Role and regulation of the superoxide dismutases of Staphylococcus aureus

Michail H. Karavolos1, Malcolm J. Horsburgh1, Eileen Ingham2 and Simon J. Foster1

1 Department of Molecular Biology and Biotechnology, University of Sheffield, Firth Court, Western Bank, Sheffield S10 2TN, UK
2 Department of Microbiology, University of Leeds, Old Medical School, Leeds, LS2 9NL, UK

Correspondence
Simon J. Foster
S.Foster{at}sheffield.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Staphylococcus aureus has two superoxide dismutases (SODs), encoded by the sodA and sodM genes, which inactivate harmful superoxide radicals () encountered during host infection or generated from aerobic metabolism. The transcriptional start sites have been mapped and expression analysis on reporter fusions in both genes has been carried out. Under standard growth conditions, manganese (Mn), a mineral superoxide scavenger, elevated total SOD activity but had no effect on the transcription of either gene. Transcription of sodA and sodM was most strongly induced by either internally or externally generated , respectively. Sensitivity to internally generated was linked with SodA deficiency. Mn supplementation completely rescued a sodA mutant when challenged by internally generated , and this was growth-phase-dependent. Sensitivity to externally generated stress was only observed in a sodA sodM mutant and was Mn-independent. In a mouse abscess model of infection, isogenic sodA, sodM and sodA sodM mutants had reduced virulence compared to the parental strain, showing the importance of the enzymic scavenging system for the survival of the pathogen.


Abbreviations: SOD, superoxide dismutase; X/XO, xanthine/xanthine oxidase


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Reactive oxygen species (ROS) such as the superoxide anion (), hydrogen peroxide (H2O2) and hydroxyl radical ({bullet}OH) are natural products of metabolism formed when oxygen becomes partially reduced and are crucial to host defence for the optimal antimicrobial activity of neutrophils and other phagocytes (Hurst & Barrette, 1989; Miller & Britigan, 1995). Internal generation of these reactive species can result in damage to DNA, proteins and lipids (Imlay, J. A. & Linn, 1988).

Superoxide dismutases (SODs) are metalloproteins that catalyse dismutation of , the first ROS produced upon reduction of oxygen. By converting to H2O2 and O2, not only do they protect against direct damage caused by , but also against indirect toxicity by preventing a Fe3+-dependent catalytic reaction leading to the production of {bullet}OH via the Haber–Weiss reaction (Haber & Weiss, 1934). There is also evidence that can reduce hypochlorous acid (HOCl), a potent oxidant derived from the interaction of H2O2 with phagocyte-derived peroxidases, to form {bullet}OH (Candeias et al., 1993; Ramos et al., 1992).

Many Gram-negative bacteria, including Escherichia coli and Salmonella typhimurium, have three SODs that are classified according to their metal cofactor and cellular localization. Iron (Fe) and manganese (Mn) SODs are present in the cytoplasm, where they protect cellular macromolecules from endogenous superoxide, whereas copper–zinc (Cu–Zn) SODs reside in the periplasm, protecting from superoxide generated externally by host phagocytes during infection (Benov & Fridovich, 1994; Imlay, K. R. & Imlay, J. A., 1996). Other metal cofactors have been found in SODs, including a cytoplasmic nickel (Ni) SOD in Streptomyces coelicolor (Kim et al., 1998).

Staphylococcus aureus has two SOD-encoding genes, sodA and sodM (Clements et al., 1999; Valderas & Hart, 2001). The products of these genes combine to form three SOD activity zones as observed by zymogram analysis (Clements et al., 1999; Valderas & Hart, 2001), two homodimers and a heterodimer. In vitro data so far indicate that the SodA homodimer is responsible for the majority of the S. aureus SOD activity. SodA is a Mn cofactor enzyme (Clements et al., 1999) and the metal requirement of SodM has been proposed to be Mn also (Valderas & Hart, 2001).

The role of SODs in virulence has been investigated in several bacterial species. Mycobacterium tuberculosis survival in macrophages was not affected by loss of its Cu–ZnSOD (Dussurget et al., 2001). However, increased expression of the periplasmic Cu–ZnSOD in invasive E. coli enhanced survival within non-phagocytic cells (Battistoni et al., 2000). Streptococcal MnSOD also contributes to virulence (Poyart et al., 2001; Yesilkaya et al., 2000). In S. aureus, there is contradicting evidence as to the roles of the SODs in virulence. A study of S. aureus clinical isolates concluded that SOD activity did not correlate with lethality in a mouse model of infection (Mandell, 1975). Furthermore, 8325-4 isogenic sodA (Clements et al., 1999) or sodM (Schneider et al., 2002) mutants were not attenuated in animal models of infection (Chan et al., 1998; Schneider et al., 2002). However, SOD activity was significantly higher in S. aureus strains isolated from patients with staphylococcal disease compared to those isolated from non-staphylococcal disease patients (Kanafani & Martin, 1985).

As well as enzymic SOD activity, it is now becoming apparent that, in many organisms, several Mn2+ complexes can catalyse the elimination of (Archibald & Fridovich, 1982; Beyer & Fridovich, 1989; Darr et al., 1987; Faulkner et al., 1994; Kono et al., 1976). In Bacillus subtilis, both sodA and Mn are essential for resisting superoxide stress (Inaoka et al., 1999). In Neisseria gonorrhoeae, accumulation of Mn correlates with resistance to oxidative killing by and is independent of SOD activity (Tseng et al., 2001). Lactic acid bacteria lack enzymic SOD but accumulate Mn to levels that can protect them from , by a mechanism that is proposed to involve oxidation of Mn2+ to Mn3+ by and then subsequent reduction of Mn3+ to Mn2+ by the H2O2 generated or an intracellular reductant (Archibald & Fridovich, 1981). Furthermore, in E. coli, Mn supplementation relieves the phenotypic defects seen in SOD-deficient strains (Al-Maghrebi et al., 2002).

We have recently characterized Mn2+ homeostasis in S. aureus and have proposed a similar non-enzymic Mn2+-associated protection mechanism. S. aureus has two Mn uptake systems that consist of an ABC-type permease, MntABC and an Nramp homologue, MntH (Horsburgh et al., 2002b). The expression of these transporters is regulated by MntR, a DtxR-like metalloregulatory protein (O'Halloran, 1993; Que & Helmann, 2000), and is required for Mn2+-dependent mediation of oxidative stress resistance in response to internal (Horsburgh et al., 2002b). Inactivation of MntA results in methyl viologen sensitivity specifically under Mn-depleted conditions. In addition to MntR, S. aureus also has three Fur homologues: Fur acts as an Fe2+-dependent transcriptional repressor of genes encoding iron uptake proteins (Hantke, 2001; Horsburgh et al., 2001b; Xiong et al., 2000), PerR controls the peroxide resistance regulon in response to Mn and Fe (Bsat et al., 1998; Horsburgh et al., 2001a, b) and Zur controls a zinc uptake system (Gaballa & Helmann, 1998; Lindsay & Foster, 2001). Fur, PerR and MntR form an integrated network controlling peroxide stress resistance as well as Fe2+ and Mn2+ homeostasis (Horsburgh et al., 2002c).

Most physiological studies on S. aureus have used strains of the 8325 lineage (including RN6390 and 8325-4). However, these strains have an 11 bp deletion in rsbU, which is involved in the positive control of SigB activity (Giachino et al., 2001). SigB is an alternative sigma factor with a role in stress resistance. To circumvent possible problems associated with the rsbU mutation, we have recently constructed an RsbU+ derivative of 8325-4 called SH1000 (Horsburgh et al., 2002a).

To further understand the mechanisms by which S. aureus resists oxidative stress and evades host defences, we have characterized the two staphylococcal SODs and describe the relationship between the enzymic and non-enzymic systems of superoxide resistance in S. aureus SH1000.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and growth conditions.
Strains and plasmids used in this study are listed in Table 1. E. coli was grown in Luria–Bertani (LB) medium at 37 °C. S. aureus was grown in brain–heart infusion (BHI; Oxoid) or Chelex-treated BHI (CT-BHI) containing 400 µM MgSO4 at 37 °C with shaking at 250 r.p.m. CT-BHI was prepared by adding 25 g Chelex-100 l-1 and stirring at room temperature for 6 h prior to removal by filtration (Horsburgh et al., 2001a). Experimental 25 ml cultures in acid-washed (0·1 M HCl, 24 h) 250 ml flasks were inoculated at a starting OD600 value of 0·005 prior to growth at 37 °C. When required, antibiotics were added at the following concentrations: 100 mg ampicillin l-1; 50 mg kanamycin l-1; 50 mg neomycin l-1; 5 mg tetracycline l-1; 5 mg erythromycin l-1; 25 mg lincomycin l-1.


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

 
Construction of reporter and mutant strains.
To construct a sodM mutant, primers (5'-AAGGATCCCATTGATGACCATTTTAGC-3' and 5'-AAGAATTCTTGATGTCATAC-3; restriction sites underlined) were designed using sequences from the S. aureus NCTC 8325 Genome Sequencing Database maintained by the University of Oklahoma's Advanced Center for Genome Technology (http://www.genome.ou.edu/staph.html) and used to PCR-amplify a 2345 bp region containing sodM. The PCR product was cut with BamHI and EcoRI and cloned into BamHI/EcoRI-digested suicide vector pOB (Horsburgh et al., 2002b) to create pMKM. A KpnI-digested tet cassette amplified from pDG1513 (Guerout-Fleury et al., 1995) using primers OL32 and OL33 (Horsburgh et al., 2001b) was ligated into KpnI (unique within sodM) digested pMKM to generate pMKMTET.

The suicide plasmid pMKMTET was used to transform electrocompetent S. aureus RN4220 (Schenk & Ladagga, 1992), resulting in erythromycin- and tetracycline-resistant strains harbouring one disrupted and one functional copy of sodM through single-crossover homologous recombination. SH1000 was transduced (Novick, 1991) with a {phi}11 lysate from one of these integrants to resolve the sodM locus by transductional outcross. Tetracycline-resistant transductants were screened for loss of erythromycin resistance and thus loss of the plasmid. One such clone was isolated, producing strain MHKM (sodM : : tet). To obtain a sodA mutant, the sodA : : Tn917-LTV1 locus was transduced from strain SPW1 (Clements et al., 1999) into SH1000 using {phi}11, giving rise to strain MHKA (sodA : : Tn917). Southern blotting was used in each case to verify the location and structural integrity of the DNA at the integration site.

Transcriptional sodA : : lacZ and sodM : : lacZ reporter fusions were made by PCR amplification (primers 5'-AAGGATCCTAGCTTTCTATATGG-3' and 5'-AAGAATTCTAAATCTGTACCTTC-3', for sodA; primers 5'-ATGGATCCAGCAATCACATAG-3' and 5'-AAGAATTCGTCACGTACGTATTG-3', for sodM; restriction sites underlined) of suitable DNA fragments incorporating 950 and 847 bp of the promoter regions of the above genes, respectively. The purified DNA fragments were digested with appropriate restriction enzymes and cloned into similarly digested pAZ106 (Kemp et al., 1991), to generate pMK1A and pMK1M, respectively. The plasmids were electroporated into RN4220 and the resulting erythromycin-resistant clones were transduced into recipient SH1000 using {phi}11, producing lacZ fusion strains MHK1A (sodA : : pAZ106 SodA+) and MHK1M (sodM : : pAZ106 SodM+). Southern blotting was used in each case to verify the location and structural integrity of the DNA at the integration site.

Transcriptional start-site mapping.
RNA was isolated from late-exponentially growing (BHI) cells of S. aureus (OD600 8·0) using the Fast-prep blue kit (BIO-101). Primer extension reactions (Horsburgh & Moir, 1999) were performed using 20 µg total RNA and 10 pmol primer (5'-AGTTTCTTTGTCAAAATGTGG-3', sodA; 5'-ATTGTTCTTTGATCTATATATGG-3', sodM) and labelling with [{alpha}-32P]ATP (3000 Ci mmol-1, 111 TBq mmol-1). Sequencing (Amersham Sequenase 2.0 kit) was performed on purified pMK1A and pMK1M (Qiagen Mini-Prep kit) using the above oligonucleotides and labelling with 35S-labelled ATP (1000 Ci mmol-1, 37 TBq mmol-1).

SOD and {beta}-galactosidase assays.
Levels of {beta}-galactosidase activity were measured, using 4-methylumbelliferyl {beta}-D-galactoside (Sigma) as substrate, as described previously (Horsburgh et al., 2001a). One unit of {beta}-galactosidase activity was defined as the amount of enzyme that catalysed the production of 1 pmol 4-methylumbelliferone (MU) min-1 (OD600 unit)-1. To assess the effect of superoxide-producing agents on SOD transcription, double-distilled H2O (control), methyl viologen (1 mM) or xanthine/xanthine oxidase (X/XO) (0·5 mM/60 mU ml-1) were added at an OD600 value of ~0·1. Growth of the cells was not significantly affected by these additions (results not shown). Assays were performed on duplicate samples and the means of the values were determined. The results presented here are representative of at least two independent experiments.

For determination of SOD activity, samples were collected at the late-exponential phase of growth (OD600 ~10), washed in PBS and lysed using osmotic stress and lysostaphin (Clements et al., 1999). Total protein concentration was determined by Bradford assay (Bio-Rad) and samples were stored at -70 °C until needed. Spectrophotometric determination of total SOD activity was determined by the inhibition of autoxidation of pyrogallol according to the method of Marklund & Marklund (1974) using 1 µg of total protein per assay. Samples were analysed twice. For enzymic activity determination, 1 µg of total protein was applied onto a native PAGE gel. Bands were visualized by negative staining using the nitro blue tetrazolium method of Beauchamp & Fridovich (1971).

Disc-diffusion assay.
Sensitivity to methyl viologen was assessed by disc diffusion (Horsburgh et al., 2002b). Twenty-millilitre CT-BHI agar (1 %, w/v) plates (+/-20 µM MnSO4) were poured. Plates were overlaid with 5 ml of CT-BHI agar (0·7 %, w/v; +/-20 µM MnSO4) that had just been inoculated with 5 µl of stationary-phase CT-BHI-grown S. aureus (OD600 14). The overlay was allowed to set and an assay disc (5 mm diameter) was placed in the middle of the plate followed by 5 µl of the test compound (Sigma) (methyl viologen, 0·2 M). Plates were incubated overnight at 37 °C. Assays were performed in duplicate.

Internal superoxide stress assay.
Experimental cultures in CT-BHI with or without 20 µM Mn2+ were inoculated from an overnight CT-BHI culture to an OD600 value of ~0·005. Cells were challenged with methyl viologen (10 mM) at the early-exponential (T=2 h), post-exponential (T=6 h) and stationary (T=12 h) phases of growth. Samples were collected at appropriate intervals post-challenge and viable bacteria determined in duplicate by plating onto BHI agar. The experiment was performed twice and a representative dataset is shown.

External superoxide stress assay.
Experimental cultures in CT-BHI with or without 20 µM Mn2+ were inoculated from an overnight CT-BHI culture to an OD600 value of ~0·005. Cells from the early-exponential (T=2 h), post-exponential (T=6 h) and stationary (T=12 h) phases of growth were diluted in PBS to an OD600 value of 0·1 and the test compounds (X/XO, 1 mM/1·5 U ml-1) were added. Cells were incubated stationary at 37 °C for 1 h. Suitable dilutions were plated in duplicate on BHI plates to determine viable counts. The experiment was repeated twice and a representative set of values is shown.

Pathogenicity study.
A mouse abscess model of infection was used, as described in Chan et al. (1998). Results from 10 mice were recorded, and their significance was determined by the Student's t-test.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
SOD expression and activity
RNA extracted from late-exponential phase cells of (BHI) SH1000 or MJH502 (sigB) was subjected to primer extension analysis and transcription start sites were visualized on a 6 % (w/v) sequencing gel (Fig. 1). Under the conditions tested, transcription of sodA initiated from two {sigma}A-type promoters (Fig. 1a). These are designated as {sigma}A1 and {sigma}A2. More transcription from {sigma}A1 was observed in a sigB mutant background. Expression of sodM occurs from a single {sigma}A-type promoter that is significantly more active in MJH502 (sigB) and almost undetectable in the wild-type SH1000 (Fig. 1b).



View larger version (51K):
[in this window]
[in a new window]
 
Fig. 1. Transcriptional analysis of the SODs in S. aureus SH1000 and MJH502 (sigB). (a) Primer extension analysis of the sodA promoter region showing the two (PA1, PA2) start sites. (b) The sodM promoter region has a single start site (PM). Promoter -10 and -35 regions of (c) sodA and (d) sodM are underlined. Transcriptional start sites are enlarged. The ribosome-binding site and translational start of each gene are shown in bold.

 
To measure gene expression during growth, reporter fusion strains were made. Expression of sodA and sodM was maximal at the mid- (Fig. 2a, 60 000 {beta}-galactoside units) and late- (Fig. 2b, 12 500 {beta}-galactoside units) exponential phases of growth, respectively. In MJH502 (sigB), expression levels of sodA were unaffected (Fig. 2a), while sodM levels were elevated (Fig. 2b, 30 000 {beta}-galactoside units). Similar expression levels were obtained in the presence or absence of added Mn (results not shown).



View larger version (34K):
[in this window]
[in a new window]
 
Fig. 2. Expression of SODs. Growth (closed symbols) and {beta}-galactosidase activity (open symbols) in CT-BHI of sodA : : lacZ (a) and sodM : : lacZ (b) in SH1000 ({square}, {blacksquare}) and MJH502 (sigB) ({circ}, {bullet}). Protein-associated SOD activity of S. aureus SH1000 (c) and MJH502 (sigB) (d) during growth in CT-BHI with no added metals (-) and 20 µM Mn2+ (+Mn). The top and bottom band represent a SodM homodimer and SodA homodimer, respectively. The middle band is thought to represent a heterodimer. The experiment was repeated twice and a representative is shown here.

 
In metal-depleted cultures, SodM homodimer was the major SOD activity on an activity gel, while SodA homodimer activity was barely visible (Fig. 2c). Addition of Mn caused a dramatic increase in SodA homodimer activity and the disappearance of the SodM homodimer activity (Fig. 2c). Conversely, in MJH502 (sigB), SodM homodimer activity was independent of Mn levels and less SodA homodimer activity was apparent in the presence of Mn when compared to SH1000 (Fig. 2d).

Spectrophotometrically measured SOD activity in metal-depleted culture extracts of S. aureus was approximately 300 U mg-1 (Fig. 3). The addition of Mn increased specific activity to ~2500 U mg-1 (Fig. 3). Strain MHKM (sodM) grown in the absence of metals had no detectable SOD activity and upon addition of Mn its activity levels reached ~1900 U mg-1 (Fig. 3). Conversely, strain MHKA (sodA) grown in the absence of metals had an activity of ~500 U mg-1 which was reduced to ~200 U mg-1 upon addition of Mn (Fig. 3). Strain MHKAM (sodA sodM) growing in the absence or presence of Mn had residual activity of ~100 U mg-1 (Fig. 3). Thus, regulation of sodA and sodM is complex and occurs at the transcriptional and post-transcriptional levels.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 3. SOD activity in S. aureus. Comparison of SOD-specific activity of S. aureus SH1000 and strains MHKA (sodA), MHKM (sodM) and MHKAM (sodA sodM) in CT-BHI under metal limitation and upon addition of 20 µM Mn. Error bars are shown.

 
Effect of oxidative stress on SOD expression and activity
To determine if the expression of the SODs was induced by stress, the lacZ fusion strains were sampled after the addition of either methyl viologen (100 µM) or X/XO (0·5 mM/60 mU ml-1) at early-exponential phase (OD600 0·1). Under the conditions tested here, the addition of the superoxide-generators had no effect on the cell growth rate. In metal-depleted cultures, addition of either methyl viologen or X/XO did not significantly alter expression of sodA (results not shown). In Mn-containing cultures (20 µM), the expression of sodA increases threefold (180 000 {beta}-galactoside units, T=3 h) after addition of methyl viologen, while addition of X/XO had no effect compared to the control (60 000 {beta}-galactoside units, T=3 h; results not shown).

In the absence of metals, expression of sodM was not increased by addition of either methyl viologen or X/XO. In the presence of Mn, sodM transcription was induced by addition of X/XO (26 000 {beta}-galactoside units, T=5 h) compared to the control (18 000 {beta}-galactoside units, T=5 h; results not shown).

Mn affects resistance to oxidative-stress-generating compounds
We have previously shown that Mn uptake contributes to superoxide stress resistance (Horsburgh et al., 2002b) and have proposed that elemental dismutation of superoxide by Mn acts in concert with the SOD enzymes. The role of SODs and Mn availability in stress resistance was tested by disc-diffusion assay. Under the metal-depleted conditions used, strains SH1000 and MHKM (sodM) were not sensitive to methyl viologen stress (Fig. 4). However, strains MHKA (sodA) and MHKAM (sodA sodM) were sensitive (Fig. 4, 8–8·5 mm inhibition zones). Upon addition of Mn, MHKA (sodA) and MHKAM (sodA sodM) became fully resistant to methyl viologen. Thus, superoxide sensitivity due to the lack of both SOD enzymes can be relieved by the presence of Mn.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 4. The effect of Mn supplementation (20 µM) on the sensitivity of wild-type S. aureus SH1000 and isogenic mutants to methyl viologen using disc-diffusion assays. Cells were grown on CT-BHI plates with discs containing 0·2 M methyl viologen (5 µl). Error bars are shown. Solid bars, SH1000; hatched bars, MHKA (sodA); open bars, MHKM (sodM); horizontal-striped bars, MHKAM (sodA sodM).

 
Role of Mn and SODs on the survival of S. aureus during oxidative stress throughout growth
Temporal expression of sodA and sodM was shown earlier (Fig. 2). Superoxide stress was applied at different points during growth to determine if there was any growth-phase-associated role for the SODs. In the absence of metals, only strain MHKAM (sodA sodM) showed a growth defect which was restored to near wild-type levels upon incorporation of Mn in the growth medium (results not shown).

Exposure to methyl viologen during the early-exponential phase of growth (2 h) and in the absence of Mn resulted in a 103- to 104-fold loss of viability for strains MHKA (sodA) and MHKAM (sodA sodM) within 12 h post-addition of the compound (Fig. 5a I). Viability levels were not affected by the addition of Mn (Fig. 5a I). Viability of SH1000 and MHKM (sodM) was largely unaffected in either Mn-depleted or replete media (Fig. 5a I).



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 5. Role of Mn in stress resistance of wild-type SH1000 ({circ}, {bullet}), MHKA (sodA) ({square}, {blacksquare}), MHKM (sodM) ({triangleup}, {blacktriangleup}) and MHKAM (sodA sodM) ({lozenge}, {blacklozenge}) upon internal (methyl viologen, 10 mM) superoxide stress. (a) Cells were grown in CT-BHI with (closed symbols) or without (open symbols) Mn (20 µM). At points corresponding to (I) early-exponential (T=2 h), (II) post-exponential (T=6 h) and (III) stationary (T=12 h) phase, cells were challenged with methyl viologen. Samples were taken thereafter, and viability was determined by performing viable counts in duplicate. In the case of external superoxide stress (b), cells were grown as before and, at points corresponding to early-exponential (T=2 h), post-exponential (T=6 h) and stationary (T=12 h) phase, cells were resuspended in PBS to an OD600 value of ~0·1. X/XO (1 mM/1·5 U ml-1) was then added and the cells were incubated stationary at 37 °C for 1 h. Viability was determined by performing viable counts in duplicate. Solid bars, SH1000; hatched bars, MHKA (sodA); open bars, MHKM (sodM); horizontal-striped bars, MHKAM (sodA sodM). All the above experiments were repeated twice and representatives are shown here.

 
During the post-exponential phase of growth (6 h) and in the absence of Mn, exposure to methyl viologen resulted in a 103- to 104-fold drop in the viability of strains MHKA (sodA) and MHKAM (sodA sodM) within 24 h post-addition of the compound (Fig. 5a II). Growth of strains MHKA (sodA) and MHKAM (sodA sodM) in Mn-replete cultures resulted in a 100-fold increase in cell survival after 24 h (Fig. 5a II). Viability of strains SH1000 and MHKM (sodM) was unaffected in either metal-depleted or Mn-containing media (Fig. 5a II).

During stationary phase (12 h) and in the absence of Mn, viability of strains MHKA (sodA) and MHKAM (sodA sodM) was reduced by 103-fold after 24 h (Fig. 5a III). Mn improved MHKA (sodA) and MHKAM (sodA sodM) viability by 10-fold (Fig. 5a III). Viability of MHKM (sodM) reduced twofold and was unaffected by the presence or absence of Mn in the culture. The viability of SH1000 was unchanged in either metal-depleted or Mn-containing media (Fig. 5a III).

Strains MHKA (sodA), MHK (sodM) and MHKAM (sodA sodM) all showed less survival than SH1000 after exposure to X/XO stress only during the early- (2 h) and post- (6 h) exponential phases of growth (Fig. 5b). Resistance could only be restored for MHKA (sodA) and MHKM (sodM) during the post-exponential phase by the addition of Mn to the medium(Fig. 5b, 6 h). In the stationary phase, only MHKAM (sodA sodM) was more sensitive than SH1000, this being irrespective of Mn level (Fig. 5b, 12 h).



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 6. Virulence of strains SH1000, MHKA (sodA), MHKM (sodM) and MHKAM (sodA sodM) in a mouse abscess model of infection. Ten mice were used and the statistical significance of the results was assessed by the Student's t-test. The mean of each strain is shown by a horizontal bar.

 
Role of SODs in virulence
The role of the SODs in virulence was tested in an established mouse subcutaneous model of infection (Chan et al., 1998). Strains MHKA (sodA), MHKM (sodM) and MHKAM (sodA sodM) had significantly reduced virulence compared to the wild-type SH1000, using the Student's t-test (MHKA, P<0·01; MHKM, P<0·02; MHKAM, P<0·02) (Fig. 6). Thus, both SodA and SodM contribute to the virulence of S. aureus SH1000 in this model.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
We have characterized important aspects of the role and regulation of S. aureus SODs in response to superoxide stress and with respect to Mn availability.

Both S. aureus SODs are transcribed from {sigma}A-type promoters. However, unlike E. coli (Dubrac & Touati, 2000; Hassan & Schrum, 1994; Imlay, K. R. & Imlay, J. A., 1996) or B. subtilis (Inaoka et al., 1998), where expression of SODs is carried out from single {sigma}A-type promoters, in S. aureus sodA is transcribed from two {sigma}A-type promoters (PA1 and PA2). So far, the significance of this finding remains unclear. Furthermore, transcriptional data show an indirect repression of sodM PM and sodA PA1 promoters by {sigma}B, which is in agreement with the recent finding that {sigma}B indirectly represses a number of staphylococcal genes (Giachino et al., 2001; Horsburgh et al., 2002a). Gel activity evidence also suggests that {sigma}B is also involved in the post-transcriptional regulation of SodM homodimer activity. {sigma}B is a pleiotropic regulator in S. aureus with roles in stress resistance and virulence-determinant production (Horsburgh et al., 2002a). The indirect control of gene expression by {sigma}B alludes to the presence of further {sigma}B-controlled regulatory components.

In E. coli, Mn and Fe play an important role in MnSOD and FeSOD biosynthesis, respectively, and MnSOD or FeSOD activity increases in the presence of Mn or Fe, respectively (Pugh et al., 1984). Our work with S. aureus has shown that the presence of Mn increases SodA homodimer activity and concomitantly decreases SodM homodimer activity. This occurs primarily due to post-transcriptional effects, as the expression of the respective genes is independent of Mn availability, in the absence of superoxide-generating compounds. Synthesis of SodA in the presence of Mn (its co-factor) may enhance enzyme stability and perhaps heterodimer formation. In turn, this would deplete the level of SodM available for homodimer formation.

Superoxide induction of sodA and sodM transcription is Mn-dependent. sodM is most strongly induced in response to externally generated superoxide stress, whereas sodA is induced by internal stress. In Gram-negative bacteria, SODs are compartmentalized in order to deal with different sources of superoxide stress. MnSODs are present in the cytoplasm, protecting cell components from internal superoxide, whereas Cu–ZnSODs are periplasmic, protecting from external superoxide encountered during infection (Benov & Fridovich, 1994; Imlay, K. R. & Imlay, J. A., 1996). The importance of SodA in resisting internal superoxide stress in S. aureus has been highlighted previously (Clements et al., 1999) and is a reflection of its preferential induction by internal superoxide stress.

Regulation of SODs in response to stress has been studied extensively in E. coli (Compan & Touati, 1993; Dubrac & Touati, 2000, 2002; Fee, 1991; Greenberg et al., 1990), where sodA expression is controlled by SoxRS in response to oxidative stress (Compan & Touati, 1993; Greenberg et al., 1990). There is no apparent SoxRS in S. aureus and how superoxide influences sodA expression is unknown. Superoxide induction of the expression of the Mn transporter mntA by an unknown regulator has been noted previously (Horsburgh et al., 2002b). Interestingly, in E. coli, a small RNA molecule, RyhB, has recently been shown to be involved in the positive regulation of sodB (Masse & Gottesman, 2002).

Mn can completely rescue the sensitivity of S. aureus MHKA (sodA) and MHKAM (sodA sodM) to internal superoxide (methyl viologen) stress and the growth defect of MHKAM. Mn homeostasis is important for several micro-organisms in surviving oxidative stress (Archibald & Fridovich, 1981, 1982; Inaoka et al., 1998; Horsburgh et al., 2002b; Tseng et al., 2001). This has been proposed to occur by a direct elemental superoxide scavenging by complexes of Mn with cellular ligands (Archibald & Fridovich, 1981, 1982; Al-Maghrebi et al., 2002; Horsburgh et al., 2002c). The SOD- and Mn-dependent superoxide-scavenging systems act synergistically to relieve the burden of superoxide stress on the cell.

The importance of Mn in the response of S. aureus to internal (methyl viologen) superoxide stress is growth-phase-dependent. S. aureus uses Mn to combat internal superoxide stress encountered mostly during the post-exponential and stationary phases of growth. Previously, Valderas & Hart (2001) showed that the double mutant (sodA sodM) had a growth defect in the absence of methyl viologen. We were able to minimize this defect by supplementing the medium with Mn (results not shown). Also, in the Valderas & Hart study (2001), SodM had a role in resisting internal superoxide stress encountered during the stationary phase of growth. Here, we find a similar slight SodM-dependent effect. Furthermore, they found the sodA mutant was insensitive to methyl viologen at either the post-exponential or stationary phase of growth. Throughout our experiments, it has always been the case that methyl viologen stress induces a virtually identical phenotype for the sodA and sodA sodM mutants that can be attributed to the absence of SodA. The differences between these studies may be related to the strain backgrounds used (RN6390 vs SH1000). RN6390 is of the 8325 lineage and has an rsbU defect which results in greatly reduced SigB activity (Giachino et al., 2001; Horsburgh et al., 2002a).

So far, there have been no reports on the sensitivity of S. aureus to externally generated superoxide stress. Here, we show that Mn has a growth-phase- as well as a SOD-dependent effect on external superoxide stress resistance. Very interestingly, both SodA and SodM have a role in resisting external superoxide stress. Absence of both SODs leads to significantly reduced viability and Mn supplementation is incapable of restoring resistance. It is also apparent that the Mn system contributes to survival in the post-exponential phase of growth, where it is complementary to the SODs in overcoming external superoxide stress.

The different effect of external compared to internal superoxide on the cells is probably a reflection of the targets that might be involved in bringing about its toxicity. It is likely that proteins or lipids in the staphylococcal cell wall or membrane are damaged by external superoxide leading to loss of viability. Also, the fact that SodA, the major internal superoxide-scavenging dismutase, is sufficient to resist external stress supports the idea that external superoxide radicals may enter the cell and disrupt the internal superoxide equilibrium. External superoxide enters the cell via anion channels (Miller & Britigan, 1997). We have so far been unable to find SOD activity in the staphylococcal cell wall, despite testing under many different growth conditions (results not shown).

Contrary to previous studies (Clements et al., 1999; Schneider et al., 2002), we have found that both SodA and SodM contribute to the survival of the organism in an animal model of infection. Clements et al. (1999) used the S. aureus 8325-4 background to show that SodA was not important in virulence. This may be a reflection of the constitutively high levels of SodM homodimer activity observed in 8325-4, a SigB under-producing strain (M. H. Karavolos & S. J. Foster, unpublished data). In a different study by Schneider et al. (2002) conducted in S. aureus 8325-4, lack of SodM did not affect survival of the organism in either a systemic or a kidney model of infection.

These results accentuate the significance of SOD activity for the survival of staphylococci either within the host or in the external environment. It is becoming apparent in a wide range of organisms that this activity can be achieved by complementary mechanisms involving enzyme activity and elemental dismutation. These dual systems allow efficient resistance mechanisms to be maintained throughout growth. Oxidative stress resistance is a key component of the organism's survival machinery and further elucidation of its intricate mechanisms will allow further understanding of S. aureus physiology and pathogenesis.


   ACKNOWLEDGEMENTS
 
This work was supported by the Royal Society (S. J. F.) and the BBSRC (M. H. K., M. J. H.).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Al-Maghrebi, M., Fridovich, I. & Benov, L. (2002). Manganese supplementation relieves the phenotypic deficits seen in superoxide-dismutase-null Escherichia coli. Arch Biochem Biophys 402, 104–109.[CrossRef][Medline]

Archibald, F. S. & Fridovich, I. (1981). Manganese and defenses against oxygen toxicity in Lactobacillus plantarum. J Bacteriol 145, 442–451.[Medline]

Archibald, F. S. & Fridovich, I. (1982). The scavenging of superoxide radical by manganous complexes: in vitro. Arch Biochem Biophys 214, 452–463.[Medline]

Battistoni, A., Pacello, F., Folcarelli, S. & 7 other authors (2000). Increased expression of periplasmic Cu,Zn superoxide dismutase enhances survival of Escherichia coli invasive strains within nonphagocytic cells. Infect Immun 68, 30–37.[Abstract/Free Full Text]

Beauchamp, C. & Fridovich, I. (1971). Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem 44, 276–287.[Medline]

Benov, L. T. & Fridovich, I. (1994). Escherichia coli expresses a copper- and zinc-containing superoxide dismutase. J Biol Chem 269, 25310–25314.[Abstract/Free Full Text]

Beyer, W. F. & Fridovich, I. (1989). Characterization of a superoxide dismutase mimic prepared from desferrioxamine and MnO2. Arch Biochem Biophys 271, 149–156.[Medline]

Bsat, N., Herbig, A., Casillas-Martinez, L., Setlow, P. & Helmann, J. D. (1998). Bacillus subtilis contains multiple Fur homologues: identification of the iron uptake (Fur) and peroxide regulon (PerR) repressors. Mol Microbiol 29, 189–198.[CrossRef][Medline]

Candeias, L. P., Patel, K. B., Stratford, M. R. L. & Wardman, P. (1993). Free hydroxyl radicals are formed on reaction between the neutrophil-derived species superoxide anion and hypochlorous acid. FEBS Lett 333, 151–153.[CrossRef][Medline]

Chan, P. F., Foster, S. J., Ingham, E. & Clements, M. O. (1998). The Staphylococcus aureus alternative sigma factor {sigma}B controls the environmental stress response but not starvation survival or pathogenicity in a mouse abscess model. J Bacteriol 180, 6082–6089.[Abstract/Free Full Text]

Clements, M. O., Watson, S. P. & Foster, S. J. (1999). Characterization of the major superoxide dismutase of Staphylococcus aureus and its role in starvation survival, stress resistance, and pathogenicity. J Bacteriol 181, 3898–3903.[Abstract/Free Full Text]

Compan, I. & Touati, D. (1993). Interaction of six global transcription regulators in expression of manganese superoxide dismutase in Escherichia coli K-12. J Bacteriol 175, 1687–1696.[Abstract]

Darr, D., Zarilla, K. A. & Fridovich, I. (1987). A mimic of superoxide dismutase activity based upon desferrioxamine B and manganese(IV). Arch Biochem Biophys 258, 351–355.[Medline]

Dubrac, S. & Touati, D. (2000). Fur positive regulation of iron superoxide dismutase in Escherichia coli: functional analysis of the sodB promoter. J Bacteriol 182, 3802–3808.[Abstract/Free Full Text]

Dubrac, S. & Touati, D. (2002). Fur-mediated transcriptional and post-transcriptional regulation of FeSOD expression in Escherichia coli. Microbiology 148, 147–156.[Abstract/Free Full Text]

Dussurget, O., Stewart, G., Neyrolles, O., Pescher, P., Young, D. & Marshal, G. (2001). Role of Mycobacterium tuberculosis copper-zinc superoxide dismutase. Infect Immun 69, 529–533.[Abstract/Free Full Text]

Faulkner, K. M., Liochev, S. I. & Fridovich, I. (1994). Stable Mn(III) porphyrins mimic superoxide dismutase in vitro and substitute for it in vivo. J Biol Chem 269, 23471–23476.[Abstract/Free Full Text]

Fee, J. A. (1991). Regulation of sod genes in Escherichia coli: relevance to superoxide dismutase function. Mol Microbiol 5, 2599–2610.[Medline]

Gaballa, A. & Helmann, J. D. (1998). Identification of a zinc-specific metalloregulatory protein, Zur, controlling zinc transport operons in Bacillus subtilis. J Bacteriol 180, 5815–5821.[Abstract/Free Full Text]

Giachino, P., Engelmann, S. & Bischoff, M. (2001). {sigma}B activity depends on RsbU in Staphylococcus aureus. J Bacteriol 183, 1843–1852.[Abstract/Free Full Text]

Greenberg, J. T., Monach, P., Chou, J. H., Josephy, P. D. & Demple, B. (1990). Positive control of a global antioxidant defense regulon activated by superoxide-generating agents in Escherichia coli. Proc Natl Acad Sci U S A 87, 6181–6185.[Abstract]

Guerout-Fleury, A. M., Shazand, K., Frandsen, N. & Stragier, P. (1995). Antibiotic-resistance cassettes for Bacillus subtilis. Gene 167, 335–336.[CrossRef][Medline]

Haber, F. & Weiss, J. (1934). The catalytic decomposition of hydrogen peroxide by iron salts. Proc R Soc London Ser A 147, 332–351.

Hantke, K. (2001). Iron and metal regulation in bacteria. Curr Opin Microbiol 4, 172–177.[CrossRef][Medline]

Hassan, H. & Schrum, L. W. (1994). Roles of manganese and iron in the regulation of the biosynthesis of manganese-superoxide dismutase in Escherichia coli. FEMS Microbiol Rev 14, 315–323.[CrossRef][Medline]

Horsburgh, M. J. & Moir, A. (1999). {sigma}M, an ECF RNA polymerase sigma factor of Bacillus subtilis 168, is essential for growth and survival in high concentrations of salt. Mol Microbiol 32, 41–50.[CrossRef][Medline]

Horsburgh, M. J., Clements, M. O., Crossley, H., Ingham, E. & Foster, S. J. (2001a). PerR controls oxidative stress resistance and iron storage proteins and is required for virulence in Staphylococcus aureus. Infect Immun 69, 3744–3754.[Abstract/Free Full Text]

Horsburgh, M. J., Ingham, E. & Foster, S. J. (2001b). In Staphylococcus aureus, Fur is an interactive regulator with PerR, contributes to virulence, and is necessary for oxidative stress resistance through positive regulation of catalase and iron homeostasis. J Bacteriol 183, 468–475.[Abstract/Free Full Text]

Horsburgh, M. J., Aish, J. L., White, I. J., Shaw, L., Lithgow, J. K. & Foster, S. J. (2002a). {sigma}B modulates virulence determinant expression and stress resistance: characterization of a functional rsbU strain of Staphylococcus aureus 8325-4. J Bacteriol 184, 5457–5467.[Abstract/Free Full Text]

Horsburgh, M. J., Wharton, S. J., Cox, A. G., Ingham, E., Peacock, S. & Foster, S. J. (2002b). MntR modulates expression of the PerR regulon and superoxide resistance in Staphylococcus aureus through control of manganese uptake. Mol Microbiol 44, 1269–1286.[CrossRef][Medline]

Horsburgh, M. J., Wharton, S. J., Karavolos, M. H. & Foster, S. J. (2002c). Manganese: elemental defence for a life with oxygen? Trends Microbiol 10, 496–501.[CrossRef][Medline]

Hurst, J. K. & Barrette, W. C., Jr (1989). Leukocyte oxygen activations and microbicidal oxidative toxins. Crit Rev Biochem Mol Biol 24, 271–328.[Abstract]

Imlay, J. A. & Linn, S. (1988). DNA damage and oxygen radical toxicity. Science 240, 1302–1309.[Medline]

Imlay, K. R. & Imlay, J. A. (1996). Cloning and analysis of sodC, encoding the copper-zinc superoxide dismutase of Escherichia coli. J Bacteriol 178, 2564–2571.[Abstract]

Inaoka, T., Matsumura, Y. & Tsuchido, T. (1998). Molecular cloning and nucleotide sequence of the superoxide dismutase gene and characterization of its product from Bacillus subtilis. J Bacteriol 180, 3697–3703.[Abstract/Free Full Text]

Inaoka, T., Matsumura, Y. & Tsuchido, T. (1999). SodA and manganese are essential for resistance to oxidative stress in growing and sporulating cells of Bacillus subtilis. J Bacteriol 181, 1939–1943.[Abstract/Free Full Text]

Kanafani, H. & Martin, S. E. (1985). Catalase and superoxide dismutase activities in virulent and nonvirulent Staphylococcus aureus isolates. J Clin Microbiol 21, 607–610.[Medline]

Kemp, E. H., Sammons, R. L., Moir, A., Sun, D. & Setlow, P. (1991). Analysis of transcriptional control of the gerD spore germination gene of Bacillus subtilis 168. J Bacteriol 173, 4646–4652.[Medline]

Kim, E.-J., Chung, H.-J., Suh, B. & Roe, J.-H. (1998). Transcriptional and post-transcriptional regulation by nickel of sodN gene encoding nickel-containing superoxide dismutase from Streptomyces coelicolor Muller. Mol Microbiol 27, 187–195.[CrossRef][Medline]

Kono, Y., Takahashi, M. A. & Asada, K. (1976). Oxidation of manganous pyrophosphate by superoxide radicals and illuminated spinach chloroplasts. Arch Biochem Biophys 174, 454–462.[Medline]

Lindsay, J. A. & Foster, S. J. (2001). zur: a Zn2+-responsive regulatory element of Staphylococcus aureus. Microbiology 147, 1259–1266.[Abstract/Free Full Text]

Mandell, G. L. (1975). Catalase, superoxide dismutase, and virulence of Staphylococcus aureus. In vitro and in vivo studies with emphasis on staphylococcal–leukocyte interaction. J Clin Invest 55, 561–566.[Medline]

Marklund, S. & Marklund, G. (1974). Involvement of the superoxide anion radical in the autoxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur J Biochem 47, 469–474.[Medline]

Masse, E. & Gottesman, S. (2002). A small RNA regulates the expression of genes involved in iron metabolism in Escherichia coli. Proc Natl Acad Sci U S A 99, 4620–4625.[Abstract/Free Full Text]

Miller, R. A. & Britigan, B. E. (1995). The formation and biologic significance of phagocyte-derived oxidants. J Investig Med 43, 39–49.[Medline]

Miller, R. A. & Britigan, B. E. (1997). Role of oxidants in microbial pathophysiology. Clin Microbiol Rev 10, 1–18.[Abstract]

Novick, R. P. (1991). Genetic systems in staphylococci. Methods Enzymol 204, 587–636.[Medline]

O'Halloran, T. V. (1993). Transition metals in control of gene expression. Science 261, 715–725.[Medline]

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]

Pugh, S. Y. R., DiGuiseppi, J. L. & Fridovich, I. (1984). Induction of superoxide dismutases in Escherichia coli by manganese and iron. J Bacteriol 160, 137–142.[Medline]

Que, Q. & Helmann, J. D. (2000). Manganese homeostasis in Bacillus subtilis is regulated by MntR, a bifunctional regulator related to the diphtheria toxin repressor family of proteins. Mol Microbiol 35, 1454–1468.[CrossRef][Medline]

Ramos, C. L., Pou, S., Britigan, B. E., Cohen, M. S. & Rosen, G. M. (1992). Spin trapping evidence for myeloperoxidase-dependent hydroxyl radical formation by human neutrophils and monocytes. J Biol Chem 267, 8307–8312.[Abstract/Free Full Text]

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

Schenk, S. & Ladagga, R. A. (1992). Improved method for electroporation of Staphylococcus aureus. FEMS Microbiol Lett 73, 133–138.[Medline]

Schneider, W. P., Ho, S. K., Christine, J., Yao, M., Marra, A. & Hromockyj, A. E. (2002). Virulence gene identification by differential fluorescence induction analysis of Staphylococcus aureus gene expression during infection-simulating culture. Infect Immun 70, 1326–1333.[Abstract/Free Full Text]

Tseng, H.-J., Srikhanta, Y., McEwan, A. G. & Jennings, M. P. (2001). Accumulation of manganese in Neisseria gonorrhoeae correlates with resistance to oxidative killing by superoxide anion and is independent of superoxide dismutase activity. Mol Microbiol 40, 1175–1186.[CrossRef][Medline]

Valderas, M. W. & Hart, M. E. (2001). Identification and characterization of a second superoxide dismutase gene (sodM) from Staphylococcus aureus. J Bacteriol 183, 3399–3407.[Abstract/Free Full Text]

Xiong, A., Singh, V. K., Cabrera, G. & Jayaswal, R. K. (2000). Molecular characterization of the ferric-uptake regulator, fur, from Staphylococcus aureus. Microbiology 146, 659–668.[Abstract/Free Full Text]

Yesilkaya, H., Kadioglu, P., Gingles, N., Alexander, J. E., Mitchell, T. J. & Andrew, P. W. (2000). Role of manganese-containing superoxide dismutase in oxidative stress and virulence of Streptococcus pneumoniae. Infect Immun 68, 2819–2826.[Abstract/Free Full Text]

Received 17 March 2003; revised 22 May 2003; accepted 23 May 2003.