pmc.ncbi.nlm.nih.gov

Role of Manganese-Containing Superoxide Dismutase in Oxidative Stress and Virulence of Streptococcus pneumoniae

Abstract

Streptococcus pneumoniae was shown to contain two types of superoxide dismutase, MnSOD and FeSOD. Levels of MnSOD increased during growth in an aerobic environment. The sodA gene, encoding MnSOD, of virulent S. pneumoniae type 2 strain D39 was inactivated to give mutant D39HY1. Aerobically, D39HY1 had a lower growth rate than the wild type and exhibited susceptibility to the redox-active compound paraquat, but anaerobic growth of D39HY1 was identical to that of the wild type. Virulence studies showed that the median survival time of mice infected intranasally with D39HY1 was significantly longer than that of mice infected with the wild-type pneumococcus. In contrast to the wild type, D39HY1 did not multiply in lungs during the first 24 h but thereafter grew at the same rate as the wild type. Appearance in the bloodstream was also delayed, but growth in the blood was unimpaired by the sodA mutation. The pattern of inflammation in lungs infected with D39HY1 differed from that in wild-type-infected mice. After infection with D39HY1, neutrophils were densely packed around bronchioles, in contrast to the wild-type infection, where neutrophils were more diffusely localized.

Streptococcus pneumoniae is the etiological agent of pneumonia, otitis media, and meningitis in humans (1, 36). Increasing antibiotic resistance (26, 39) and drawbacks of the current vaccine (2) have increased the need for a better understanding of the pathogenesis of pneumococcal diseases. Several bacterial components, including the capsule, the cell wall, and some pneumococcal protein products, such as pneumococcal surface protein A, pneumolysin, and neuraminidase, are thought to play a role in the virulence of pneumococci (36). There are, however, many questions about pathogenic mechanisms and the basic metabolism of the pneumococcus that remain to be answered. One of them is how the pneumococcus copes with oxygen and its reactive derivatives.

Reactive oxygen intermediates, including superoxide anion (·O2−), hydrogen peroxide (H2O2), and hydroxyl radical (·OH−), have many deleterious effects on living organisms, ranging from DNA strand damage to peroxidation of membrane lipids (15, 25, 47). Sources of reactive oxygen intermediates are abundant and include incomplete reduction of oxygen during respiration, exposure to radiation or to redox-active compounds, and the respiratory burst of phagocytes (18). Aerobic organisms, however, have developed several enzymatic and nonenzymatic mechanisms to detoxify these very active compounds. Enzymatically, oxygen radicals are removed mainly by the action of four enzymes: superoxide dismutase, catalase, glutathione peroxidase, and glutathione reductase (7).

Superoxide dismutases (SODs) are metalloenzymes that catalyze the conversion of superoxide molecules to hydrogen peroxide and molecular oxygen and therefore form one of the cell's major defense mechanisms against oxidative stress (33). In prokaryotes, depending on the metal cofactor, three types of SODs have been described: Cu-Zn- (SodC), Fe- (SodB), or Mn-type (SodA) SODs. FeSOD and MnSOD are characteristically prokaryote enzymes, but MnSOD is also present in mitochondria of eukaryotes. Cu-ZnSOD, on the other hand, is primarily found in the cytosol of many eukaryotic organisms (19). However, several prokaryotes containing Cu-ZnSOD have recently been reported (29, 45). As well as being important in the detoxification of superoxide radical during aerobic metabolism, SOD has also been shown to be a virulence factor for several pathogenic bacteria. SOD in Nocardia asteroides was found to protect the bacterium from the oxidative killing mechanisms of neutrophils during infection (3). In another study done with a Shigella flexneri ΔsodB mutant, the mutant strain was less virulent than wild-type bacteria in the rabbit ligated ileal loop model (16). Pesci and coworkers (37) showed that a sodB insertion mutation in Campylobacter jejuni caused a significant reduction in the ability of the bacterium to invade an embryonic intestinal cell line. Inactivation of sodA in Yersinia enterocolitica resulted in a marked reduction in virulence of the organism in a mouse infection model after intravenous infection when compared to its parental strain (40). More recently, it was reported that a Cu-ZnSOD mutant strain of Neisseria meningitidis was less virulent in a mouse intraperitoneal infection model than the wild-type strain (46). On the other hand, a sodA deletion mutation of Bordetella bronchiseptica and Bordetella pertussis did not affect the virulence of these species in mice following intranasal infection (17).

S. pneumoniae is a facultative anaerobe that lacks catalase (18a). Thus, the absence of this enzyme suggests that SOD may play a critical role against oxidative stress, affecting both the survival and, consequently, the virulence of the organism. In this paper, we report some characterization of a pneumococcal SOD. An isogenic sodA insertion mutant strain of S. pneumoniae was constructed in order to study the role of the enzyme during in vitro growth, and the mutant also was tested in a mouse model of bacteremia and pneumonia.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions.

Virulent, capsular type 2 S. pneumoniae strain D39 was used to construct the isogenic sodA-deficient mutant. The parent strain and its sodA-deficient derivative were routinely grown in brain heart infusion (BHI) broth or on blood agar base containing 7% (vol/vol) horse blood. Where appropriate, a defined medium containing 0.4% (wt/vol) sucrose, originally described by Lacks (27), was used. Overnight incubation of the pneumococcal cultures was done at 37°C in an anaerobic jar (Gas Pak; Becton Dickinson) or on a shaker (aerobic growth). Escherichia coli strains DH5α and XL1 Blue were grown in Luria-Bertani broth or on Luria-Bertani agar plates. Where appropriate, spectinomycin and ampicillin were added to growth media at the following concentrations: ampicillin at 100 μg/ml and spectinomycin at 200 μg/ml for E. coli or at 1 mg/ml for S. pneumoniae.

Recombinant DNA techniques.

Plasmid DNA preparation from E. coli was done using ion exchange columns from QIAGEN (Hilden, Germany). Pneumococcal chromosomal DNA was isolated as described previously (41). Standard methods were followed for restriction digestion, ligation, gel electrophoresis of DNA, and Southern blotting (42). DNA was introduced in E. coli by electrotransformation (48). Competence-stimulating peptide was used to transform encapsulated S. pneumoniae (23). PCR amplifications were carried out in a thermal cycler (Omnigene HBTR3CM; Hybaid Ltd., Teddington, United Kingdom) with 50 ng of template DNA primers at a final concentration of 0.1 μM, each deoxynucleotide at 200 μM, 1× reaction buffer (20 mM Tris-HCl, pH 8.4, and 50 mM KCl), 2 mM MgCl2, and 2 U of Taq DNA polymerase. The PCR mixture was subjected to a denaturation step (3 min at 95°C), followed by 35 cycles of amplification (30 s of denaturation at 95°C, 2 min of annealing at 50°C, 90 s of elongation at 72°C) and elongation (12 min at 72°C) steps. Sequence analyses were done manually by using a Sequenase 2.0 sequencing kit (U. S. Biochemicals).

Primers.

A 480-bp internal region of the sodA gene (sodint) was amplified by PCR from S. pneumoniae by using the degenerate primers HYK1 (5′-CCGTA[CT]GCGTA[CT]GA[CT]GCG[CT]TGGA[AG]CC-3′) and HYK2 (5′-A[AG][AG]TA[AG]TAGGC[AG]TG[CT]TCCCAGAC[AG]TC-3′), whose sequences were based on conserved amino acids from microbial SODs PY(PAT)YDALEP and DVWEHAYYL, respectively (38). (Nucleotides in brackets were used as alternatives to the one left of the bracket; for example, C and T could be used instead of A.) The primer SOD4 (5′-GGCGCGCCCTCCTTCAGCAGAACTGGCA-3′) anneals to nucleotides 257 to 276 within sodint and generated an AscI site. Primer SOD5 (5′-GGCGCGCCCAATTCCCAGAAAAGAGCGT-3′) binds to nucleotides 244 to 263 of the complementary strand of sodint and created an AscI site.

The oligonucleotide primers spec1 (5′-CAGGCGCGCCATCGATTTTCGTTCGTGAAT-3′) and spec2 (5′-AAGGCGCGCCCATATGCAAGGGTTTATTGT-3′), which add AscI sites to 5′ and 3′ ends, respectively, were used to amplify a spectinomycin resistance gene from pDL278 (30). The SOD6 (5′-TAGCAGCTGCGTACAATTCAT-3′) primer based on the 3′ end of the entire sod gene, which was discovered during the course of this work (data not shown), was used to detect insertion of the spectinomycin resistance cassette into the pneumococcal chromosome. When this primer was used in combination with primer HYK1, a 577-bp sod gene fragment was amplified.

Construction of S. pneumoniae ΔsodA mutant.

A 480-bp internal fragment of the sodA gene (sodint), representing approximately 85% of the gene, was obtained by PCR by using the primers HYK1 and HYK2. This fragment was cloned into pPCR-Script Amp SK(+) to produce pSOD1. The primers SOD4 and SOD5 were used to introduce a unique AscI site into the sodint fragment in pSOD1 by reverse PCR. This site was needed to ligate a spectinomycin resistance gene cassette that was used as a selective marker for transformants. The resulting plasmid was designated pSOD2. A spectinomycin resistance gene carrying 5′ and 3′ AscI sites was amplified from pDL278 (14) with the spec1 and spec2 primers. Following digestion with AscI, pSOD2 and the spectinomycin gene were ligated, and the resulting construct was designated pHY2. Pneumococci were transformed with pHY2, and spectinomycin-resistant colonies were selected. Because pHY2 is unable to replicate in gram-positive bacteria, recombinants were anticipated to be spectinomycin resistant by virtue of the cassette entering the chromosome by homologous recombination via sod sequences.

Sequencing strategy.

The sequence of sodA in D39 that corresponded to that reported by Poyart et al. (38) was determined by sequencing PCR fragments generated with primers HYK1 and HYK2. To obtain the 5′ end of the gene, we took advantage of the presence of a spectinomycin resistance gene cassette within sodA of D39HY1. PstI-digested DNA from D39HY1 was inserted into pBluescript SK(+), selecting for spectinomycin-resistant recombinants. The 5′ sequence of sodA was obtained from analysis of the plasmid (pPSTSOD) from one of these recombinants. The sequence of the 3′ end of the gene was obtained by first screening an available lambda ZAP expression library of a type 1 pneumococcus by using radiolabelled sodint fragment as the probe. As a result, a strongly hybridizing clone, designated pHYSOD1, was isolated. Sequence analysis of pHYSOD1 provided the 3′ end, and downstream sequences, of sodA in the type 1 strain. From this information, a primer downstream of sodA was used with a primer from within the gene to amplify by PCR the 3′ end of sodA from D39. These PCR fragments were then sequenced.

Preparation of bacterial cell protein.

All manipulations were carried out at 4°C. Cell pellets of late-log-phase cultures were collected by centrifugation at 2,000 × g for 15 min and were resuspended in 50 mM phosphate buffer, pH 7. This cell suspension was sonicated by using a Sanyo Soniprep model 150 sonicator, and cell debris was removed by centrifugation at 22,000 × g for 30 min. Total protein concentration of the lysate was determined by the Bradford assay; bovine serum albumin was used as a standard (5).

SOD activity assays.

Extracts of S. pneumoniae, prepared as described above, were separated by polyacrylamide gel electrophoresis (PAGE) under nondenaturing conditions. Electrophoresis was carried out by a modification of the Davis (12) procedure as described by Langford et al. (28). The gels were then stained for SOD activity by nitroblue tetrazolium, as described previously (4). To assess the type of SOD, 5 mM KCN or 5 mM H2O2 was added to the staining solution (11). Where appropriate, a cell extract of E. coli QC771, as a source of MnSOD and FeSOD (prepared as described above), or bovine erythrocyte Cu-ZnSOD (Sigma) was used as a control. SOD was assayed quantitatively by the xanthine oxidase-cytochrome c method (33).

Hydrogen peroxide assay.

Bacteria were grown until mid-log phase, were harvested by centrifugation, and were resuspended in an equal volume of a solution containing 50 mM Tris-HCl, pH 7.5, and 10 mM glucose. The bacterial culture was incubated at 37°C for 30 min, and cells were removed by centrifugation, as described by Spellerberg et al. (44). Bacterial cells (4 × 107) were used to measure hydrogen peroxide production fluorimetrically by the method of Jackett et al. (24), in which the level of H2O2 was assayed by following the oxidation of p-hydroxyphenyl acetic acid by H2O2 catalyzed by horseradish peroxidase.

In vivo virulence test.

Female MFI outbred mice, weighing 30 to 35 g, were obtained from Harlan Olac (Bicester, United Kingdom). A standardized inoculum was prepared as follows: 200 μl of an overnight culture of pneumococcal strains was injected into the peritoneal cavity of mice and was recovered 24 h later from cardiac blood. This was done to standardize the virulence of inocula. Passaged bacteria were grown overnight at 37°C in BHI broth, were harvested by centrifugation, were resuspended in 1 ml of BHI containing 20% (vol/vol) fetal calf serum, and then were diluted with fresh BHI-fetal calf serum to give an A600 of 0.7. The culture was incubated at 37°C for 4 to 5 h. The viability of the suspension was determined by serial dilution in sterile nanopure water and plating on blood agar plates in duplicate. Aliquots of the suspension were stored at −70°C until required. Under these conditions, the viability of strains was unaffected for at least 1 month. When needed, an aliquot was thawed slowly at room temperature, and bacteria were harvested by centrifugation before resuspension in sterile phosphate-buffered saline (PBS).

To determine the virulence of pneumococcal strains, mice were lightly anesthetized with 2.5% (vol/vol) fluothane (Zeneca, Macclesfield, United Kingdom) over oxygen (1.5 to 2 liters/min), administered with a calibrated vaporizer. A 50-μl sample of PBS containing 106 CFU of S. pneumoniae was given into the nostrils. Mice were monitored for clinical symptoms (progressively starry coat, hunched, lethargic, and then moribund) for 14 days, and those that became moribund were accepted to have reached the end point of the assay and were killed humanely.

To measure in vivo growth of bacteria, preselected groups of mice were deeply anesthetized with 5% (vol/vol) fluothane at prechosen intervals following infection and blood was collected by cardiac puncture. Following this, mice were killed by cervical dislocation. The lungs were removed into 10 ml of sterile distilled water, were weighed, and then were homogenized in a stomacher laboratory blender (Seward Medical, London, United Kingdom). Viable counts in homogenates and in blood were determined as above on selective and nonselective media. In the case of intravenous infection, the desired number of bacteria was given in 100 μl of sterile PBS through a tail vein.

Histology.

Lung samples were embedded in Tissue Tek OCT (Miles) tissue embedding compound, were snap-frozen in liquid nitrogen, and were stored at −70°C until required. Before sectioning, samples were kept at −20°C overnight. Sections (20 μm) were cut with a cryostat and were subsequently stained with hematoxylin and eosin. One observer did a histological analysis of sections, blinded to the identity of samples, by using bright field microscopy.

Statistical analysis.

Data were analyzed by analysis of variance followed by Tukey-Kramer multiple comparisons test or by Mann-Whitney U test, as appropriate. Growth rates were analyzed by comparison of regression slopes.

Nucleotide sequence accession number.

The sequence of pneumococcal sod has been deposited in GenBank under accession no. AF162664.

RESULTS

Construction of S. pneumoniae ΔsodA mutant.

A ΔsodA mutant was made in the virulent type 2 strain D39, as described in Materials and Methods. PCR and Southern blot analysis (data not shown) confirmed disruption of the sod gene in the chromosome of transformants. Furthermore, none of the transformants hybridized with the pPCR-Script Amp SK(+) probe, indicating that a double-crossover event had occurred (data not shown). One transformant, designated D39HY1, was chosen for further study.

Sequence of the sodA gene and site of insertion in D39HY1.

A partial sequence of pneumococcal sod had been published (38). We determined the sequence of the remaining parts of the gene and deposited the information in GenBank (accession no. AF162664). The internal sequence of sodA agreed exactly with the sodint sequence of Poyart et al. (38). By comparison, the pneumococcal SOD most resembled SodA of known bacterial species (data not shown). Furthermore, analysis of the amino acid sequence of the pneumococcal SOD showed that it contained residues that are unique to MnSODs (35), such as glycine at positions 76 and 77 and phenylalanine, glutamine, and aspartic acid at positions 85, 151, and 152, respectively (data not shown). A putative rho-independent transcription terminator (GGAGGGAAGAATTGTTCTTCTCTTT; ΔG = −12.04 kcal/mol) occurs downstream of sodA. The sequencing also showed that the site of insertion of the spectinomycin resistance gene was between nucleotides 279 and 280.

SOD activity detection in D39HY1 and in the wild-type strain.

Separation of protein extracts from the whole cells under nondenaturing polyacrylamide gel conditions showed two electrophoretically distinct activity bands for the wild type (Fig. 1A, lane 2). The intensities of these bands were considerably different, as were their mobilities. The activity of these bands was not affected by treatment with 5 mM KCN, indicating the lack of a Cu-ZnSOD in the pneumococcus (data not shown). However, H2O2 inhibited the activity of the minor band in the wild type (Fig. 1B, lane 2) and in D39HY1 (data not shown). This suggests that the minor band was an iron-cofactored SOD. The major band was partially inhibited by H2O2. By exclusion, the major SOD in the pneumococcus appears to be MnSOD; MnSOD is not affected by KCN and is either fully or partially resistant to H2O2 (7, 11, 32). In the mutant strain, however, only the minor activity band was seen (Fig. 1C, compare lanes 1 and 2), confirming that D39HY1 is a sodA mutant.

FIG. 1.

FIG. 1

Type inhibition assay and insertional inactivation of SOD activity. Nondenaturing PAGE gels for determination of SOD activity were treated (B) or not treated (A) with 5 mM H2O2. (A and B) Lanes 1 contain E. coli extract and lanes 2 contain S. pneumoniae extract. (C) Extract of the wild type (lane 2) or mutant (D39HY1) (lane 1) was separated on nondenaturing PAGE gels. The lane containing the wild type had two bands of activity, a minor and a major band, whereas D39HY1 had only the minor activity band. Arrows indicate the minor and major SODs.

Aeration increases the expression of SOD.

The SOD activity of wild-type D39 grown anaerobically was 2.4 U of protein per mg (±0.1; n = 3), whereas aerated cultures had 4.1 U of protein per mg (±0.49; n = 3) (P < 0.05). In contrast, the SOD activity of D39HY1 was not statistically different (P > 0.05) when grown anaerobically or aerobically: 0.5 U/mg (±0.04; n = 3) anaerobically and 1.0 U/mg (±0.06; n = 3) aerobically. This activity in D39HY1 represents the activity of the FeSOD.

Hydrogen peroxide production in D39HY1.

Since the catalytic activity of superoxide dismutase generates H2O2, we measured the ability of D39HY1 to generate this molecule. H2O2 production was quantified in the mid-log-phase cultures by a fluorimetric assay. There was no statistical difference (P > 0.05) between the mutant and the wild type in H2O2 production, the mean of the four experiments being 51 nmol/107 bacteria (±1.72) for D39HY1 and 50 nmol/107 bacteria (±1.35) for the wild type.

In vitro growth characteristics of the D39HY1.

The colony size of the mutant strain was comparable to that of the wild type on blood agar plates kept under anaerobic conditions. Aerobically, however, D39HY1 grew slower than wild type. D39HY1 formed alpha-hemolytic colonies under both conditions on blood agar. To examine possible effects of the sodA mutation, the SOD-deficient strain was tested under aerobic and anaerobic growth conditions in broth culture. In addition, the redox-active compound paraquat was added to the culture medium as a superoxide generator to assess any effect of excess superoxide on the mutant strain.

The rate of increase in the optical density (OD) of the wild-type strain did not change under aerobic or anaerobic conditions (P > 0.05) (Fig. 2A and B). However, D39HY1 had a lower rate of aerobic increase than anaerobic increase (P < 0.001) (Fig. 2A and B). D39HY1 had a rate statistically similar to the wild type under anaerobic conditions (P > 0.05), but aerobically, it had a significantly lower rate of increase in OD than the wild type (P < 0.001). The rate of increase of the wild-type strain was not significantly affected by any concentration of paraquat used, under either aerobic or anaerobic conditions (P > 0.05) (Fig. 2A and B), but addition of 0.001 M paraquat affected the rate of increase of D39HY1 in both conditions (P < 0.001).

FIG. 2.

FIG. 2

In vitro growth of wild type and sod mutant strain (D39HY1). Aerobic growth (A) and anaerobic growth (B) in BHI broth with and without paraquat and growth in defined medium (C) are shown. (A and B) Strains and concentrations of paraquat are represented with symbols as follows: wild type with no paraquat (Inline graphic), wild type with 0.001 M paraquat (Inline graphic), wild type with 0.0001 M paraquat (Inline graphic), D39HY1 with no paraquat (Inline graphic), D39HY1 with 0.001 M paraquat (Inline graphic), and D39HY1 with 0.0001 M paraquat (Inline graphic). (C) Strains and growth conditions are represented with symbols as follows: wild type, anaerobic conditions (Inline graphic); wild type, aerobic conditions (Inline graphic); D39HY1, anaerobic conditions (Inline graphic); and D39HY1, aerobic conditions (Inline graphic). Each point is the mean of three experiments.

In addition to growth characteristics in complex medium, we investigated growth of the mutant strain in defined medium (Fig. 2C). Anaerobically, the wild type and D39HY1 showed statistically similar rates of increase of OD (P > 0.05), but D39HY1 was unable to grow under aerobic conditions in defined medium.

In vivo studies.

The median survival time of mice was determined after intranasal infection with D39HY1 or the wild type to investigate if the sodA mutation had any effect on the ability of S. pneumoniae to cause disease. Infected mice were observed for 7 days, and the development of clinical symptoms was recorded. Around 19 h postinfection, individual mice within the wild-type-infected group began to show symptoms (starry coat), and all mice in this group showed symptoms by 45 h. However, none of the D39HY1-infected mice showed any sign of illness by 45 h postinfection, at which point three mice in the wild-type-infected group were moribund. No animal in the D39HY1-infected group showed the initial symptoms until 47 h postinfection, and the first moribund case in this group was not recorded until 69 h after infection. Within 7 days, all 14 mice in each group became moribund (the end point of the assay). Statistical analysis showed that the D39HY1-infected group survived significantly longer (107 h) than the wild-type-infected mice (50 h) (P < 0.001) (Fig. 3).

FIG. 3.

FIG. 3

Survival of mice infected intranasally with D39HY1 or the wild-type (WT) pneumococcus. Symbols show the time that individual mice became moribund. The horizontal bars mark the median time to the moribund state.

To further investigate the nature of impaired virulence, we went on to investigate the growth of bacteria in blood and lungs of mice. Significantly fewer bacteria were recovered in the lungs of mice infected with D39HY1 than in the wild-type-infected mice 24 and 48 h after infection (P < 0.05) (Fig. 4A). The explanation for this appears to be a lag in the first 24 h with D39HY1, which did not occur with the wild type. After 24 h, the rates of growth in the lungs of D39HY1 and the wild type were not significantly different (0.15 h−1 and 0.09 h−1, respectively; P > 0.05). D39HY1 is apparently less invasive than the wild type, since, in contrast to the wild type, the mutant was not recovered from blood at 24 h (Fig. 4B). However, by 48 h, the level of D39HY1 in the blood was similar to that of the wild type, and the concentration of D39HY1 in blood had reached approximately 109 bacteria/ml by 72 h. The maximum growth rates for D39HY1 and the wild type were the same (0.51 h−1).

FIG. 4.

FIG. 4

Time course of bacterial growth in the lungs (A) and blood (B) of mice infected intranasally with D39HY1 or the wild-type pneumococcus. Symbols indicate mice infected with the wild type (Inline graphic) or D39HY1 (Inline graphic). Each point is the mean log data from five MF1 female mice except 72 h, which is from three mice. Error bars show the standard errors of the means.

Changes in growth rate due to emergence of revertants were discounted because at each time point the number of colonies was the same when determined on selective or nonselective medium (data not shown). Furthermore, Southern blotting of mutants recovered from lungs, using the spectinomycin resistance cassette as the probe, confirmed the presence of the antibiotic resistance gene (data not shown).

In addition to intranasal challenge, the virulence of the mutant strain was tested by observing survival for 7 days after intravenous infection. All mice survived infection with 50 CFU, but both of the two mice in each group infected with 5 × 102 and 1 × 103 CFU became moribund. Following intravenous infection with 104 CFU, the rates of growth of the wild type and D39HY1 were found to be similar: 0.17 and 0.22 h−1, respectively (P > 0.05). There also was no difference seen in the median survival times of the mice infected intravenously with D39HY1 (46 h) and those infected with the wild type (45 h) (P > 0.05).

Histological examination of lung tissue.

Histological analysis of lung tissue sections from mice infected with wild-type pneumococci showed inflammation and cellular infiltration centered around bronchioles and perivascular areas (Fig. 5A). The foci of inflammation were restricted to certain bronchioles and perivascular areas close to these bronchioles at 24 h postinfection. Inflammation presented itself as bronchiole wall thickening and heavy cellular infiltration around such bronchioles.

FIG. 5.

FIG. 5

Lung histology. Light microscopy of cryostat cut frozen sections (20 μm) of lung tissue from mice infected with 106 wild-type bacteria and sacrificed at 24 h (panel A; magnification, ×400; large arrow, bronchiole wall; pair of small arrows, cellular infiltration; open arrowhead, perivascular space) and 48 h (panel B; magnification, ×400; large arrow, heavy cellular infiltration; pair of large arrows, fibrosis; pair of small arrows, edema; open arrowhead, bronchiole wall) and from mice infected with 106 D39HY1 bacteria and sacrificed at 24 h (panel C; magnification, ×300; large arrow, bronchiole wall; pair of small arrows, heavy cellular infiltration; open arrowhead, perivascular space) and 48 h (panel D; magnification, ×400; large arrow, bronchiole wall; pair of small arrows, heavy cellular infiltration; open arrowhead, perivascular space). Sections were stained with hematoxylin and eosin.

By 48 h postinfection, bronchiole wall thickening had increased, along with the emergence of fibrous tissue (Fig. 5B). Additionally, cellular infiltration had increased with the extension of inflammatory cells from bronchioles and perivascular areas into the surrounding lung parenchyma, with focal areas of consolidation becoming larger and more diffuse. The presence of alveoli in lung sections at this time point was hardly distinguishable due to intensive tissue edema.

Mice infected with D39HY1 showed less tissue inflammation and cellular infiltration into perivascular areas between infected bronchioles than wild-type-infected mice at both time points (Fig. 5C and D). However, despite this and lower levels of pathologic tissue damage, some bronchioles exhibited levels of inflammation comparable to wild-type-infected mice at 24 h. Bronchioles also exhibited substantial levels of cellular infiltration 48 h postinfection. Compared to wild-type-infected mice, however, the cellular infiltration around such bronchioles appeared to be more intense and localized and did not extend into the perivascular areas.

DISCUSSION

Our data revealed that the pneumococcus has two types of SOD, an MnSOD as the major fraction and an FeSOD as a minor fraction. Identification of the minor band as FeSOD was on the basis of its sensitivity to inhibition by H2O2. The occurrence of this band from D39HY1 confirmed that it was a product of a separate gene and not due to Fe cofactoring the sodA product. The major band was identified as an MnSOD because of the extent of its resistance to H2O2 and on the basis of the gene sequence. MnSODs are classically described as being completely resistant to H2O2, but the pneumococcal SOD appeared to be slightly sensitive to H2O2. This might be because a proportion of the SOD in the major band is in the form of a hybrid SOD, cofactored in the main by Mn but with some Fe also. Such a hybrid would have limited sensitivity to H2O2 (10). However, the expectation that a hybrid would have a pI different from those of the other two SODs and appear as a third band (5) contradicts this idea.

Lack of Cu-ZnSOD in the pneumococcal whole-cell extract was suggested by the type inhibition assays. This conclusion was supported by an analysis of the pneumococcal genome, available from The Institute for Genomic Research through the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov), by using two Cu-ZnSOD consensus sequences: GFHIHENGSCG and GGGGARIACGVI. No significant similarities were found.

The amount of SOD in pneumococci varied depending on aeration. Although no significant change in the amount of FeSOD occurred, more MnSOD activity could be detected under aerobic conditions than under anaerobic conditions. These findings are consistent with observations with other bacteria (13, 20, 21). We took this observation as suggesting that the ability to adapt to oxidative stress was part of the virulence of this respiratory pathogen. Thus, to examine the role of the MnSOD for pneumococci, we mutated the sodA gene of S. pneumoniae. Our supposition was supported by the observation that the mutant, D39HY1, had an impaired ability to survive oxidative stress either from an oxygenated environment or due to a flux of superoxide from paraquat. In a defined medium, which presumably lacks the superoxide scavengers present in complex media, inactivation of sodA was lethal for pneumococci in oxygen. Impaired growth appears to be a common characteristic of bacteria deficient in cytosolic SOD, being also observed in sodA mutants of Streptococcus mutans (34), Haemophilus influenzae (13), Pseudomonas aeruginosa (22), and Lactobacillus lactis (43) and in E. coli sodA sodB mutants (9).

The susceptibility of D39HY1 to paraquat under anaerobic conditions was unexpected, because in the absence of oxygen, paraquat does not give rise to superoxide (31). An explanation is that strict anaerobic conditions were not achieved in this study. If this is so, then we are probably underestimating the increase in SodA when pneumococci move from anaerobic to aerobic conditions.

In vivo analysis of D39HY1 in a mouse model of pneumonia and bacteremia showed that sod mutation reduced the virulence of S. pneumoniae after intranasal infection. Reduction in virulence was dependent, however, on the route of infection, as no difference was seen between the wild type and D39HY1 when infection was intravenous. These observations presumably reflect differences in oxidative stress for pneumococci in the lungs and the blood. When infection is by the intranasal route, the crucial oxidative stress is found within the lungs, since once the bacteria invaded the bloodstream, no effect of impairment of growth in oxygen was seen. The conclusion that SodA protects against externally derived oxidative stress rather than against pneumococcus-derived superoxide is supported by the identity in peroxide production by D39HY1 and the wild type. This observation was consistent with the report (44) that 99% of the peroxide produced by the pneumococci is derived from pyruvate oxidase.

It was during the early stages after intranasal infection of the lungs that the effect of impairment of ability to grow in an aerobic environment was seen. Only after the first 24 h in the lungs did D39HY1 begin to grow in the lungs, in contrast to the wild type. This could reflect the time taken for the bacteria to move from oxygen-rich environments in the alveoli and bronchioles into more microaerophilic areas. The ability to survive in oxygen-rich areas appears to be part of the repertoire for invasion. Thus, wild-type pneumococci, with oxygen resistance mechanisms intact, rapidly move from the bronchioles and invade the bloodstream, whereas pneumococci without SodA remain for a time within the lungs. Nevertheless, these pneumococci do eventually become invasive, although not as a result of the loss of the mutagenizing antibiotic resistance gene. Why D39HY1 should eventually begin to grow within the lungs and also become invasive is not clear. The microenvironment of the pneumococci may become anoxygenic due to the pathological processes and enable invasive mechanisms to operate. Alternatively, the pneumococci become adapted to the in vivo oxygen environment. The nature of any adaptive mechanisms occurring during this period and the effect of oxygen and oxygen radicals on pneumococcal virulence mechanisms will be of interest.

The different patterns of cellular infiltration into the lungs may be explained by the different invasive properties of the pneumococci. D39HY1 remains in the bronchioles longer than the wild type, and inflammatory cells congregate around these areas of the lung, whereas wild-type pneumococci spread throughout the lung tissue and, therefore, the distribution of neutrophils is more diffuse. Experiments to determine the precise localization of pneumococci in the lungs have been started.

ACKNOWLEDGMENTS

Hasan Yesilkaya is sponsored by Kafkas University, Kars, Turkey.

We thank John Beckett, Leicester University Computer Centre, for advice on statistical methods.

REFERENCES

  • 1.Andrew P W, Mitchell T J, Myint S H. Prevention of respiratory infections. Curr Opin Infect Dis. 1993;6:146–149. [Google Scholar]
  • 2.Andrew P W, Boulnois G J, Mitchell T J, Lee C J, Paton J C, Poolman J T, Peeters C C A M. Pneumococcal vaccines. Zentbl Bakteriol. 1994;24:453–466. [Google Scholar]
  • 3.Beaman B L, Black C M, Doughty F, Beaman L. Role of superoxide dismutase and catalase as determinants of pathogenicity of Nocardia asteroides: importance in resistance to microbicidal activities of human polymorphonuclear neutrophils. Infect Immun. 1985;47:135–141. doi: 10.1128/iai.47.1.135-141.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Beauchamp C O, Fridovich I. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal Biochem. 1971;44:276–287. doi: 10.1016/0003-2697(71)90370-8. [DOI] [PubMed] [Google Scholar]
  • 5.Beyer W F, Fridovich I. In vivo competition between iron and manganese for occupancy of the active site region of the manganese-superoxide dismutase of Escherichia coli. J Biol Chem. 1991;266:303–308. [PubMed] [Google Scholar]
  • 6.Bradford M M. A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 1976;72:248–254. doi: 10.1016/0003-2697(76)90527-3. [DOI] [PubMed] [Google Scholar]
  • 7.Bridges S M, Salin M L. Distribution of iron containing superoxide dismutase in vascular plants. Plant Physiol. 1981;68:275–278. doi: 10.1104/pp.68.2.275. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Camhi S L, Lee P, Choi A M K. The oxidative stress response. New Horiz. 1995;3:170–182. [PubMed] [Google Scholar]
  • 9.Carlioz A, Touati D. Isolation of superoxide dismutase mutants in Escherichia coli: is superoxide dismutase necessary for aerobic life? EMBO J. 1986;5:623–630. doi: 10.1002/j.1460-2075.1986.tb04256.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Clare D A, Blum J, Fridovich I. A hybrid superoxide dismutase containing both functional iron and manganese. J Biol Chem. 1984;259:5932–5936. [PubMed] [Google Scholar]
  • 11.Crapo J D, McCord J M, Fridovich I. Preparation and assay of superoxide dismutases. Methods Enzymol. 1978;53:382–393. doi: 10.1016/s0076-6879(78)53044-9. [DOI] [PubMed] [Google Scholar]
  • 12.Davis B J. Disc electrophoresis. II. Method and applications to human serum proteins. Ann N Y Acad Sci. 1964;121:404–427. doi: 10.1111/j.1749-6632.1964.tb14213.x. [DOI] [PubMed] [Google Scholar]
  • 13.D'Mello R A, Langford P R, Kroll J S. Role of bacterial Mn-cofactored superoxide dismutase in oxidative stress responses, nasopharyngeal colonization, and sustained bacteremia caused by Haemophilus influenzae type b. Infect Immun. 1997;65:2700–2706. doi: 10.1128/iai.65.7.2700-2706.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Dunny G M, Lee L N, LeBlanc D J. Improved electroporation and cloning vector system for gram-positive bacteria. Appl Environ Microbiol. 1991;57:1194–1201. doi: 10.1128/aem.57.4.1194-1201.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Fee J A. Is superoxide toxic? In: Bannister W H, Bannister J V, editors. Biological and clinical aspects of superoxide and superoxide dismutase. New York, N.Y: Elsevier Science Publishing, Inc.; 1980. pp. 41–48. [Google Scholar]
  • 16.Franzon V L, Arondel J, Sansonetti P J. Contribution of superoxide dismutase and catalase activities to Shigella flexneri pathogenesis. Infect Immun. 1990;58:529–535. doi: 10.1128/iai.58.2.529-535.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Graeff-Wohlleben H, Killat S, Banemann A, Guiso N, Gross R. Cloning and characterization of an Mn-containing superoxide dismutase (SodA) of Bordetella pertussis. J Bacteriol. 1997;179:2194–2201. doi: 10.1128/jb.179.7.2194-2201.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Halliwell B, Gutteridge J M C. Role of free radicals and catalytic metal ions in human disease: an overview. In: Packer L, Glazer A N, editors. Oxygen radicals in biological systems. New York, N.Y: Academic Press; 1990. pp. 1–85. [DOI] [PubMed] [Google Scholar]
  • 18a.Hardie J M. Genus Streptococcus. In: Sneath P H A, Mair N S, Sharpe M E, Holt J G, editors. Bergey's manual of systematic bacteriology. Baltimore, Md: Williams & Wilkins; 1986. pp. 1043–1071. [Google Scholar]
  • 19.Hassan H M. Microbial superoxide dismutases. Adv Genet. 1989;26:65–97. doi: 10.1016/s0065-2660(08)60223-0. [DOI] [PubMed] [Google Scholar]
  • 20.Hassan H M, Fridovich I. Enzymatic defenses against the toxicity of oxygen and of streptonigrin in Escherichia coli. J Bacteriol. 1977;129:1574–1583. doi: 10.1128/jb.129.3.1574-1583.1977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hassan H M, Fridovich I. Physiological function of superoxide dismutase in glucose-limited chemostat cultures of Escherichia coli. J Bacteriol. 1977;130:805–811. doi: 10.1128/jb.130.2.805-811.1977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Hassett D J, Schweizer H P, Ohman D E. Pseudomonas aeruginosa sodA and sodB mutants defective in manganese- and iron-cofactored superoxide dismutase activity demonstrate the importance of the iron-cofactored form in aerobic metabolism. J Bacteriol. 1995;177:6330–6337. doi: 10.1128/jb.177.22.6330-6337.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Haverstein L S, Coomaraswamy G, Morrison D A. An unmodified heptadecapeptide pheromone induces competence for genetic transformation in Streptococcus pneumoniae. Proc Natl Acad Sci USA. 1995;92:11140–11144. doi: 10.1073/pnas.92.24.11140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Jackett P S, Andrew P W, Aber V R, Lowrie D B. Hydrogen peroxide and superoxide release from alveolar macrophages from normal and BCG-vaccinated guinea pigs after intravenous challenge with Mycobacterium tuberculosis. Br J Exp Pathol. 1981;62:419–428. [PMC free article] [PubMed] [Google Scholar]
  • 25.Janssen Y M W, Van Houten B, Borm P J A, et al. Biology of disease. Cell and tissue response to oxidative damage. Lab Investig. 1993;69:261–274. [PubMed] [Google Scholar]
  • 26.Klugman K P. Worldwide increase in pneumococcal antibiotic resistance. Lancet. 1989;ii:444. doi: 10.1016/s0140-6736(89)90617-x. [DOI] [PubMed] [Google Scholar]
  • 27.Lacks S. Integration efficiency and genetic recombination in pneumococcal transformation. Genetics. 1966;53:207–235. doi: 10.1093/genetics/53.1.207. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Langford P R, Loynds B M, Kroll J S. Copper-zinc superoxide dismutase in Haemophilus species. J Gen Microbiol. 1992;138:517–522. doi: 10.1099/00221287-138-3-517. [DOI] [PubMed] [Google Scholar]
  • 29.Langford P R, Loynds B M, Kroll J S. Cloning and molecular characterization of Cu,Zn superoxide dismutase from Actinobacillus pleuropneumoniae. Infect Immun. 1996;64:5035–5041. doi: 10.1128/iai.64.12.5035-5041.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.LeBlanc D J, Lee L N, Inamine J M. Cloning and nucleotide base sequence-analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob Agents Chemother. 1991;35:1804–1810. doi: 10.1128/aac.35.9.1804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Liochev S V, Fridovich I. Paraquat diaphorases in Escherichia coli. Free Radic Biol Med. 1994;16:555–559. doi: 10.1016/0891-5849(94)90055-8. [DOI] [PubMed] [Google Scholar]
  • 32.Martin M E, Strachan R C, Aranha H, Evans S L, Salin M L, Welch B, Arceneaux J E L, Byers B R. Oxygen toxicity in Streptococcus mutans: manganese iron and superoxide dismutase. J Bacteriol. 1984;159:745–749. doi: 10.1128/jb.159.2.745-749.1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.McCord J M, Fridovich I. Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein) J Biol Chem. 1969;244:6049–6055. [PubMed] [Google Scholar]
  • 34.Nakayama K. Nucleotide sequence of Streptococcus mutans superoxide dismutase gene and isolation of insertion mutants. J Bacteriol. 1992;174:4928–4934. doi: 10.1128/jb.174.15.4928-4934.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Parker M W, Blake C C F. Iron- and manganese-containing superoxide dismutases can be distinguished by analysis of their primary structures. FEBS Lett. 1988;229:377–382. doi: 10.1016/0014-5793(88)81160-8. [DOI] [PubMed] [Google Scholar]
  • 36.Paton J C, Andrew P W, Boulnois G J, Mitchell T J. Molecular analysis of the pathogenicity of Streptococcus pneumoniae—the role of pneumococcal proteins. Annu Rev Microbiol. 1993;47:89–115. doi: 10.1146/annurev.mi.47.100193.000513. [DOI] [PubMed] [Google Scholar]
  • 37.Pesci E C, Cottle D L, Pickett C L. Genetic, enzymatic, and pathogenic studies of the iron superoxide dismutase of Campylobacter jejuni. Infect Immun. 1994;62:2687–2694. doi: 10.1128/iai.62.7.2687-2694.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Poyart C, Berche P, Trieu-Cuot P. Characterization of superoxide dismutase genes from Gram-positive bacteria by polymerase chain reaction using degenerate primers. FEMS Microbiol Lett. 1995;131:41–45. doi: 10.1016/0378-1097(95)00232-t. [DOI] [PubMed] [Google Scholar]
  • 39.Ridgway E J, Allen K D. Penicillin resistance in pneumococci. J Antimicrob Chemother. 1991;27:251–252. doi: 10.1093/jac/27.2.251. [DOI] [PubMed] [Google Scholar]
  • 40.Roggenkamp A, Bittener T, Leitritz L, Sing A, Heesemann J. Contribution of the Mn-cofactored superoxide dismutase (SodA) to the virulence of Yersinia enterocolitica serotype O8. Infect Immun. 1997;65:4705–4710. doi: 10.1128/iai.65.11.4705-4710.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Saito H, Maiura K I. Preparation of transforming deoxyribonucleic acid by phenol treatment. Biochim Biophys Acta. 1963;72:619–629. [PubMed] [Google Scholar]
  • 42.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
  • 43.Sanders J W, Leenhouts K J, Haandrikman A J, Venema G, Kok J. Stress response in Lactococcus lactis: cloning, expression analysis, and mutation of the lactococcal superoxide dismutase gene. J Bacteriol. 1995;177:5254–5260. doi: 10.1128/jb.177.18.5254-5260.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Spellerberg B, Cundell D R, Sandros J, Pearce B J, Heikkila-Idanpaan I, Rosenow C, Masure H R. Pyruvate oxidase, as a determinant of virulence in Streptococcus pneumoniae. Mol Microbiol. 1996;19:803–813. doi: 10.1046/j.1365-2958.1996.425954.x. [DOI] [PubMed] [Google Scholar]
  • 45.Steinman H M. Bacteriocuprein superoxide dismutases in pseudomonads. J Bacteriol. 1985;162:1255–1260. doi: 10.1128/jb.162.3.1255-1260.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Wilks K E, Dunn K L, Farrant J L, Reddin K M, Gorringe A R, Langford P R, Kroll J S. Periplasmic superoxide dismutase in meningococcal pathogenicity. Infect Immun. 1998;66:213–217. doi: 10.1128/iai.66.1.213-217.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Uchida Y, Shigematu H, Yamafuji K. The mode of action of hydrogen peroxide on DNA. Enzymologia. 1965;6:369–376. [PubMed] [Google Scholar]
  • 48.Zabarovsky E R, Winberg G. High efficiency electroporation of ligated DNA into bacteria. Nucleic Acid Res. 1990;18:5912. doi: 10.1093/nar/18.19.5912. [DOI] [PMC free article] [PubMed] [Google Scholar]