of c.f.u. obtained was not affected by prior incubation for more than 6 h in PB. Sensitivity to air and to other oxidative stress conditions was described previously (Briolat & Reysset, 2002).

Kinetics of C. perfringens cell survival under aerobic conditions.
Bacteria were grown to mid-exponential phase in MMP medium supplemented with 0·01 % yeast extract at 37 °C in an anaerobic chamber. The bacterial cultures supplemented without and with 2·5 µM plumbagin (Pl) were then transferred to aerobic conditions at 37 °C with shaking in a rotary incubator (250 r.p.m.), in flasks with a capacity 5–10 times greater than the volume of the culture. At various times, aliquots were returned to the anaerobic chamber and viability was measured. The rate of cell survival was calculated as the number of viable cells after exposure to air divided by the total number of viable cells before treatment.

Inhibition of E. coli growth with paraquat (PQ), H2O2 and ter-butylhydroperoxide (t-BT) on solid medium.
E. coli cells were grown overnight in M63casa liquid medium and then spread onto M63casa solid medium, with or without Ap (50 µg ml–1). Then, 5 µl of 25, 50 or 25 mM PQ, H2O2 or t-BT solutions, respectively, was spotted onto small circles (diam. 6 mm) of blotting paper that had been placed on the agar plate. The plates were then incubated for 24 h at 37 °C and the diameter of inhibition was measured. Means of four to eight inhibition zones were determined for each strain.

Peroxidase activity gels.
Peroxidase activity was detected in native polyacrylamide gels as described by Heym & Cole (1992). A crude extract of E. coli cells was prepared as follows. The strains were cultivated in LB medium until stationary phase, harvested by centrifugation and washed twice in 50 mM PB buffer, pH 7·0. The cells were then resuspended in 1/25 of the initial volume of PB buffer and disrupted by sonication (3x3 min with 5 min cooling intervals). Cell debris was removed by centrifugation (20 min at 15 000 g) and the supernatant was stored frozen in aliquots. Proteins were quantified using the method of Bradford (1976).

Molecular and genetic techniques.
Unless otherwise stated, all DNA manipulations were performed by standard techniques (Sambrook et al., 1989). DNA preparation and C. perfringens electroporation were done as described by Briolat & Reysset (2002).

Cloning of genes or genetic regions of C. perfringens.
The genes encoding the alkyl hydroperoxide reductase (ahpC), the SOD (sod), the rubrerythrins (rbr1, rbr2 and rbr3), the rubredoxins (rub1 and rub2) and the genetic regions GRI (from CPE0776 to CPE0778) and GRII (from CPE0779 to CPE0781) (see Fig. 1) were amplified by PCR using specific sets of primers (Table 2). After purification, the PCR products were cloned into the pGEMTeasy vector (Promega) and introduced into E. coli DH5{alpha} by electroporation. The multiple cloning sequence of the pGEMTeasy vector is flanked by promoters recognized by the T7 and SP6 RNA polymerases, respectively. However, as the two polymerases are both absent from the E. coli strains used in this study, the C. perfringens genes are likely to be expressed from their own promoter.



View larger version (11K):
[in this window]
[in a new window]
 
Fig. 1. Genetic organization of the chromosomal region containing the two rubredoxin (rub) genes of C. perfringens. The GRI and GRII genetic regions are indicated. The putative location of the CRP boxes are indicated by arrows.

 

View this table:
[in this window]
[in a new window]
 
Table 2. Primers used to amplify the genes or genetic regions of interest

 
Gene inactivation.
The CPE0776 and CPE0779 genes were disrupted by single crossover insertion of pKNT21DJ776 and pKNT21DJ779, respectively. The CmR transformants obtained after electroporation of the 13R strain were reisolated and total DNA was prepared. PCR and Southern hybridization using pUC19 as a probe validated the disruption of the two genes.

Preparation and quantification of RNA.
C. perfringens (50 ml) was harvested 0, 20, 40 and 60 min after the onset of each stress condition and total RNA was extracted using the trizol reagent (Gibco-BRL) (Dupuy & Sonenshein, 1998). RNA concentration was determined by the orcinol reaction (Schneider, 1957). Various concentrations (5, 2·5, 1 and 0·5 µg) of the denatured DNA-free RNA preparations from each time point were blotted onto two nylon filter membranes. The filters were subjected to Northern hybridization with an internal 106 d.p.m. [{alpha}-32P]dCTP (Amersham) radiolabelled probe of the gene of interest and 106 d.p.m. of a specific C. perfringens 16S RNA probe. Filters were hybridized overnight at 55 °C in rapid-hyb buffer (Amersham) and then washed once for 15 min in 1x SSC (0·15 M NaCl plus 0·015 M sodium citrate)/0·1 % SDS, and three times for 10 min each in 0·2x SSC/0·1 % SDS at 55 °C. Radioactivity was quantified using the Storm Phospho-Imager (Molecular Dynamics). Standardization was done by dividing the value obtained for a given probe by the corresponding 16S rRNA value. The results are expressed as the means of the various RNA concentrations used.

Computer methods.
The whole-genome sequence of C. perfringens was visualized and analysed using Artemis software (Rutherford et al., 2000). C. perfringens gene sequences were compared with sequences present in public databases using FASTA, BLASTN and BLASTX (Altschul et al., 1994). Protein and regulatory or consensus sequences were determined using the MOTIF programs of the Staden package (Staden et al., 2003).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In silico identification of C. perfringens genes possibly involved in the scavenging of H2O2, organic peroxide and superoxide radicals
The publication of the complete sequence of C. perfringens strain 13 by Shimizu et al. (2002) allowed us to identify genes that might scavenge H2O2, t-BT and/or superoxide radicals. Some of them were selected because they encoded proteins with similar sequences to oxidative stress proteins from anaerobic bacteria, particularly Desulfovibrio species, which have been shown to participate in the oxidative stress response. These included:

(i) The sod gene potentially encoding a SOD showing extensive similarity to amino acid sequences typically conserved in Mn-SOD from both Gram-positive and Gram-negative bacteria. An orthologous SOD gene has previously been cloned from C. perfringens strain NCIMB8875 by functional complementation of an E. coli strain lacking superoxide activity (Geissmann et al., 1999).

(ii) The rbr1, rbr2 and rbr3 genes, potentially encoding three rubrerythrin proteins. These proteins, which are 39 to 48 % identical, are closely related to the well characterized rubrerythrin of Desulfovibrio vulgaris (GenBank accession no. AAB39991-1) and to other rubrerythrins from strictly anaerobic bacteria and archaea. The rubrerythrin and related nigerythrin protein from the anaerobic sulphate-reducing bacterium D. vulgaris are suspected to possess NADH peroxidase activity at least in an in vitro assay system (Coulter et al., 1999).

(iii) The rub1 and rub2 genes, both of which encode highly conserved rubredoxin proteins. Such non-haem iron proteins have been found in numerous species of anaerobic eubacteria and archaea. These two rub genes, which encode polypeptides of 53 and 54 aa, respectively, are located in the same chromosomal region of C. perfringens strain 13 and share a similar genetic organization. Both genes are associated with a gene encoding a flavoprotein and a gene that encodes a product displaying similarity with transcriptional regulators. The rub1 gene is located 664 bp upstream of CPE0778, which encodes a probable flavohaemoprotein (haemoglobin-like protein) sharing similarity to other proteins possessing a probable NADH : rubredoxin-oxidoreductase (Rbo) activity. Likewise, rub2 forms an operon with the CPE0781 gene, encoding a hypothetical flavoprotein. The product shares 29 % identity and 47 % similarity with the rubredoxin-oxygen-oxidoreductase (Roo) from Desulfovibrio gigas, which is believed to catalyse the reduction of dioxygen to water (Gomes et al., 1997; Kurtz & Coulter, 2002). Two putative regulatory genes (CPE0776 and CPE0779) are located upstream of two rubredoxin-encoding genes. The two corresponding proteins, share 76 % identity at the amino acid level and belong to the cAMP-binding protein (CRP)/FNR superfamily of transcriptional factors which includes the CRP, FNR and FLP (FNR-like protein) regulatory proteins (Green et al., 2001). The genetic region including CPE0776, rub1 and CPE0778 is designated GRI and that including CPE0779, rub2 and CPE0781 is designated GRII (Fig. 1).

The sequence data also show that the rub2-CPE0781 operon is separated by 208 bp from the ahpC gene of C. perfringens. Associated with the specific electron transporter (AhpF), AhpC is suspected of being the main scavenger of both endogenous alkyl peroxide and H2O2 in E. coli (Seaver & Imlay, 2001). No clear orthologue of the ahpF gene was found in the entire C. perfringens genome. However, a BLAST search identified three genes (CPE0783, CPE2440, CPE2354) sharing about 30 % identity with ahpF, but annotated as probable thioredoxin reductase proteins. Such similarity with a family of thioredoxin reductase proteins has already been pinpointed for the AhpF polypeptide of Salmonella typhimurium (Tartaglia et al., 1990). Notice that one of these genes (CPE0783) is located just downstream (247 bp) of the ahpC gene of C. perfringens (Fig. 1).

Functional complementation of a {Delta}sodA {Delta}sodB E. coli strain
To evaluate the superoxide scavenging activity of the sod, rub and rbr genes of C. perfringens, the recombinant plasmids carrying these genes were introduced into E. coli strain QC1799 deficient in both sodA and sodB genes, encoding manganese- and iron-SODs, respectively. This strain was very sensitive to superoxide-generating compounds (e.g. PQ, Pl) and unable to grow in minimal medium in the absence of amino acids. This last property was due to superoxide damage of iron–sulphur enzymes in the biosynthetic pathway of branched-chain amino acids (Touati, 2000). The diameter of growth inhibition of the transformants with 25 mM PQ revealed that all the C. perfringens genes tested in strain QC1799 provided partial (rbr2, rbr3, GRI and sod) or complete (rbr1 and GrII) protection against this agent (Fig. 2). These results were in good agreement with those obtained in liquid minimal medium (Fig. 3). Plasmids carrying the rbr1 gene and the GRII region of C. perfringens fully restored the aerobic growth of strain QC1799, whereas the expression of sod, rbr2 and rbr3 genes or the entire GRI region were less efficient.



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 2. Diameter of growth inhibition with 25 mM PQ of E. coli QC1799 ({Delta}sodA {Delta}sodB) complemented with the sod, rbr, rub genes of C. perfringens. The mean inhibitory diameter was measured after 24 h on M63 solid medium supplemented with Casamino acids. The wild-type E. coli QC1301 and QC2481 ({Delta}ahp) strains were used as controls.

 


View larger version (22K):
[in this window]
[in a new window]
 
Fig. 3. Growth of E. coli strain QC1799 ({Delta}sodA {Delta}sodB) complemented with the sod, rbr, rub genes from C. perfringens in minimal M63 liquid medium. The cells were pre-grown overnight aerobically in M63 supplemented with amino acids at 37 °C, and then diluted at 1/100 (v/v) in pre-warmed minimal M63 medium. The wild-type E. coli QC1301 and QC1799 ({Delta}sodA {Delta}sodB) strains were used as positive and negative controls, respectively. (a) Growth of QC1799 complemented with the pGRI (rub1), pGRII (rub2) and sod genes of C. perfringens; (b) growth of QC1799 complemented with the three rubrerythrin (rbr) genes of C. perfringens; (c) growth curves of QC1799 complemented with the pGRI and pGRII genetic regions deleted or not of their regulatory genes CPE0776 and CPE0779, respectively.

 
To evaluate the potential transcriptional regulatory properties of CPE0776 and CPE0779 polypeptides, the corresponding genes were deleted from the GRI and GRII regions, generating pGRI{Delta}CPE0776 and pGRII{Delta}CPE0779, respectively. These plasmids were introduced into strain QC1799 and the transformants were assayed. Inactivation of the CPE0776 gene completely abolished the ability of the {Delta}sodA {Delta}sodB E. coli strain to grow in minimal medium and did not enhance resistance to PQ (Figs 2 and 3c). These results suggested that rub1 and/or CPE0778 genes were no longer expressed by pGRI{Delta}CPE0776. In contrast, inactivation of CPE0779 from the GRII region only slowed down the growth rate of the transformants in minimal medium and only partially protected the cells from PQ. The results suggested that CPE0779 can activate the transcription of the rub2-CPE0781 operon, but that the operon can also be transcribed, albeit less efficiently, from a promoter not regulated by the CPE0779 protein. Moreover, although the amino acid sequences of CPE0776 and CPE0779 are 76 % identical, the two regulatory proteins are very specific. When cloned into low-copy-number plasmid pHSG576, CPE0779, but not CPE0776, restored the resistance to PQ of QC1799(pGRII{Delta}CPE0779) in minimal medium (diameter of inhibition=15·0±1·15 vs 36·5±2·89 mm for 25 mM PQ). These results strongly suggested that CPE0776 and CPE0779 are positive regulatory proteins. Based on the protein motif (COG0664), CPE0776 and CPE0779 are more specifically related to the CRP family, and possessed the specific CRP DNA-binding motif RE(T)(V)(S)R located at the C-terminal end of the proteins. Two putative consensus CRP binding sites containing the imperfect palindrome TGTGAN6TCACA (see Green et al., 2001) were identified upstream of the two rubredoxin genes. Our working hypothesis is that CPE0776 and CPE0779 belong to the CRP subfamily, which controls the response to glucose starvation in a broad spectrum of bacterial species

Functional complementation of a catalase-deficient E. coli strain
The recombinant plasmids described above were also introduced into E. coli QC2476 devoid of catalase activity. Whatever the constructions used, the inhibition concentrations of H2O2 were identical to that of the E. coli mutant strain in either aerobic or anaerobic conditions (data not shown). However, a peroxidase activity was detected in a protein crude extract of strain QC2476(pGrbr1) (Fig. 4). This suggests that only one of the three rubrerythrin genes (CPE0135) possesses an in vitro peroxidase activity and that this activity is not sufficient to protect in vivo a catalase-negative strain of E. coli from the toxic action of H2O2.



View larger version (86K):
[in this window]
[in a new window]
 
Fig. 4. Peroxidase activity in crude protein extracts of strain QC2476 (katE : : Tn10 katG : : Tn10) complemented with the rubrerythrin (rbr) and rubredoxin (rub) genes of C. perfringens. (a) Non-denaturing polyacrylamide gel of crude protein extracts (50 µg) stained with Coomassie blue. (b) Activity staining of a native polyacrylamide gel for peroxidase activity with 50 µg crude protein extract. Lanes: L, molecular mass marker; T+, 25 ng purified KatG from Mycobacterium tuberculosis used as a positive control; 1, QC1301 (wild-type); 2, QC2476(pGEMT); 3, QC2476(pGrbr3); 4, QC2476(pGrbr1); 5, QC2476(pGrbr2); 6, QC2476(pGRI) (rub1); 7, QC2476(pGRII) (rub2).

 
Functional complementation of a {Delta}ahp E. coli strain with the ahpC gene of C. perfringens
The ahpC gene of C. perfringens was cloned into both pGEMTeasy, a multicopy plasmid, and pACYC184, a low-copy-number vector and the resulting constructs were introduced into E. coli QC2481 ({Delta}ahp). Whatever the plasmid used, the ahpC gene increased resistance to t-BT. The diameter of growth inhibition for 200 mM t-BT was >=50, 12·3 and 14·5 mm for strains QC2481, QC2481(pGahpC) and QC2481(pACYCahpC), respectively, versus <=6 mm for the wild-type strain QC1301. As the alkyl hydroperoxide reductase of E. coli is also active against H2O2, we studied the ability of the AhpC protein of C. perfringens to restore H2O2 resistance. The H2O2 inhibition concentrations were similar for strains QC2481 ({Delta}ahp) and QC2476 (katE : : Tn10 katG : : Tn10) complemented or not complemented with the ahpC gene of C. perfringens (data not shown).

Transcriptional analysis of the sod, rub, rbr and ahpC genes of C. perfringens
To evaluate the roles of the sod, rub, rbr and ahpC genes in the oxidative stress response, we monitored the transcriptional level of these genes in stressed C. perfringens cells. Induction (-fold difference) was determined by comparing stressed samples to the t=0 control sample. The cells were grown to mid-exponential phase in MMP medium under anaerobic conditions and then maintained under anaerobic conditions after the addition of subinhibitory H2O2 (50 µM) or t-BT (25 µM) to the medium, or transferred into the normal atmosphere with vigorous shaking with or without Pl (2·5 µM), H2O2 (50 µM) and t-BT (25 µM). Samples were harvested after 0, 20, 40 and 60 min. RNA was then extracted and hybridized either with radiolabelled internal probes corresponding to the various genes or with a 16S rRNA probe used as an internal standard. Some of the data are illustrated in Fig. 5. The main results may be summarized as follows.



View larger version (35K):
[in this window]
[in a new window]
 
Fig. 5. Levels of rub, rbr, sod and ahpC transcripts in C. perfringens under various oxidative stress conditions. Mid-exponential-phase C. perfringens strain 13 cells were maintained under anaerobic (Ana) conditions after addition at t=0 of 50 µM H2O2 or 25 µM t-BT or transferred into normal atmosphere (Air) after addition or not of 2·5 µM Pl, 50 µM H2O2 or 25 µM t-BT. Aliquots were harvested at 0 (white bars), 20 (light grey bars), 40 (dark grey bars) and 60 min (black bars) and total RNA was prepared. Northern blotting was performed with an internal probe for rub1 and CPE0778 from the GRI region, CPE0781 from the GRII region, sod, rbr1, rbu2, rbr3 and ahpC. Results were normalized using a specific 16S rRNA probe as internal standard. The ratios of specific mRNAs/16S rRNAs at t=0 in air were: 0·064±0·013 (rub1 probe); 0·217±0·076 (CPE0778 probe); 0·034±0·01 (CPE0781 probe); 0·287±0·051 (sod probe); 0·035±0·005 (rbr1 probe); 0·137±0·03 (rbr2 probe); 0·367±0·129 (rbr3 probe); 0·070±0·027 (ahpC probe).

 
(a) The rub1 and CPE0778 genes of the GRI region were not induced when C. perfringens cells were shaken in air with or without Pl. In contrast, under the same aerobic conditions, the CPE0781 gene of the GRII region was induced three- to sevenfold, the lowest value being obtained with 2·5 µM Pl. These results suggested that the genes of the GRII region, which are fully activated under aerobic conditions but not activated under anaerobic conditions, participate one way or another in the stress response.

(b) The sod gene was induced three- to fourfold under aerobic conditions when the intracellular superoxide-producing drug Pl was present. The induction was detected 20 min after addition and remained maximal for at least 60 min. This suggested that the sod gene also takes part in the superoxide stress response. When H2O2 or t-BT was added to the medium in both aerobic and anaerobic conditions, the sod gene was not significantly induced. It is nevertheless possible that the basal level of transcription was sufficient to produce superoxide activity in the presence of oxygen, H2O2 or t-BT, as already suggested by Geissmann et al. (1999).

(c) The mRNA induction level of the three rubrerythrin-encoding genes (rbr1, rbr2 and rbr3) was only significant (three- to fourfold) when the cells were shaken in air in the presence of t-BT.

(d) The ahpC gene of C. perfringens was highly induced (up to 50-fold) whatever the aerobic stress conditions used. Under anaerobic conditions, the induction was less important, about eightfold in cells challenged with H2O2 and between 10- and 20-fold in cells challenged with t-BT.

Physiological properties of mutants impaired in the potentially regulatory genes of GRI and GRII of C. perfringens
To confirm the physiological role of the genes studied in the oxidative stress response of C. perfringens, we attempted to create insertional mutants of the genes of interest. Unfortunately, despite numerous experiments, we failed to inactivate the sod gene. However, we successfully disrupted the regulatory genes of the GRI and GRII regions (CPE0776 and CPE0779, respectively). The survival of the CPE0776 : : catP and CPE0779 : : catP mutants was monitored under aerobic conditions with and without Pl. The survival curves were identical to those obtained with the wild-type strain, but the sizes of the colonies were considerably smaller compared to those of strain 13R (data not shown). The results suggested that even if the activity of Rub1 and CPE0778 is not essential for the survival of C. perfringens under aerobic conditions, and that lowering the activity of the CPE0781-rub2 operon was not sufficient to impair survival at least during the time-course (6 h) of the experiment, they might be both required when the cells are maintained for a longer period of time under aerobic conditions.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The functional complementation of E. coli strains lacking SOD showed that the SOD protein, the two rubredoxin/haemoprotein and the three rubrerythrin polypeptides of C. perfringens strain 13 protect the mutant strain from the adverse effect of superoxide to some extent. In contrast, they were unable to protect an E. coli mutant lacking catalase activity from H2O2. However, physiological studies of C. perfringens mutants and mRNA expression levels showed that these results were not sufficient to conclude that these genes are definitely involved in the oxidative stress response of this species.

Geissmann et al. (1999) have shown that the sod mRNA level of C. perfringens strain NCIMB8875 increases during exponential growth and peaks upon entry into stationary phase, independently of whether C. perfringens is grown under anaerobic or partially aerobic conditions. In the more drastic stress conditions used in this study, exposure to oxygen induced the sod gene by a maximum of threefold. Due to the rather high level of SOD still present under anaerobic conditions, this response is undoubtedly sufficient to protect cells against superoxide generated under aerobic conditions. The fact that we failed to generate a sod mutant of C. perfringens strain 13R suggests that the sod gene is essential for survival. However, superoxide-deficient mutants of aerobic organisms [e.g. E. coli (Carlioz & Touati, 1986), Candida albicans (Hwang et al., 2002), Streptococcus agalactiae (Poyart et al., 2001)], or of microaerophilic bacteria exemplified by Helicobacter pylori (Seyler et al., 2001), have been readily obtained. These mutants are generally more sensitive to oxygen or to oxidative stress than the wild-type and display an increased susceptibility to bacterial killing by macrophages. No such mutants have been obtained in strictly anaerobic Gram-positive bacteria, but sod mutants have been reported in Porphyromonas gingivalis (Nakayama, 1994; Lynch & Kuramitsu, 1999).

C. perfringens 13R possesses three genes encoding a non-haem iron protein belonging to the rubrerythrin family. Some members of this family possess NADH peroxidase activity in an in vitro system containing NADH and H2O2, and a bacterial NADH oxidoreductase (Coulter et al., 1999). However, the specific activity is very low and the activity detected in vitro may be different from that expressed in vivo. As some of these proteins can complement E. coli strains devoid of SOD, they may possess SOD-like activity (Lehmann et al., 1996). However, according to Lumppio et al. (1997), this activity was one or two orders of magnitude lower than that reported for classical iron SOD. They thus suggested that these proteins protect cells against oxidative stress by both scavenging iron from and donating iron to oxidatively damaged [4Fe–4S] proteins. The proteins may also have peroxidase activity. A mutant of P. gingivalis, an obligate anaerobe, devoid of rubrerythrin-like proteins, was more sensitive to dioxygen and H2O2 than the wild-type strain (Sztukowska et al., 2002). Our data indicated that the three rubrerythrin genes were only significantly induced when the cultures were shaken in air in the presence of t-BT. Thus, a possible role of the rubrerythrin proteins in the redox reaction required to eliminate organic peroxides under aerobic conditions is not excluded, but an efficient hydrogen peroxidase activity of these proteins in vivo remains questionable. This latter hypothesis was also suggested by Geissmann et al. (1999), who showed that in C. perfringens NCIMB8875, rubrerythrin (rbr) gene transcription was not influenced by oxidative stress.

The C. perfringens genes or operons able to restore the growth of a sodA sodB E. coli mutant on minimal medium included the two rubredoxin genes associated with flavoprotein- or flavohaemoprotein-encoding genes (CPE0778 and CPE0781, respectively). The rbo genes from Desulfovibrio species have the same properties (Pianzzola et al., 1996). It has been proposed that the Rbo protein catalyses the reduction of the superoxide anion to H2O2 at the expense of cellular reducing agents such NAD(P)H (Liochev & Fridovich, 1997; Emerson et al., 2002; Kurtz & Coulter, 2002). It has also been proposed that the flavohaemoprotein is a rubredoxin-oxygen oxidoreductase (Roo), the final component of a soluble electron transfer chain that produces energy by recycling NAD(P)+ from carbon reserves in an aerobic environment (Santos et al., 1993). The fact that in C. perfringens these genes might be regulated by a CRP-like protein, which controls the response to glucose starvation, and that the size of the colonies of the mutants impaired in these regulatory genes is smaller than those of the wild-type strain when plated after oxidative stress, favour the latter hypothesis. Moreover, orthologues of putative NADH : rubredoxin-oxidoreductase and rubredoxin-oxygen-oxidoreductase genes are widespread in anaerobic micro-organisms, even though the genetic organization may be different (Das et al., 2001; Lumppio et al., 2001)

The ahpC gene was rapidly and fully induced when exponentially growing wild-type C. perfringens was exposed to air or to various oxidative stress under anaerobic conditions. As a general rule, anaerobic bacteria do not possess catalase activity. It is thus possible that in strict anaerobic bacteria AhpC plays a pivotal role in scavenging the various peroxide species, and accumulates the activity of the two major scavenging enzymes, alkyl hydroperoxide reductase and catalase found in aerobic bacteria (Seaver & Imlay, 2001).

Thus, taken together this specialized enzymic equipment may explain how C. perfringens and other strictly anaerobic bacteria can survive occasional oxidative stress, despite the fact that the key metabolic enzymes required for normal growth are promptly damaged by oxygen or superoxide (Imlay, 2002; Pan & Imlay, 2001).


   ACKNOWLEDGEMENTS
 
We wish to thank Professor Stewart Cole for his support and advice throughout this work and his team for helpful discussions. We also wish to thank Professor Danièle Touati for the gift of the E. coli strains used in the study.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Altschul, S. F., Boguski, M. S., Gish, W. & Wooton, J. C. (1994). Issues in searching molecular sequence databases. Nat Genet 6, 119–129.[Medline]

Azeddoug, H., Hubert, J. & Reysset, G. (1992). Stable inheritance of shuttle vectors based on plasmid pIM13 in a mutant strain of Clostridium acetobutylicum. J Gen Microbiol 138, 1371–1378.[Medline]

Bebien, M., Lagniel, G., Garin, J., Touati, D., Verméglio, A. & Labarre, J. (2002). Involvement of superoxide dismutases in the response of Escherichia coli to selenium oxides. J Bacteriol 184, 1556–1564.[Abstract/Free Full Text]

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

Briolat, V. & Reysset, G. (2002). Identification of the Clostridium perfringens genes involved in the adaptive response to oxidative stress. J Bacteriol 184, 2333–2343.[Abstract/Free Full Text]

Carlioz, A. & Touati, D. (1986). Isolation of superoxide dismutase mutants in Escherichia coli: is superoxide dismutase necessary for aerobic life? EMBO J 5, 623–630.[Abstract]

Coulter, E. D., Shenvi, N. V. & Kurtz, D. M., Jr (1999). NADH peroxidase activity of rubrerythrin. Biochem Biophys Res Commun 255, 317–323.[CrossRef][Medline]

Das, A., Coulter, E. D., Kurtz, D. M., Jr & Ljungdahl, L. G. (2001). Five-gene cluster in Clostridium thermoaceticum consisting of two divergent operons encoding rubredoxin oxidoreductase-rubredoxin and rubrerythrin-type A flavoprotein-high-molecular-weight rubredoxin. J Bacteriol 183, 1560–1567.[Abstract/Free Full Text]

Dupuy, B. & Sonenshein, A. L. (1998). Regulated transcription of Clostridium difficile toxin genes. Mol Microbiol 27, 107–120.[CrossRef][Medline]

Emerson, J. P., Coulter, E. D., Cabelli, D. E., Phillips, R. S. & Kurtz, D. M., Jr (2002). Kinetics and mechanism of superoxide reduction by two-iron superoxide reductase from Desulfovibrio vulgaris. Biochemistry 41, 4348–4357.[CrossRef][Medline]

Geissmann, T. A., Teuber, M. & Meile, L. (1999). Transcriptional analysis of the rubrerythrin and superoxide dismutase genes of Clostridium perfringens. J Bacteriol 181, 7136–7139.[Abstract/Free Full Text]

Gomes, C. M., Silva, G., Oliveira, S., Legall, J., Liu, M.-Y., Antonio, X. V., Rodrigues-Pousada, C. & Teixeira, M. (1997). Studies on the redox centers of the terminal oxidase from Desulfovibrio gigas and evidence for its interaction with rubredoxin. J Biol Chem 272, 22502–22508.[Abstract/Free Full Text]

Green, J., Scott, C. & Guest, J. R. (2001). Functional versatility in the CRP-FNR superfamily of transcription factors: FNR and FLP. Adv Microbiol Physiol 44, 1–34.[Medline]

Hanahan, D. (1983). Studies on transformation of Escherichia coli with plasmids. J Mol Biol 166, 557–580.[Medline]

Hérouart, D., Sigaud, S., Moreau, S., Frendo, P., Touati, D. & Puppo, A. (1996). Cloning and characterization of the katA gene of Rhizobium meliloti encoding a hydrogen peroxide-inducible catalase. J Bacteriol 178, 6802–6809.[Abstract]

Heym, B. & Cole, S. (1992). Isolation and characterisation of isoniazid-resistant mutants of Mycobacterium smegmatis and M. aurum. Res Microbiol 143, 721–730.[CrossRef][Medline]

Hwang, C. S., Rhie, G. E., Oh, J. H., Huh, W. K., Yim, H. S. & Kang, S. O. (2002). Copper- and zinc-containing superoxide dismutase (Cu/ZnSOD) is required for the protection of Candida albicans against oxidative stresses and the expression of its full virulence. Microbiology 148, 3705–3713.[Abstract/Free Full Text]

Imlay, J. A. (2002). How oxygen damages microbes: oxygen tolerance and obligate anaerobiosis. Adv Microbial Physiol 46, 111–153.[Medline]

Jenney, F. E., Jr Verhagen, M. F. J. M., Cui, X. & Adams, M. W. W. (1999). Anaerobic microbes: oxygen detoxification without superoxide dismutase. Science 286, 306–309.[Abstract/Free Full Text]

Kurtz, D. M., Jr & Coulter, E. D. (2002). The mechanism(s) of superoxide reduction by superoxide reductases in vitro and in vivo. J Biological Inorg Chem 7, 653–658.[CrossRef]

Lehmann, Y., Meile, L. & Teuber, M. (1996). Rubrerythrin from Clostridium perfringens – cloning of the gene, purification of the protein, and characterization of its superoxide dismutase function. J Bacteriol 178, 7152–7158.[Abstract]

Liochev, S. I. & Fridovich, I. (1997). A mechanism for complementation of the sodA sodB defect in Escherichia coli by overproduction of the rub gene product (desulfoferrodoxin) from Desulfofoarculus baarsii. J Biol Chem 272, 25573–25575.[Abstract/Free Full Text]

Lumppio, H. L., Shenvi, N. V., Garg, R. P., Summers, A. O. & Kurtz, D. M., Jr (1997). A rubrerythrin operon and nigerythrin gene in Desulfovibrio vulgaris (Hildenborough). J Bacteriol 179, 4607–4615.[Abstract]

Lumppio, H. L., Shenvi, N. V., Summers, A. O., Voordouw, G. & Kurtz, D. M., Jr (2001). Rubrerythrin and rubredoxin oxidoreductase in Desulfovibrio vulgaris: a novel oxidative stress protection system. J Bacteriol 183, 101–108.[Abstract/Free Full Text]

Lynch, M. C. & Kuramitsu, H. K. (1999). Role of superoxide dismutase activity in the physiology of Porphyromonas gingivalis. Infect Immun 67, 3367–3375.[Abstract/Free Full Text]

Mahony, D. E. & Moore, T. I. (1976). Stable L-forms of Clostridium perfringens and their growth on glass surface. Can J Microbiol 22, 953–959.[Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Nakayama, K. (1994). Rapid viability loss on exposure to air in a superoxide dismutase-deficient mutant of Porphyromonas gingivalis. J Bacteriol 176, 1939–1943.[Abstract]

O'Brien, D. K. & Melville, S. B. (2000). The anaerobic pathogen Clostridium perfringens can escape the phagosome of macrophages under aerobic conditions. Cell Microbiol 2, 505–519.[CrossRef][Medline]

Pan, N. & Imlay, J. A. (2001). How does oxygen inhibit central metabolism in the obligate anaerobe Bacteroides thetaiotaomicron. Mol Microbiol 39, 1562–1571.[CrossRef][Medline]

Pianzzola, M. J., Soubes, M. & Touati, D. (1996). Overproduction of the rbo gene product from Desulfovibrio species suppresses all deleterious effects of lack of superoxide dismutase in Escherichia coli. J Bacteriol 178, 6736–6742.[Abstract]

Poyart, C., Pellegrini, E., Gaillot, O., Boumaila, C., Baptista, M. & Trieu-Cuot, P. (2001). Contribution of Mn-cofactored superoxide dismutase (SodA) to the virulence of Streptococcus agalactiae. Infect Immun 69, 5098–5106.[Abstract/Free Full Text]

Privitera, G., Dublanchet, A. & Sebald, M. (1979). Transfer of multiple antibiotic resistance between subspecies of Bacteroides fragilis. J Infect Dis 139, 97–101.[Medline]

Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.-A. & Barrell, B. (2000). Artemis: sequence visualization and annotation. Bioinformatics 16, 944–945.[Abstract]

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

Santos, H., Fareleira, P., Xavier, A. V., Chen, L., Liu, M.-Y. & LeGall, J. (1993). Aerobic metabolism of carbon reserves by the ‘obligate anaerobe’ Desulfovibrio gigas. Biochem Biophys Res Commun 193, 100–105.[CrossRef][Medline]

Schneider, W. C. (1957). Determination of nucleic acids in tissues by pentose analysis. Methods Enzymol 3, 680–684.

Seaver, L. C. & Imlay, A. J. (2001). Alkyl hydroperoxide reductase is the primary scavenger of endogenous hydrogen peroxide in Escherichia coli. J Bacteriol 183, 7173–7181.[Abstract/Free Full Text]

Seyler, R. W., Olson, J. W. & Maier, R. J. (2001). Superoxide dismutase-deficient mutants of Helicobacter pylori are hypersensitive to oxidative stress and defective in host colonization. Infect Immun 69, 4034–4040.[Abstract/Free Full Text]

Shimizu, T., Ohtani, K., Hirakawa, H. & 7 other authors (2002). Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc Natl Acad Sci U S A 99, 996–1001.[Abstract/Free Full Text]

Sies, H. (1986). Biochemistry of oxidative stress. Angew Chem Int Ed Engl 25, 1058–1071.[CrossRef]

Staden, R., Judge, D. P. & Bonfield, J. K. (2003). Analysing sequences using the Staden package and EMBOSS. In Introduction to Bioinformatics. A Theoretical and Practical Approach, pp. 393–410. Edited by S. A. Krawetz & D. D. Womble. Totawa: Humana Press.

Sztukowska, M., Bugno, M., Potempa, J., Travis, J. & Kurtz, D. M. (2002). Role of rubrerythrin in the oxidative stress response of Porphyromonas gingivalis. Mol Microbiol 44, 479–488.[CrossRef][Medline]

Takeshita, S. M., Sato, M., Toba, M., Masahashi, W. & Hashimoto-Gotoh, T. (1987). High-copy-number and low-copy-number plasmid vectors for lacZ alpha-complementation and chloramphenicol- or kanamycin-resistance selection. Gene 61, 63–74.[CrossRef][Medline]

Tartaglia, L. A., Storz, G., Brodsky, M. H., Lai, A. & Ames, B. N. (1990). Alkyl hydroperoxide reductase from Salmonella typhimurium. J Biol Chem 265, 10535–10540.[Abstract/Free Full Text]

Touati, D. (2000). Iron and oxidative stress in bacteria. Arch Biochem Biophys 373, 1–6.[CrossRef][Medline]

Touati, D., Jaques, M., Tardat, B., Bouchard, L. & Despied, S. (1995). Lethal oxidative damage and mutagenesis are generated by iron in {Delta}fur mutants of Escherichia coli: protective role of superoxide dismutase. J Bacteriol 177, 2305–2314.[Abstract]

Trinh, S., Briolat, V. & Reysset, G. (2000). Growth response of Clostridium perfringens to oxidative stress. Anaerobe 6, 233–240.[CrossRef]

Voordouw, J. K. & Voordouw, G. (1998). Deletion of the rbo gene increases the oxygen sensitivity of the sulfate-reducing bacterium Desulfovibrio vulgaris Hildenborough. Appl Environ Microbiol 64, 2882–2887.[Abstract/Free Full Text]

Voyich, J. M., Sturdevant, D. E., Braughton, K. R., Kobayashi, S. D., Lei, B., Virtaneva, K., Dorward, D. W., Musser, J. M. & DeLeo, F. R. (2003). Genome-wide protective response used by group A Streptococcus to evade destruction by human polymorphonuclear leukocytes. Proc Natl Acad Sci U S A 100, 1996–2001.[Abstract/Free Full Text]

Yanisch-Perron, C., Vieira, J. & Messing, J. (1985). Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors. Gene 33, 103–119.[CrossRef][Medline]

Received 6 January 2004; revised 4 March 2004; accepted 19 March 2004.