Role of the thioredoxin system and the thiol-peroxidases Tpx and Bcp in mediating resistance to oxidative and nitrosative stress in Helicobacter pylori

Spencer L. Comtois{dagger}, Mark D. Gidley and David J. Kelly

Department of Molecular Biology & Biotechnology, University of Sheffield, Firth Court, Western Bank, Sheffield S10 2TN, UK

Correspondence
David J. Kelly
d.kelly{at}sheffield.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Helicobacter pylori possesses two distinct thioredoxin proteins (Trx1 and Trx2) which may play important roles in the ability of this bacterium to survive oxidative stress. Trx1 has previously been shown to be an electron donor in vitro for alkyl-hydroperoxide reductase (AhpC), one of three members of the peroxiredoxin family of antioxidant peroxidases present in H. pylori. In this study, mutants in the trxA1 and trxA2 genes encoding Trx1 and Trx2, respectively, and in the tpx and bcp genes, which encode the remaining two members of the H. pylori peroxiredoxin family, were constructed in order to determine their roles in resistance to damage by reactive oxygen and nitrogen species. Mutation of trxA1 led to a pronounced increase in sensitivity to oxygen, hydrogen peroxide and the superoxide generator paraquat, as well as to the nitric oxide (NO) releasers sodium nitroprusside (SNP) and S-nitrosoglutathione (GSNO), consistent with an in vivo role for Trx1 as a reductant for AhpC. A trxA2 single mutant grew normally in an atmosphere of 2 % (v/v) O2 but grew very poorly in 10 % (v/v) O2. It showed slight increases in killing by hydrogen peroxide, paraquat, SNP and GSNO compared to the wild-type, but was significantly more sensitive to cumene hydroperoxide in disc-diffusion assays. A trxA1 trxA2 double mutant was very sensitive to all of the oxidative and nitrosative stresses applied. Comparison of the phenotypes of the tpx and bcp mutants showed that Tpx plays a significant role in peroxide and superoxide resistance in H. pylori, while the role of Bcp is minimal. No evidence was obtained for a role for either Tpx or Bcp in resistance to the toxic effects of NO. The results show that a functional thioredoxin system is necessary for both oxidative and nitrosative stress resistance in H. pylori but, surprisingly, is not essential for viability despite the absence of glutathione and a glutaredoxin system in this bacterium.

Abbreviations: GSNO, S-nitrosoglutathione; SNP, sodium nitroprusside

{dagger}Present address: Microscience Ltd, 545 Eskdale Road, Winnersh Triangle, Wokingham, Berkshire RG41 5TU, UK.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Helicobacter pylori is a ubiquitous human pathogen and is the causative agent of type B gastritis and the majority of cases of duodenal and gastric ulceration (Dixon, 2001). Moreover, long-term infection with the bacterium is a known risk-factor for the development of gastric cancer (Forman et al., 1991). H. pylori is a microaerophilic bacterium that is unable to grow at normal atmospheric oxygen concentrations, and is routinely cultured in gas atmospheres containing 3–10 % (v/v) O2 and 5–10 % (v/v) CO2 (Andersen & Wadström, 2001). The molecular basis of the sensitivity of H. pylori to molecular oxygen is now beginning to be understood. The bacterium is known to possess oxygen-sensitive essential enzymes such as the pyruvate and 2-oxoglutarate oxidoreductases, which contain labile FeS redox centres (Hughes et al., 1995, 1998), and H. pylori also appears to generate larger amounts of superoxide and has a lower specific activity of superoxide dismutase than some other bacteria, e.g. Escherichia coli (Nagata et al., 1998).

Oxidative and nitrosative stress resistance are key properties that enable pathogenic bacteria to survive the effects of the production of reactive oxygen and nitrogen species by the host (Storz & Zheng, 2000). H. pylori seems well equipped to deal with peroxide stress, as it contains an active catalase and a number of peroxidases, including a periplasmic cytochrome c peroxidase (Tomb et al., 1997; Kelly, 1998; Alm et al., 1999). Interestingly, it possesses several enzymes of the peroxiredoxin family which may play key roles. The best studied of these enzymes is alkyl-hydroperoxide reductase (AhpC), which was proposed to be essential for viability (Baker et al., 2001). However, ahpC mutants can be isolated if mutant selection is performed under low oxygen tensions (Olczak et al., 2002). The AhpC enzyme probably has several roles in protecting against oxidative stress by reducing a variety of organic peroxides produced during oxidative damage and, by analogy with E. coli (Costa Seaver & Imlay, 2001), may be a major contributor to hydrogen peroxide disposal at concentrations too low to be dealt with efficiently by catalase. AhpC may also be important in protecting H. pylori against the damaging effects of nitric oxide (NO). NO is a bactericidal agent produced in the host by macrophages and other cells, which can combine with superoxide to form peroxynitrite (ONOO-), an extremely toxic and reactive nitrogen species (Hughes, 1999). H. pylori AhpC can reduce peroxynitrite to the relatively harmless nitrite (Bryk et al., 2000), at least in vitro. At least two other peroxidases are encoded in the genome of H. pylori. One, encoded by HP390 in strain 26695, is a thiol-peroxidase that was originally identified by Wan et al. (1997) as ‘scavengase p20’ and is divergently transcribed from the superoxide dismutase gene (sodB). The other enzyme is encoded by HP0136 in strain 26695 and is a homologue of the E. coli ‘bacterioferritin co-migratory protein’ (Bcp), now known to be a thiol-peroxidase (Jeong et al., 2000).

AhpC, Tpx and Bcp are all members of the peroxiredoxin family (Schröder & Ponting, 1998; Jeong et al., 2000), which use reduced thioredoxin as an electron donor for the catalytic reduction of their respective substrates. H. pylori possesses two distinct thioredoxins, Trx1 and Trx2 (Windle et al., 2000; Baker et al., 2001), encoded by HP0824 and HP1458, respectively, in strain 26695. A thioredoxin reductase is encoded by trxB (HP0825 in strain 26695; Tomb et al., 1997). The H. pylori Trx1 thioredoxin has been characterized biochemically (Windle et al., 2000). Experiments with the purified proteins in vitro have shown that Trx1 (but not Trx2) acts as the electron donor to AhpC (Baker et al., 2001). In view of the fact that AhpC plays a critical role in the H. pylori oxidative stress response, it would be predicted that Trx1 is also of central importance, but this and the possibility that Trx2 or another reductant can substitute in vivo has not been investigated. Indeed, the normal physiological role of Trx2 is unclear, as are the functions of the additional peroxidases known to be encoded in the genome of H. pylori. In this study, we have constructed mutants in the trxA1, trxA2, tpx and bcp genes in order to determine the role of the thioredoxins and thiol-peroxidase proteins of H. pylori in contributing to the resistance of this bacterium to the effects of oxidative and nitrosative stress. Phenotypic analyses of these mutants show that both Trx1 and Trx2 have roles in oxidative and nitrosative stress resistance. tpx mutants were found to be more sensitive to killing by peroxide and superoxide (but not oxygen or NO) compared to the wild-type parent strain, while bcp mutants had a similar but much weaker phenotype.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and growth conditions.
A list of bacterial strains and plasmids used in this study is shown in Table 1. H. pylori was routinely grown on Columbia agar (Oxoid), supplemented with 5 % (v/v) lysed horse blood (hereafter called blood agar). A selective antibiotic supplement of vancomycin, polymyxin B and amphotericin B, each at a final concentration of 10 µg ml-1, was added to all H. pylori growth media. Plates were incubated at 37 °C for 48 h in a Variable Atmosphere Incubator (VAIN; Don Whitley) in an atmosphere of 5 % (v/v) CO2, 85 % (v/v) N2 and 10 % (v/v) O2. For growth experiments at low oxygen concentrations, the gas atmosphere was changed to 5 % (v/v) CO2, 93 % (v/v) N2 and 2 % (v/v) O2. Liquid cultures of H. pylori were grown microaerobically in brain–heart infusion (BHI) broth supplemented with 5 % (v/v) fetal calf serum (BHI/FCS) and the above antibiotics, in 25 or 250 ml media contained in 50 or 500 ml conical flasks shaken orbitally at 120 r.p.m. To select for H. pylori mutants carrying antibiotic-resistant determinants, either kanamycin or chloramphenicol was added to media at a final concentration of 30 µg ml-1. For growth experiments, H. pylori overnight BHI/FCS starter cultures were inoculated into fresh BHI/FCS broth to an initial OD710 value of between 0·1 and 0·2; growth was monitored by regular optical density readings. E. coli strains for plasmid subcloning were grown on Luria–Bertani (LB) agar or in LB broth. Supplements of 100 µg ampicillinml-1, 30 µg chloramphenicolml-1 and 30 µg kanamycinml-1 were used as appropriate for the selection and maintenance of plasmids.


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

 
DNA isolation and manipulation.
Plasmid DNA for screening clones, sequencing and gene disruption experiments was routinely isolated using anion exchange resin spin-columns (Qiagen), according to the manufacturer's instructions. H. pylori total chromosomal DNA was extracted using a modified SDS lysis procedure (Marmur, 1961). Restriction endonucleases, T4 DNA ligase and Taq polymerases were purchased from Promega and Pfu polymerase from Stratagene, and used according to manufacturer's instructions. Standard techniques were used for the cloning, transformation, preparation and restriction analysis of plasmid DNA from E. coli.

Inactivation of H. pylori genes.
A set of PCR primers were designed to amplify the entire coding regions of the trxA1, trxA2, tpx and bcp genes for insertional activation after cloning of the products into pGEM T-easy or pGEM-3-Zf (Promega). The primers used were TrxA1C-F (5'-GCTCGAATTCACGATCCGCTGTTATT-3'), TrxA1C-R (5'-CGAGGAATTCTAACACCGCCTCTAGT-3'), TrxA1K-F (5'-CTGAAGAAGAAGCTTGCGGA-3'), TrxA1K-R (5'-CATCTGGCGCAAGTGCTAAC-3'), TrxA2(a)-F (5'-ATCACTGCAGGCTAATGCGGCTAATG-3'), TrxA2(a)-R (5'-GCACCTGCAGCGCCCTTTATGATACC-3'), TrxA2(b)-F (5'-ACGCAGAGAAAATCGCTCAT-3'), TrxA2(b)-R (5'-GCGTCTTCAATCGGTTTTTG-3'), Bcp-F (5'-GCCGGACAGCTGACATTGAA-3'), Bcp-R (5'-AGCGATCTTAGGGTTGTAGG-3'), Tpx-F (5'-CTCCTTGTGGATCAGATAGCC-3') and Tpx-R (5'-GCCAGAGACTTTGAGCGATA-3'). The TrxA1C and TrxA2(a) primers introduced EcoRI or PstI restriction sites (shown in bold italics) for the cloning of trxA1 and trxA2, respectively, into pGEM-3-Zf(-). PCRs were performed with Taq polymerase using genomic DNA from strain 26695 as a template. pGEM-3Zf(-) was linearized with EcoRI or PstI and ligated with the products resulting from PCR with the above primers after digestion with the same restriction enzymes, generating plasmids pSLC1 and pSLC4 (Table 1). The TrxA2(b) primers amplified an internal fragment of trxA2 and were used for mutant verification. The remaining primers were used for cloning into pGEM T-easy. All the resulting plasmids were linearized at unique restriction sites within the H. pylori gene inserts, as detailed in Table 1. Protruding 5' termini were in-filled using the Klenow fragment of the E. coli DNA polymerase I. For insertional inactivation, a chloramphenicol acetyltransferase (cat) cassette or an aminoglycoside phosphotransferase gene (aphAIII) derived from Campylobacter coli (Wang & Taylor, 1990; Pittman et al., 2001) were used. The 749 bp cat cassette excised from pUCAT by HincII digestion was used to inactivate trxA1 and trxA2. The cassette was purified using a Gel Extraction Kit (Qiagen) and blunt-end ligated into pSLC1 and pSLC4, respectively. Plasmids pSLC7 and pSLC10 were generated (see Table 1). To construct a plasmid suitable for generating a H. pylori trxA1 trxA2 double mutant, the aphAIII gene was amplified from pUKAN with Pfu polymerase, using primers Kan-F (5'-ACTGAGATCTACTCTATGAAGCGCCATATT-3') and Kan-R (5'-CAATAGATCTTTTAGACATCATCTAAATCTAGG-3'). pSLC3 was linearized at the unique BstXI site within the trxA2 gene, in-filled using the Klenow fragment and blunt-end ligated with the aphAIII gene. The resulting plasmid was designated pSLC9. Plasmids pSLC11 and pSLC12 were derived from pSLC5 and pSLC6 by insertion of aphAIII into the bcp and tpx genes, respectively.

Transformation of H. pylori 26695 was carried out with pSLC7, pSLC10, pSLC11 and pSLC12 to generate single gene mutants in trxA1 (strain SLC100), trxA2 (strain SLC200), bcp (strain SLC400) and tpx (strain SLC500), respectively. Strain SLC100 was transformed with pSLC9 to generate a trxA1 trxA2 double mutant (strain SLC300). H. pylori 1061 was transformed with pSLC8 to construct a non-polar trxA1 mutant. Natural transformations were carried out as described by Ferrero et al. (1992). Genomic DNA was extracted from putative recombinants, and correct insertion of the resistance cassettes was evaluated by PCR using the above primers.

Effect of oxidative and NO stress on cell viability.
Wild-type and mutant H. pylori strains were grown on Columbia blood agar plates for 48–72 h, harvested and resuspended in BHI/FCS to an OD600 value of 1·0. Viable counts showed that this corresponded to about 107–109 c.f.u. ml-1. Aliquots (2 ml) of this cell suspension were distributed to a series of 20 ml tissue-culture flasks, to which either no additions were made (control) or 50 mM H2O2, 10 mM methyl viologen (paraquat; a superoxide generator), 10 mM sodium nitroprusside (SNP; an NO+ donor) or 5 mM S-nitrosoglutathione (GSNO; an NO{bullet} releaser and NO+ donor) were added. The flasks were incubated at 37 °C under microaerobic conditions with orbital shaking at 120 r.p.m. One flask with no additions was also incubated under fully aerobic conditions (21 %, v/v, O2) at 37 °C with orbital shaking at 250r.p.m. to determine the effect of oxygen stress. Periodically, samples were taken from the flasks, serially diluted in BHI and then plated onto blood agar. Colonies were counted after 96–120 h incubation.

Cumene hydroperoxide disc-diffusion assay.
H. pylori was grown on a blood agar plate for 48–72 h, after which time the growth was scraped off and resuspended in 1 ml BHI broth. Aliquots (100 µl) of this suspension were spread out onto fresh blood agar plates and sterile 5 mm diameter Whatman filter paper discs were placed in the centre of the plates. Five microlitres of a 10 % (v/v) cumene hydroperoxide solution in DMSO was pipetted onto the discs and the plates were incubated for 48–72 h at 37 °C. The diameter of the zone of growth inhibition was measured. No growth inhibition was observed when DMSO alone was placed on the filter discs. Six replicate assays were performed for each strain, and the data were subjected to Student's t-test to evaluate their statistical significance.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Mutagenesis of the trx, tpx and bcp genes
Fig. 1 shows the genes analysed in this study and the strategy used for mutagenesis. PCR-amplified genes were insertionally inactivated at unique restriction sites using antibiotic-resistance cassettes and transferred into H. pylori 26695 by natural transformation. For each mutant, verification that correct chromosomal insertion of the cassettes had occurred by a double homologous recombination event was obtained by PCR analysis of parental and mutant genomic DNA (Fig. 1a–f).



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 1. Mutagenesis of thioredoxin and thiol-peroxidase genes in H. pylori, and verification of mutant strains by PCR. Mutants were constructed by cloning cat or aphAIII cassettes ({blacktriangledown}) into unique restriction sites in the insert DNA of the relevant plasmids shown (see also Table 1), followed by transformation into H. pylori. (a) Agarose gel of PCR products obtained from parental strain 26695 (P) and trxA1 mutant (M) genomic DNA, using primers TrxA1C-F and TrxA1C-R, showing an increase in size of ~750 bp of the mutant PCR product due to insertion of the cat cassette. (b) Agarose gel of PCR products obtained from strain 1061 (P) and trxA1 mutant (M) genomic DNA, using primers TrxA1K-F and TrxA1K-R, showing an increase in size of ~1·5 kb of the mutant PCR product due to insertion of the aphAIII gene. (c) Agarose gel of PCR products obtained from parental strain 26695 (P) and trxA2 mutant (M) genomic DNA, using primers TrxA2(b)-F and TrxA2(b)-R, showing an increase in size of ~750 bp of the mutant PCR product due to insertion of the cat cassette. (d) Agarose gel of PCR products obtained from strain 26695 (P) and trxA1 : : cat trxA2 : : aphAIII mutant (M) genomic DNA, using primers TrxA2-F and TrxA2-R, showing an increase in size of ~1·5 kb of the mutant PCR product due to insertion of the aphAIII gene. (e) Agarose gel of PCR products obtained from strain 26695 (P) and bcp mutant (M) genomic DNA, using primers Bcp-F and Bcp-R, showing an increase in size of ~1·5 kb of the mutant PCR product due to insertion of the aphAIII gene. (f) Agarose gel of PCR products obtained from strain 26695 (P) and tpx mutant (M) genomic DNA, using primers Tpx-F and Tpx-R, showing an increase in size of ~1·5 kb of the mutant PCR product due to insertion of the aphAIII gene.

 
Polar effects were a consideration in constructing mutants in the trxA1–trxB operon. In strain 26695, trxA1 (HP0824) was successfully mutagenized using the cat cassette (Wang & Taylor, 1990), which may also prevent transcription of trxB. However, transformation of strain 26695 by pSLC8, containing a terminatorless aphAIII gene in trxA1 (allowing read-through into trxB, driven by the aphAIII promoter), was unsuccessful, but such mutants were obtained using the more highly transformable H. pylori strain 1061 (Fig. 1b). One such mutant was designated SLC600, and it showed phenotypic properties identical to those of strain SLC100 when compared by each of the experiments described below (data not shown), indicating that any polar effects on trxB were minimal and did not contribute to the observed phenotype of the trxA mutants. A trxA1 trxA2 double mutant was also constructed by transformation of strain SLC100 (26695 trxA1 : : cat) with pSLC9, which contains the C. coli aphAIII kanamycin-resistance gene inserted at the unique BstXI site in HP1458 (Fig. 1d). Mutants in the bcp and tpx genes were successfully constructed using the aphAIII gene (Fig. 1e, f).

Growth characteristics of the trx, tpx and bcp mutants: thioredoxin deficiency results in growth inhibition by molecular oxygen
Strains SLC100 (26695 trxA1 : : cat), SLC200 (26695 trxA2 : : cat), SLC300 (26695 trxA1 : : cat trxA2 : : aphAIII) and SLC600 (1061 trxA1 : : aphAIII) all grew poorly on blood agar plates incubated under standard microaerobic conditions (with 10 %, v/v, O2), forming smaller colonies than the respective 26695 or 1061 parent strains. The nature of the growth defects in the 26695 thioredoxin mutants was determined in BHI/FCS liquid batch cultures incubated in gas atmospheres containing either 10 or 2 % (v/v) O2 (Fig. 2). The wild-type grew at a similar rate in either gas atmosphere (doubling time of ~3 h), but reached slightly higher final cell densities when cultured under 10 % (v/v) O2 (Fig. 2a). Strain SLC100 (trxA1 : : cat) showed an extended lag period and a significantly reduced exponential growth rate (doubling time of ~6 h) at 10 % (v/v) O2 (Fig. 2b), but cultures eventually reached the same final cell density as the wild-type parent strain (compare Fig. 2a, b). However, when cultured with 2 % (v/v) O2 in the gas atmosphere, strain SLC100 grew as well as the wild-type (Fig. 2a, b), even reaching slightly higher final cell densities. At 10 % (v/v) O2, strain SLC200 (trxA2 : : cat) showed a severe reduction in both the rate and extent of growth compared to its isogenic parent, but grew as well or better than the parent strain at 2 % (v/v) O2 (Fig. 2a, c). As expected, the SLC300 trxA1 trxA2 double mutant showed a severe growth defect at 10 % (v/v) O2 but not at 2 % (v/v) O2 (Fig. 2d), similar to that of strain SLC200. The data clearly show that both thioredoxin 1 and 2 are important for the normal microaerobic growth of H. pylori, with a deficiency of either resulting in an increased sensitivity to molecular oxygen. The isolation of strain SLC300, however, indicates that neither thioredoxin is absolutely essential for viability.



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 2. Growth of wild-type and trx mutants in gas atmospheres of 2 or 10 % (v/v) O2. BHI/FCS medium was equilibrated in conical flasks for at least 24 h at 37 °C in a controlled gas atmosphere of either 2 % (v/v) O2/5 % (v/v) CO2/93 % (v/v) N2 or 10 % (v/v) O2/5 % (v/v) CO2/85 % (v/v) N2. All strains were inoculated to a starting OD710 value of between 0·1 and 0·2; growth was monitored spectrophotometrically. Open symbols, growth at 2 % (v/v) O2; closed symbols, growth at 10 % (v/v) O2. (a) Wild-type strain 26695; (b) strain SLC100 (trxA1); (c) strain SLC200 (trxA2); (d) strain SLC300 (trxA1 trxA2).

 
In contrast to the thioredoxin mutants, the bcp and tpx mutants (strains SLC400 and SLC500, respectively) displayed only a slight reduction in growth rate compared to the parent strain when incubated under 10 % (v/v) O2, and both mutant and wild-type cultures eventually reached the same final cell density (data not shown).

Differential roles of thioredoxins 1 and 2 in protection against oxidative stress
Fig. 3 shows the effects of exposure to air, hydrogen peroxide or superoxide stress on the viability of cell suspensions of the wild-type and trx mutants, in comparison to control cells incubated under standard microaerobic conditions at 10 % (v/v) O2. H. pylori wild-type cells taken from 2–3-day-old plate cultures and resuspended to densities of 107–109 c.f.u. ml-1 showed a biphasic loss of viability after exposure to 21 % (v/v) O2 in air, 50 mM H2O2 or 10 mM paraquat. An initial gradual decline was followed by a phase in which the cells lost viability much more rapidly (Fig. 3b–d). In contrast, incubation under microaerobic conditions resulted in maintenance of the initial level of viability in each of the strains for the duration of the experiment (Fig. 3a). Strain SLC100 (trxA1) was killed much more rapidly than its isogenic parent by each of the oxidative stress treatments imposed (Fig. 3b–d) and was particularly sensitive to hydrogen peroxide killing (Fig. 3c). Strain SLC200 (trxA2) also showed an increase in sensitivity to killing by oxidative stress, but this was much less pronounced than with SLC100. As expected from these results, strain SLC300 (trxA1 trxA2) was slightly more sensitive than SLC100 to loss of viability upon exposure to oxygen, hydrogen peroxide or paraquat. The data show that a deficiency in TrxA1 results in greater sensitivity of H. pylori to killing by several types of oxidative stress compared to TrxA2 deficiency.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 3. Effect of oxidative stress on the viability of wild-type and trx mutants. Cell suspensions in BHI/FCS were incubated at 37 °C (a) in a control atmosphere of 10 % (v/v) O2/5 % (v/v) CO2/85 % (v/v) N2, (b) in air, (c) in the control atmosphere plus a final concentration of 50 mM H2O2 or (d) in the control atmosphere plus a final concentration of 10 mM methyl viologen (paraquat) as a superoxide generator (O2-). At intervals aliquots were removed, serially diluted and plated onto blood agar plates to determine viability as c.f.u. (ml original suspension)-1. {square}, Wild-type strain 26695; {blacksquare}, strain SLC100 (26695 trxA1 : : cat); {bullet}, strain SLC200 (26695 trxA2 : : cat); {circ}, strain SLC300 (26695 trxA1 : : cat trxA2 : : aphAIII). Data from a single representative experiment are shown.

 
The experiment shown in Fig. 4(a) shows the sensitivity of the wild-type and trx mutants to growth inhibition by the potent organic oxidant cumene hydroperoxide, as measured by a disc-diffusion assay. The trxA1 and trxA2 mutants both showed a statistically significant increase in the diameter of the zone of growth inhibition in this assay compared to the parent strain (P < 0·001 in each case), indicating a role for both thioredoxins in organic peroxide detoxification. The zone of growth inhibition was significantly greater in the trxA2 mutant compared to the trxA1 mutant (P = 0·008). Combining these mutations was even more deleterious, and the trxA1 trxA2 double mutant was significantly more sensitive to cumene hydroperoxide compared to the trxA2 single mutant (P = 0·001).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 4. Sensitivity of wild-type and mutant strains to cumene hydroperoxide. (a) Comparison of wild-type 26695 and trx mutants. (b) Separate experiment comparing wild-type 26695 with bcp and tpx mutants. A filter disc containing 5 µl of 10 % (v/v) cumene hydroperoxide in DMSO was placed in the centre of a blood agar plate spread with about 108 cells of the appropriate strain. The histograms represent the mean of six replicate assays in which the diameters of the zone of growth inhibition around the filter discs were measured after 2–3 days microaerobic incubation at 37 °C. The error bars represent the standard deviation of the means. Application of Student's t-test showed that the differences between the wild-type and each mutant were significant at P < 0·05 (see text for more details).

 
The thiol peroxidase Tpx provides protection against peroxide and superoxide stress
Fig. 5 shows an analogous experiment to that in Fig. 3, in which the oxidative stress phenotypes of the bcp and tpx mutants were examined by comparisons of the kinetics of killing after exposure to air, hydrogen peroxide or superoxide generated in situ by paraquat. In contrast to the thioredoxins, Tpx or Bcp deficiency did not affect the sensitivity of H. pylori to killing by oxygen (Fig. 5b) consistent with the similar microaerobic growth characteristics of the respective mutants compared to the wild-type, as noted above. However, a clear increase in the rate of killing of the tpx mutant was observed upon exposure to hydrogen peroxide (Fig. 5c) and paraquat (Fig. 5d) compared to the parent strain. The bcp mutant showed only a very slightly increased sensitivity to hydrogen peroxide killing, but an increase in sensitivity to superoxide stress was more evident (Fig. 5d). The more significant role of Tpx compared to Bcp was also evident in the cumene hydroperoxide disc-diffusion assay (Fig. 4), where much greater growth inhibition by this oxidant was observed with the tpx mutant compared to that seen with the bcp mutant.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 5. Effect of oxidative stress on the viability of wild-type, bcp and tpx mutants. Conditions were exactly as described in the legend to Fig. 3. {square}, Wild-type strain 26695; {blacksquare}, strain SLC400 (26695 bcp : : aphAIII); {blacktriangleup}, strain SLC500 (26695 tpx : : aphAIII). Data from a single representative experiment are shown.

 
Resistance to nitrosative stress is mediated mainly by thioredoxin 1 but not by Tpx or Bcp
Fig. 6 shows the results of an experiment to test the involvement of the thioredoxins and the thiol-peroxidases Tpx and Bcp in mediating resistance to the bactericidal effects of NO. Due to the high reactivity of this molecule, compounds that decompose to release NO in situ were employed. The kinetics of killing of the wild-type and mutant strains by two different NO-releasing agents, SNP (predominantly a nitrosating agent by donation of the nitrosonium cation, NO+) and GSNO (an NO{bullet} releaser and NO+ donor), were compared. Strain 26695 was killed by both reagents, but the trxA1 mutant was killed much more rapidly than its wild-type parent while the trxA2 mutant displayed very similar killing kinetics compared to the wild-type (Fig. 6a, b). However, the trxA1 trxA2 double mutant was killed more rapidly than the trxA1 single mutant, particularly by GSNO, indicating some involvement of thioredoxin 2. In contrast to these results, neither SNP nor GSNO affected the kinetics of killing of the tpx or bcp mutants in comparison to the wild-type (Fig. 6c, d), suggesting that the peroxiredoxins encoded by these genes are not involved in NO detoxification.



View larger version (20K):
[in this window]
[in a new window]
 
Fig. 6. Effect of nitrosative stress on the viability of wild-type, trx, bcp and tpx mutants. Cell suspensions in BHI/FCS were incubated at 37 °C in an atmosphere of 10 % (v/v) O2, 5 % (v/v) CO2 and 85 % (v/v) N2 in the presence of either 10 mM SNP (a, c) or 5 mM GSNO (b, d). {square}, Wild-type strain 26695; {diamondsuit}, strain SLC100 (26695 trxA1 : : cat); {bullet}, strain SLC200 (26695 trxA2 : : cat); {circ}, strain SLC300 (26695 trxA1 : : cat trxA2 : : aphAIII); {blacksquare}, strain SLC400 (26695 bcp : : aphAIII); {blacktriangleup}, strain SLC500 (26695 tpx : : aphAIII). Data from a single representative experiment are shown.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Thioredoxins are key proteins in many crucial cellular functions, including oxidative stress management (Ritz & Beckwith, 2001), and act as electron donors to a number of enzymes involved in the detoxification of reactive oxygen or nitrogen species, including the AhpC family of peroxiredoxins. Several recent studies have begun to elucidate the important role played by AhpC in the management of oxidative stress in H. pylori (Lundstrom & Bolin, 2000; Bryk et al., 2000; Baker et al., 2001; Olczak et al., 2002). The source of electrons for the catalytic activity of this enzyme has been shown to be thioredoxin 1 rather than thioredoxin 2 by biochemical studies in vitro (Baker et al., 2001), although it is possible other electron donors might contribute in vivo. In this study, we have shown that mutation of the genes encoding either thioredoxin results in an increase in sensitivity to several forms of oxidative and nitrosative stress. The phenotypes observed for the trxA1 and trxA2 single mutants and the trxA1 trxA2 double mutant were different, implying that Trx1 and Trx2 have distinct roles in H. pylori. However, changes in the relative abundance of these proteins under various stresses or in different mutant backgrounds might also be important, but although both thioredoxins have been identified on two-dimensional gels by proteomic techniques (Jungblut et al., 2000; Windle et al., 2000), there is as yet no information on how distinct oxidative or nitrosative stresses might differentially regulate their expression.

The H. pylori thioredoxins are likely to interact with many cellular proteins, but analysis of the effects of the trxA1 mutation fit in well with a major in vivo role for TrxA1 as the electron donor to AhpC. Like mutants in the ahpC gene (Olczak et al., 2002), the trxA1 mutant was more oxygen sensitive than the parent strain; it grew less well in 10 % (v/v) O2 compared to 2 % (v/v) O2 and cell suspensions were killed more rapidly than the parent strain after exposure to 21 % (v/v) O2. It also showed greater sensitivity to hydrogen peroxide, cumene hydroperoxide and the superoxide generator paraquat. Our results also demonstrate a particularly important role for Trx1 in resistance to nitrosative stress. NO reacts rapidly in aqueous solution with superoxide to form the extremely toxic peroxynitrite (Hughes, 1999). Studies with purified AhpC from H. pylori have demonstrated that it can catalyse the reduction of peroxynitrite to nitrite (Bryk et al., 2000) and hence detoxify this compound. An increased sensitivity of the trxA1 mutant to NO releasers would thus be expected if a major source of electrons for the reduction of peroxynitrite by AhpC was Trx1.

The role of Trx2 in oxidative stress management in H. pylori is less clear than with Trx1. It has been shown that, in vitro, Trx2 does not act as an electron donor to AhpC (Baker et al., 2001). Nevertheless, the trxA2 mutant constructed in this study was more sensitive to both oxidative and nitrosative stress than the parent strain. One dramatic effect of the inactivation of trxA2 was a much greater inhibition of growth at 10 % (v/v) O2 compared to that seen with the trxA1 mutant. This was particularly apparent in BHI liquid batch cultures, whereas growth on blood agar plates was affected less severely, possibly due to a protective effect of haem in removing reactive oxygen species. Some of the effects of the trxA2 mutation were more apparent in a trxA1 background. The two thioredoxins may be able to substitute for some functions in vivo, but it is likely that TrxA2 is acting as an electron donor for another reductase, peroxidase or peroxiredoxin that contributes to the removal of reactive oxygen or nitrogen species, in addition to the Trx1/AhpC system. Obvious candidates would be Tpx or Bcp, although these enzymes seem not to be important for NO detoxification as judged from the phenotypes of the cognate mutants. Studies with the purified proteins will be needed to clarify this.

In many bacteria, the thioredoxin system acts as the electron donor for ribonucelotide reductase (RNR; Jordan & Reichard, 1998). H. pylori contains only a single RNR, a heterodimeric oxygen-dependent class I type enzyme encoded by the nrdA and nrdB genes (Tomb et al., 1997; Alm et al., 1999). Reliance on this type of RNR would prevent H. pylori from growing strictly anaerobically, as is the case with its close relative Campylobacter jejuni (Sellars et al., 2002). The activity of this enzyme is therefore essential for DNA synthesis and thus for cell viability under normal microaerobic conditions. Our finding that it was possible to isolate a trxA1 trxA2 double mutant must imply that the H. pylori RNR can be supplied with electrons from a source other than Trx1 or Trx2, which is interesting in view of the fact that the bacterium appears to be deficient in other reduction systems that might substitute for Trx, particularly glutathione and the glutaredoxin system (Tomb et al., 1997; Alm et al., 1999; Baker et al., 2001). Although the double mutant clearly had a very severe deficiency in oxidative stress resistance, compensatory up-regulation of a Trx-independent antioxidant system may be able to compensate to some extent, as has been reported in the case of ahpC mutants of H. pylori in which expression of the iron-binding NapA protein is increased (Olczak et al., 2002).

The role of the thiol peroxidase Tpx in oxidative stress resistance has also been investigated in this study. Wan et al. (1997) first identified the Tpx protein in H. pylori and the partially purified enzyme was shown to protect glutamine synthetase from oxidative inactivation and to possess thioredoxin-linked peroxidase activity, when assayed using E. coli Trx. Tpx is an abundant protein in H. pylori as judged by two-dimensional gel analyses (Jungblut et al., 2000) and the tpx gene is divergently transcribed from sodB, encoding superoxide dismutase (Tomb et al., 1997; Wan et al., 1997). The increased sensitivity of the tpx mutant to hydrogen peroxide, cumene hydroperoxide and the superoxide generator paraquat confirms a significant in vivo role for Tpx as an antioxidant protein in H. pylori. That the tpx mutant did not show an increased sensitivity to killing by nitrosative stress indicates a specific role for Tpx in detoxifying reactive oxygen species, contrasting with AhpC, which appears to be a more general peroxiredoxin.

H. pylori also contains another member of the peroxiredoxin family, the Bcp protein encoded by HP0136 in 26695. This is a homologue of the E. coli Bcp protein, which was shown by Jeong et al. (2000) to be a thioredoxin-dependent thiol peroxidase with a preference for certain organic peroxides such as linoleic acid hydroperoxide. Unlike an E. coli bcp null-mutant, which had a prolonged lag phase in liquid batch cultures and a pronounced hypersensitivity to hydrogen peroxide and organic peroxides (Jeong et al., 2000), the H. pylori bcp mutant constructed in this study had only a weak phenotype. A slight perturbation in growth characteristics (slightly slower growth rate) compared to the wild-type was reproducibly noted, but the mutant was not significantly more hydrogen peroxide sensitive and only slightly more superoxide sensitive than the parent strain. Thus, Bcp is not a major contributor to general oxidative stress resistance in H. pylori, but instead could have a more specific role, for example, in the removal of fatty acid hydroperoxides produced during metabolism or by oxidative stress.


   ACKNOWLEDGEMENTS
 
This work was supported by studentships to S. L. C. and M. G. from the UK Biotechnology and Biological Sciences Research Council.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Akopyants, N. S., Eaton, K. A. & Berg, D. E. (1995). Adaptive mutation and co-colonization during infection of gnotobiotic piglets. Infect Immun 63, 116–121.[Abstract]

Alm, R. A., Ling, L.-S. L., Moir, D. T. & 20 other authors (1999). Genomic sequencing comparison of two unrelated isolates of the human gastric pathogen Helicobacter pylori. Nature 397, 176–180.[CrossRef][Medline]

Andersen, L. P. & Wadström, T. (2001). Basic bacteriology and culture. In Helicobacter pylori: Physiology and Genetics, pp. 27–52. Edited by H. L. T. Mobley and others. Washington, DC: American Society for Microbiology.

Baker, L. M. S., Raudonikiene, A., Hoffman, P. S. & Poole, L. B. (2001). Essential thioredoxin-dependent peroxiredoxin system from Helicobacter pylori: genetic and kinetic characterization. J Bacteriol 183, 1961–1973.[Abstract/Free Full Text]

Bryk, R., Griffin, P. & Nathan, C. (2000). Peroxynitrite reductase activity of bacterial peroxiredoxins. Nature 407, 211–215.[CrossRef][Medline]

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

Dixon, M. F. (2001). Pathology of gastritis and peptic ulceration. In Helicobacter pylori: Physiology and Genetics, pp. 459–470. Edited by H. L. T. Mobley and others. Washington, DC: American Society for Microbiology.

Ferrero, R. L., Cussac, V., Courcoux, P. & Labigne-Roussel, A. (1992). Construction of isogenic urease-negative mutants of Helicobacter pylori by allelic exchange. J Bacteriol 174, 4212–4217.[Abstract]

Forman, D., Newell, D. G., Fullerton, F., Yarnell, J. W. G., Stacey, A. R., Wald, N. & Sitas, F. (1991). Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. Br Med J 302, 1302–1305.[Medline]

Goodwin, A., Kersulyte, D., Sisson, G., Veldhuyzen van Zanten, S. J. O., Berg, D. E. & Hoffman, P. S. (1998). Metronidazole resistance in Helicobacter pylori is due to null mutations in a gene (rdxA) that encodes an oxygen-insensitive NADPH nitroreductase. Mol Microbiol 28, 383–393.[CrossRef][Medline]

Hughes, M. N. (1999). Relationships between nitric oxide, nitroxyl ion, nitrosonium cation and peroxynitrite. Biochim Biophys Acta 1411, 263–272.[Medline]

Hughes, N. J., Chalk, P. A., Clayton, C. L. & Kelly, D. J. (1995). Identification of carboxylation enzymes and characterization of a novel four-subunit pyruvate : flavodoxin oxidoreductase from Helicobacter pylori. J Bacteriol 177, 3953–3959.[Abstract]

Hughes, N. J., Clayton, C. L., Chalk, P. A. & Kelly, D. J. (1998). Helicobacter pylori porCDAB and oorDABC genes encode distinct pyruvate : flavodoxin and 2-oxoglutarateacceptor oxidoreductases which mediate electron transport to NADP. J Bacteriol 180, 1119–1128.[Abstract/Free Full Text]

Jeong, W., Cha, M.-K. & Kim, I.-H. (2000). Thioredoxin dependent hydroperoxidase activity of bacterioferritin comigratory protein (BCP) as a new member of the thiol-specific antioxidant protein (TSA)/alkyl hydroperoxide peroxidase C (AhpC) family. J Biol Chem 275, 2924–2930.[Abstract/Free Full Text]

Jordan, A. & Reichard, P. (1998). Ribonucleotide reductases. Annu Rev Biochem 67, 71–98.[CrossRef][Medline]

Jungblut, P. R., Bumann, D., Haas, G., Zimny-Arndt, U., Holland, P., Lamer, S., Siejak, F., Aebischer, A. & Meyer, T. F. (2000). Comparative proteome analysis of Helicobacter pylori. Mol Microbiol 36, 710–725.[CrossRef][Medline]

Kelly, D. J. (1998). The physiology and metabolism of the human gastric pathogen Helicobacter pylori. Adv Microb Phys 40, 137–189.[Medline]

Lundstrom, A. M. & Bolin, I. (2000). A 26 kDa protein of Helicobacter pylori shows alkyl hydroperoxide reductase (AhpC) activity and the mono-cistronic transcription of the gene is affected by pH. Microb Pathog 29, 257–266.[CrossRef][Medline]

Marmur, J. (1961). A procedure for the isolation of deoxyribonucleic acid from microorganisms. J Mol Biol 3, 208–218.

Nagata, K., Yu, H., Nishikawa, M., Kashiba, M., Nakamura, A., Sato, E. F., Tamura, T. & Inouye, M. (1998). Helicobacter pylori generates superoxide radicals and modulates nitric oxide metabolism. J Biol Chem 273, 14071–14073.[Abstract/Free Full Text]

Olczak, A. A., Olson, J. W. & Maier, R. J. (2002). Oxidative stress resistance mutants of Helicobacter pylori. J Bacteriol 184, 3186–3193.[Abstract/Free Full Text]

Pittman, M. S., Goodwin, M. & Kelly, D. J. (2001). Chemotaxis in the human gastric pathogen Helicobacter pylori: different roles for CheW and the three CheV paralogues, and evidence for CheV2 phosphorylation. Microbiology 147, 2493–2504.[Abstract/Free Full Text]

Ritz, D. & Beckwith, J. (2001). Roles of thiol-redox pathways in bacteria. Annu Rev Microbiol 55, 21–48.[CrossRef][Medline]

Schröder, E. & Ponting, C. P. (1998). Evidence that peroxiredoxins are novel members of the thioredoxin fold superfamily. Protein Sci 7, 2465–2468.[Abstract/Free Full Text]

Sellars, M. J., Hall, S. J. & Kelly, D. J. (2002). Growth of Campylobacter jejuni supported by respiration of fumarate, nitrate, nitrite, trimethylamine-N-oxide or dimethylsulphoxide requires oxygen. J Bacteriol 184, 4187–4196.[Abstract/Free Full Text]

Storz, G. & Zheng, M. (2000). Oxidative stress. In Bacterial Stress Responses, pp. 47–60. Edited by G. Storz & R. Hengge-Aronis. Washington, DC: American Society for Microbiology.

Tomb, J.-F., White, O., Kerlavage, A. R. & 39 other authors (1997). The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388, 539–547.[CrossRef][Medline]

Wan, X.-Y., Zhou, Y., Yan, Z.-Y., Wang, H.-L., Hu, Y.-D. & Jin, D.-Y. (1997). Scavengase p20: a novel family of bacterial antioxidant enzymes. FEBS Lett 407, 32–36.[CrossRef][Medline]

Wang, Y. & Taylor, D. E. (1990). Chloramphenicol resistance in Campylobacter coli – nucleotide-sequence, expression, and cloning vector construction. Gene 94, 23–28.[CrossRef][Medline]

Windle, H. J., Fox, A., Ni Eidhin, D. & Kelleher, D. (2000). The thioredoxin system of Helicobacter pylori. J Biol Chem 275, 5081–5089.[Abstract/Free Full Text]

Received 23 July 2002; revised 8 October 2002; accepted 9 October 2002.