Influence of the spxB Gene on Competence in Streptococcus pneumoniae

Abstract

In Streptococcus pneumoniae expression of pyruvate oxidase (SpxB) peaks during the early growth phase, coincident with the time of natural competence. This study investigated whether SpxB influences parameters of competence, such as spontaneous transformation frequency, expression of competence genes, and DNA release. Knockout of the spxB gene in strain D39 abolished spontaneous transformation (compared to a frequency of 6.3 × 10⁻⁶ in the parent strain [P < 0.01]). It also reduced expression levels of comC and recA as well as DNA release from bacterial cells significantly during the early growth phase, coincident with the time of spontaneous competence in the parent strain. In the spxB mutant, supplementation with competence-stimulating peptide 1 (CSP-1) restored transformation (rate, 1.8 × 10⁻²). This speaks against the role of SpxB as a necessary source of energy for competence. Neither supplementation with CSP-1 nor supplementation with the SpxB products H₂O₂ and acetate altered DNA release. Supplementation of the parent strain with catalase did not reduce DNA release significantly. In conclusion, the pneumococcal spxB gene influences competence; however, the mechanism remains elusive.

The human nasopharynx is the natural habitat of Streptococcus pneumoniae. Colonization is the prerequisite for invasion, transmission, and genetic evolution of pneumococci. An increasing number of adherence factors that support colonization are being discovered (13). In addition, production of hydrogen peroxide (H₂O₂) is thought to inhibit competitive nasopharyngeal flora (23). In the pneumococcus, H₂O₂ is produced under rich and aerobic conditions by the enzyme pyruvate oxidase (SpxB), encoded by the spxB gene. Besides the production of H₂O₂, SpxB decarboxylates pyruvate to acetyl phosphate plus CO₂ (24). Spellerberg et al. (29) showed that an spxB-deficient mutant exhibits reduced virulence for nasopharyngeal colonization, pneumonia, and sepsis. Expression of SpxB peaks during the early growth phase (18). This coincides with the time of pneumococcal competence.

S. pneumoniae shares with at least 40 bacterial species the property of natural transformation (19). The importance of transformation for genetic evolution has been illustrated by the emergence of penicillin-resistant pneumococcal isolates and pneumococcal strains undergoing capsule switch (8, 9, 22). In S. pneumoniae, competence is induced by the competence-stimulating peptide (CSP) (7). Induction of competence-specific (com) genes leads to DNA uptake and processing. Competence also triggers cell lysis and DNA release from a fraction of bacterial cells.

This study investigated whether the spxB gene or the products of SpxB have any influence on competence in S. pneumoniae. It is shown that in strain D39, deletion of spxB abolished spontaneous transformation, reduced expression levels of an early competence gene (comC) and a late competence gene (recA), and reduced competence-associated DNA release. There was, however, no evidence for a role of the products of SpxB, i.e., H₂O₂ and acetyl phosphate. The mechanisms connecting spxB gene with competence remain to be unraveled.

MATERIALS AND METHODS

Bacterial strains and culture.

Streptococcus pneumoniae strain D39 (serotype 2) (2) was kindly provided by Jeffrey Weiser (University of Pennsylvania, Philadelphia, PA) and strain R6 (a spontaneous nonencapsulated derivative of D39) by Philippe Moreillon (Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerland).

Bacteria were grown on Columbia sheep blood agar (CSBA) plates at 37°C in a 5% CO₂-enriched atmosphere. Liquid culture was performed in brain heart infusion (BHI) broth, pH 7.4 to 7.5 (Becton Dickinson and Company, le Pont de Claix, France), containing 5% fetal calf serum (FCS) (Biochrom KG, Berlin, Germany) in a water bath at 37°C without shaking. Transformation experiments were performed either in BHI broth supplemented with 5% FCS or in TSB competence medium (pH 8) (27). Where stated, 100 ng/ml CSP-1 (NeoMPS S.A, Strasbourg, France), 0.1% sodium acetate (Merck, Darmstadt, Germany), or 5,000 U/ml catalase (Sigma-Aldrich, Buchs, Switzerland) was added to the BHI broth with 5% FCS. Bacteria were stored at −80°C using Protect bacterial preservers (TSC, Heywood, United Kingdom). Optical density at 600 nm (OD₆₀₀) was measured using a Perkin-Elmer Lambda-2 spectrometer (Perkin-Elmer [Schweiz] AG, Schwerzenbach, Switzerland).

Construction of the D39 spxB mutant.

Transformation of Escherichia coli and S. pneumoniae was performed as described previously (14, 20). A 1.3-kb fragment (spxB_S2_F, 5′-TAAATTCGGCGGCTCAATC-3′; spxB_S1_B, 5′-CAGCGTTTGTGAAGTCTACACC-3′) of spxB was amplified and cloned into pGEM-T Easy Vector (Promega, Wallisellen, Switzerland). An erythromycin cassette (ermB) (5) was inserted at the HindIII restriction site of the amplified spxB fragment. Strain D39 Sm^r was transformed with the whole plasmid, and recombinants were selected on CSBA plates containing erythromycin (2 μg/ml). Knockout of spxB was confirmed by PCR and phenotypically by a 4-fold-decreased H₂O₂ release per cell and a 10-fold-decreased H₂O₂ concentration in the supernatant at an OD₆₀₀ of 0.75 compared to the wild type (data not shown).

Assay for transformation frequency.

Transformations were performed as described below with the following adaptations. Rifampin-susceptible strain D39 Sm^r or its spxB-deficient mutant were grown in BHI broth with 5% FCS to an OD₆₀₀ of 0.05, 0.15, 0.25, 0.35, or 0.45 and transformed with a total of 1 μg DNA consisting of the rpoB rifampin resistance (Rif^r) gene (20). Transformation was performed with or without addition of CSP-1; 100 μl of culture was spread on CSBA plates containing 0.5 μg/ml rifampin. Transformants were counted after 48 h. Total cell counts were obtained by plating serial dilutions of culture onto CSBA plates.

RNA isolation for reverse transcription-PCR (RT-PCR).

Bacteria were prepared as previously described (15), transferred to a 1.5-ml tube containing 0.05 g of 100-μm acid-washed glass beads (Sigma-Aldrich, Buchs, Switzerland), and vibrated for 10 min at half-maximum speed using a Mickle vibratory tissue disintegrator (Mickle Laboratory Engineering, Gomshall, United Kingdom). The mixture was then centrifuged and RNA extracted from the supernatant using the Qiagen RNeasy minikit (Qiagen AG, Hombrechtikon, Switzerland) according to the manufacturer's instructions. The RNA recovered was treated with DNase I (Stratagene Europe, Amsterdam, Netherlands) according to the manufacturer's instructions to remove any contaminating DNA. RNA concentration and purity were determined by measuring absorbance at both 260 nm and 280 nm (Lambda-2 spectrometer; Perkin-Elmer [Schweiz], Schwerzenbach, Switzerland).

Quantitative gene expression using a cRNA standard curve.

Quantitation of absolute mRNA copy numbers by real-time RT-PCR was performed by using a standard curve generated based on in vitro-transcribed RNA (cRNA) as previously described (12, 16). For the in vitro transcription of the spxB, comC, and recA genes, the following primers were used: spxB_F1_T7 (5′-TAATACGACTCACTATAGGGAGAGTGGAATAGTAAAAATTTGGAGAACG-3′) and spxB_Bac1 (5′-CGATCTTTTAAAGTTCTGCTCTATG-3′), comC_Start1_T7 (5′-TAATACGACTCACTATAGGGAGAAATCTTTCTGTCAGTTTTGGTCG-3′) and comC_End1 (5′-GTCCCAAATCCAAATAAATCCAT-3′), and recA_Start1_T7 (5′-TAATACGACTCACTATAGGGAGAGTACGTCACATTGCGGTTATGC-3′) and recA_End1 (5′-GAATCAAAAATCGAAAAAGTAGCG-3′) (the T7 promoter is underlined).

Three microliters of RNA extracted from 10 ml of bacterial culture at an OD₆₀₀ of 0.05, 0.15, 0.25, 0.35, 0.45, 0.55, or 0.65 was reversed transcribed to cDNA using Superscript II (Amersham, Buckinghamshire, United Kingdom) and random hexamer primers (Promega, Wallisellen, Switzerland) according to the supplier's protocol. For the cultures of each OD₆₀₀ from which we extracted RNA, serial dilutions were spread onto CSBA plates to determine the cell count (CFU).

Quantification of gene expression was achieved by real-time RT-PCR using TaqMan primers and probes created by the Assay-by-Design service of Applied Biosystems (Rotkreuz, Switzerland) based on the most conserved regions of the spxB, comC, and recA genes in S. pneumoniae strains TIGR4 (AE005672), R6 (AE007317, NC003098), D39 (NC008533, AY254852), and AB15 (AY254854) and in Streptococcus pyogenes strain MGAS10750 (CP000262) (forward primer spxB-tgt3F [5′-ACAGGTTCTGCTTACCGTGTTG-3′], reverse primer spxB-tgt3R [5′-AGGAAAAGAACTGTGTCTGCTTCAA-3′], and probe spxB-tgt3M2 [6-carboxyfluorescein {6-FAM}-TCGTTGGCTGGTTTCCAA-MGB]; forward primer comC-anyF [5′-TGGAACAGTTTGTAGCTTTGAAGGA-3′], reverse primer comC-anyR [5′-TCACGGAAGAATTTTGACAACCTCAT-3′], and probe comC-anyM2 [6-FAM-TCCCCACCTTTAATCTT-minor groove binder (MGB)]; and forward primer recA-tgt1F [5′-GGTAAAGGATCAATCATGCGTTTGG-3′], reverse primer recA-tgt1R [5′-AACCTGAGCTCATCACTTGCA-3′], and probe recA-tgt1M2 [6-FAM-TTTTGCTCCGCACGTTCA-MGB]). cDNA was diluted fourfold in the assay, and an RT-negative control was performed for every sample. Real-time RT-PCR was performed in 96-well plates using the ABI Prism 7000 sequence detection system (Applied Biosystems). The experiment was performed on three different days, and real-time RT-PCR was carried out in triplicates.

Finally, the threshold cycle values and the cell counts of the extracted RNA cultures were used to calculate the copy numbers of the samples per 10² CFU.

Quantification of DNA in the supernatants of D39 Sm^r and its spxB mutant using real-time PCR.

The relative quantification of DNA in the supernatants of D39 Sm^r and its spxB mutant at different ODs was determined by real-time PCR using TaqMan primers and probes created by the Assay-by-Design service of Applied Biosystems (Rotkreuz, Switzerland) based on the 16S rRNA gene in S. pneumoniae (forward primer 16S, 5′-GACGATACATAGCCGACCTGAGA-3′; reverse primer 16S, 5′-GTAGGAGTCTGGGCCGTGTCT-3′; and 16S probe, 6-FAM-CCAGTGTGGCCGATC-MGB). Supernatants of strain D39 Sm^r and its spxB mutant were harvest using Micropure-EZ enzyme removers (Millipore AG, Volketswil, Switzerland), and real-time PCR was performed as described above with undiluted supernatant. The threshold cycle difference between strain D39 Sm^r and its spxB-deficient mutant at each OD was calculated. A difference in threshold cycle number of 1 equates to a twofold difference in initial template concentration.

Transformation assay for DNA release.

DNA release was measured by transformation of competent cells of an Sm^s recipient strain R6 hexA mutant with cell-free filtrates from the D39 Sm^r strain or its spxB mutant as described earlier (21).

D39 Sm^r was a spontaneous mutant selected from CSBA plates containing 300 μg/ml streptomycin after inoculation with 200 μl of a stationary-phase culture (OD₆₀₀, 0.8) for 24 h at 37°C in a 5% CO₂-enriched atmosphere.

To delete hexA in R6, a 2.6-kb fragment of the hexA gene of strain 108.21 (nonencapsulated S. pneumoniae) was amplified (with primers hexA_f11 [5′-AGAGACAGAAAATGGCGATAGAAA-3′] and hexA_b2641 [5′-ATAGACAAAAGGGAGCGACAATG-3′]) and cloned into pGEM-T Easy vector (Promega, Wallisellen, Switzerland). A 1,000-bp fragment of hexA was cut out with HindIII and was replaced with an erythromycin cassette (ermB of pJDC9). S. pneumoniae strain R6 was transformed with the whole plasmid, and recombinants were selected on CSBA plates containing erythromycin (2 μg/ml). Positive recombinants were confirmed by Southern blotting, PCR and an increased mutation rate to rifampin resistance (data not shown).

An overnight culture of strain D39 Sm^r or its spxB mutant was prepared with 3 to 10 colonies in 5 ml BHI broth containing 5% FCS. One hundred microliters of culture was subcultured in 5 ml BHI broth with 5% FCS and grown to mid-log phase (OD₆₀₀, 0.5). Two milliliters of culture was then pelleted at 2,500 × g and 4°C for 10 min and washed twice with 5 ml Hanks medium to remove extracellular DNA. The pellet was diluted 100-fold in BHI broth with 5% FCS, and 20 ml was incubated at 37°C.

Samples of 0.2 ml were taken at OD₆₀₀s of 0.05, 0.15, 0.25, 0.35, 0.45, 0.55, 0.65, and 0.75. Cells were removed by filtration and centrifugation for 60 s using Micropure-EZ enzyme removers (Millipore AG, Volketswil, Switzerland). To maximize DNA recovery, 50 μl Tris-EDTA buffer (pH 8.0) was added to the Micropure-EZ reservoir and spun for 30 s. Filtrates were stored for 1 to 2 days at −20°C until use.

Competent cells of the R6 hexA mutant were prepared by inoculating BHI broth containing 5% FCS with 3 to 10 colonies. Overnight cultures were diluted 1:100 in fresh broth and grown to an OD₆₀₀ of 0.15. The culture was diluted 1:20 in BHI broth with 5% FCS or TSB competence 8.0 medium prewarmed to 30°C, and aliquots of 750 μl were incubated for 15 min. CSP-1 (Neosystems, Strasbourg, France) was added to a final concentration of 100 ng/ml and incubated for 15 min at 30°C, and then 250 μl of cell-free filtrates was added and incubated for 40 min at 30°C and then for 90 min at 37°C. Aliquots of 200 μl were spread in duplicates on CSBA plates containing 200 μg/ml streptomycin, and transformants were counted after 24 h. For each experiment, three independent measurements were performed on different days.

Supplementation of culture medium of mutant D39 Sm^r ΔspxB with H₂O₂.

In order to complement the spxB-deficient mutant with physiological concentrations of H₂O₂, the accumulation of H₂O₂ in the supernatant of strain D39 Sm^r was first determined. An overnight culture was diluted 1:50 in BHI broth with 5% FCS and was grown to mid-log phase (OD₆₀₀, 0.5). Cells from 2 ml culture were pelleted at 2,500 × g and 4°C for 10 min and washed twice with 5 ml Hanks medium. The pellet was diluted 100-fold in BHI broth with 5% FCS and incubated at 37°C. Samples of 0.2 ml were withdrawn at different OD₆₀₀s. The H₂O₂ concentration in the supernatant was measured by using the Amplex Red hydrogen peroxide/peroxidase assay kit (Molecular Probes, Eugene, OR). Fifty microliters of bacterial culture was applied to a 96-well plate (Nunclon Nalge Nunc, Roskilde, Denmark), and 50 μl Hanks medium containing 0.2 U/ml horseradish peroxidase and 100 μM Amplex Red reagent were added. The absorbance was read at a wavelength of 563 nm (SpectraMax GeminiXS; Molecular Devices, Sunnyvale, CA). Concentrations were determined based on a standard. The concentrations of H₂O₂ in the supernatant of D39 Sm^r at OD₆₀₀s of 0.005, 0.05, 0.15, 0.25, 0.45, and 0.75 were 1.5, 150, 270, 320, 525, and 480 μM, respectively.

To mimic H₂O₂ concentrations in the growth medium, the spxB-deficient mutant was grown in 20 ml BHI broth with 5% FCS. At the start, 0.05 μmol H₂O₂ (Merck, Darmstadt, Germany) was added to reach an initial concentration of 2.5 μM. Thereafter, H₂O₂ was added at intervals of 10 min for a total of 300 min. The amount of H₂O₂ added was incrementally increased by 0.05 μmol for each subsequent addition.

Statistical analyses.

Statistical analyses were done in StatView version 5.0 (SAS Institute Inc., Cary, NC). Proportions were compared with the chi square test or Fisher's exact test, and mean differences were assessed by Student's t test. A cutoff P value of ≤0.05 (two tailed) was used for all statistical analyses.

RESULTS

Kinetics of spxB gene expression.

Figure 1 confirms that transcription levels of the spxB gene in strain D39 were highest at an OD₆₀₀ of 0.150 (P < 0.05 compared to values at all other ODs).

FIG. 1.

spxB gene expression in strain D39 during the lag and log phases. D39 was cultured in BHI broth with 5% FCS. Left y axis, the spxB transcription levels were determined by real-time RT-PCR and expressed as the copy number per 100 CFU. Mean values of triplicates from three independent experiments (±SE) are presented. Right y axis, CFU per ml at each OD at which the spxB expression was measured. Presented are mean values of triplicates from three independent experiments.

Influence of spxB gene on transformation frequency.

The frequencies of spontaneous transformation in the parent strain D39 Sm^r and its spxB-deficient mutant were compared at different ODs (Fig. 2 and 3A and B). No transformants were obtained for the spxB mutant at OD₆₀₀s of 0.05, 0.15, 0.25, and 0.45. A few transformants were obtained at an OD₆₀₀ 0.35, for a mean transformation frequency of 4.9 × 10⁻⁶. Spontaneous transformation in the parent strain was highest at an OD₆₀₀ of 0.05 and decreased thereafter (OD₆₀₀ of 0.05, 5.9 × 10⁻⁴ [standard error {SE}, 2.3 × 10⁻⁴; OD₆₀₀ of 0.45, 3.5 × 10⁻⁵ [SE, 2.6 × 10⁻⁵ at) (Fig. 3A and B).

FIG. 2.

Transformation frequency of strain D39 Sm^r and its spxB mutant at an OD₆₀₀ of 0.15. Strain D39 Sm^r and its spxB mutant were grown to an OD₆₀₀ of 0.15 in BHI broth with 5% FCS. The transformation frequency was measured in TSB competence 8.0 medium with and without the addition of CSP-1, and the number of Rif^r transformants per CFU was calculated. Mean values of triplicates from three independent experiments (±SE) are presented.

FIG. 3.

comC and recA gene expression and spontaneous transformation frequencies of strain D39 Sm^r and its spxB mutant during the time of competence. Strain D39 Sm^r and its spxB mutant were grown to OD₆₀₀s of 0.05 to 0.45 in BHI broth with 5% FCS. (A and B) Left Y axes, the comC (A) and recA (B) gene transcription levels (gray and white bars) were determined by real-time RT-PCR and expressed as the copy number per 100 CFU in strain D39 Sm^r compared to its spxB-deficient mutant at each OD. Mean values of triplicates from three independent experiments (±SE) are presented. *, P < 0.05; the P values were calculated by comparing the comC or recA gene expression in D39 Sm^r with the comC or recA gene expression in D39 Sm^r ΔspxB at each OD. Right y axes, the spontaneous transformation frequencies were measured in TSB competence 8.0 medium, and the number of Rif^r transformants per CFU was calculated. Mean values of duplicates from two independent experiments (±SE) are presented. Gray squares and solid lines, D39 Sm^r; white squares and broken lines, D39 Sm^r spxB knockout mutant. (C) Growth curves in BHI broth supplemented with 5% FCS of strain D39 Sm^r and its spxB knockout mutant. (D) CFU per ml of each OD presented in panel C for strain D39 Sm^r and its spxB knockout mutant.

We investigated the effect of CSP addition on the transformation frequency in the spxB mutant compared to its parent strain D39 Sm^r at an OD₆₀₀ of 0.15 (Fig. 2). Addition of CSP-1 increased the transformation frequency in the spxB mutant significantly to 1.8 × 10⁻² (SE, 7 × 10⁻²) (comparison between frequencies with or without CSP-1, P = 0.025). The comparison with the parent strain (mean transformation frequency, 2.3 × 10⁻²; SE, 6 × 10⁻²) no longer showed a significant difference (P = 0.5). In order to better understand at which stage spxB interfered with spontaneous transformation, expression of an early and a late competence gene was analyzed.

Expression of comC and recA genes in D39 Sm^r and the spxB knockout mutant.

In strain D39 Sm^r, transcription levels of comC (an early competence gene) and recA (a late competence gene) were highest between OD₆₀₀s of 0.05 and 0.15, which is consistent with spontaneous competence. Deletion of the spxB gene significantly reduced comC and recA expression at an OD₆₀₀ of 0.05 (P = 0.004 and P = 0.040, respectively) and at an OD₆₀₀ of 0.15 (P = 0.028 and P = 0.033, respectively) (Fig. 3A and B). Growth of the spxB mutant in BHI broth supplemented with 5% FCS was slightly delayed compared to that of its parent strain D39 Sm^r and required 30 to 45 min longer to reach an OD₆₀₀ of 0.75 (Fig. 3C).

Influence of the spxB gene on DNA release.

DNA release has been shown to be associated with competence (21). Therefore, we investigated whether deletion of spxB also influenced DNA release. Amounts of DNA in the supernatants of D39 Sm^r strain and its spxB mutant were determined by real-time PCR at OD₆₀₀s of 0.05, 0.15, 0.25, and 0.35 with and without the addition of CSP-1. Deletion of the spxB gene lowered the DNA content in the supernatant up to 3.5-fold at an OD₆₀₀ of 0.15. Addition of CSP-1 increased the DNA quantity in the supernatants of both the D39 Sm^r strain and the spxB mutant. However, addition of CSP-1 could not fully restore DNA release in the supernatant of the spxB mutant (Fig. 4A).

FIG. 4.

DNA release of strain D39 Sm^r and its spxB mutant during the lag and log phases. Bacteria were grown to different ODs in BHI broth with 5% FCS. DNA release into the supernatant was measured with and without the addition of CSP-1 by real-time PCR (A) or as the number of transformants as described by Moscoso and Claverys (21) (B). (A) The DNA quantity in the supernatant is expressed as the fold difference between DNA in the supernatant and that for D39 Sm^r without the addition of CSP-1 (±SE). (B) Values for DNA release represent streptomycin-resistant transformants per ml. Mean values of duplicates from three independent experiments (±SE) are presented. * and **, P < 0.05; the P values were calculated by comparing the DNA release of D39 Sm^r with the DNA release of D39 Sm^r ΔspxB at each OD without (*) or with (**) the addition of CSP-1.

DNA release was also measured using the method described by Moscoso and Claverys (21), using the strain R6 hexA mutant with the supernatants of D39 Sm^r strain and its spxB mutant. The results were in line with those obtained by measuring DNA in the supernatant by real-time PCR (Fig. 4B). In strain D39 Sm^r, release of DNA peaked at an OD₆₀₀ of 0.15. The spxB mutant showed significantly reduced DNA release between OD₆₀₀s of 0.05 and 0.35 (P ≤ 0.025). Addition of CSP-1 increased DNA release significantly between OD₆₀₀s of 0.05 and 0.55 in strain D39 Sm^r (P ≤ 0.003) and in the spxB mutant (P ≤ 0.0392). However, DNA levels in the mutants never reached those in the parent strain (P ≤ 0.0023).

Influence of supplementation with acetate and H₂O₂ on DNA release in the spxB mutant.

In order to investigate whether the end products of SpxB, i.e., H₂O₂ and/or acetyl phosphate, played a role, DNA release was measured in the spxB mutant with H₂O₂ supplementation and/or the addition of acetate. However, the compounds alone or in combination did not influence DNA release (Fig. 5).

FIG. 5.

DNA release of the spxB-deficient mutant grown in culture medium supplemented with acetate and/or H₂O₂. Bacteria were grown to different ODs in BHI broth plus 5% FCS with and without acetate and/or H₂O₂. No CSP-1 was added to the cultures. Values for DNA release represent streptomycin-resistant transformants per ml. Mean values of triplicates from three independent experiments (±SE) are presented.

Influence of supplementation with catalase on DNA release in strain D39 Sm^r.

In order to test whether the reduction of DNA release in the spxB mutant was due to the lack of a toxic effect of H₂O₂, DNA release in parental strain D39 Sm^r was measured in the presence of catalase. Catalase reduced the H₂O₂ concentration in the supernatant of strain D39 Sm^r to an undetectable level (data not shown). DNA release was not significantly reduced upon addition of catalase compared to that in cultures without catalase at ODs ranging from 0.15 to 0.75. Catalase tended to reduce DNA release slightly in the early growth phase (reduction factor of 1.8 at OD₆₀₀s of 0.025 and 0.05 [P = 0.016 if values for both time points were pooled] [data not shown]). However, overall these results were not sufficiently convincing for indicating a direct toxic effect of H₂O₂ in the early growth phase.

DISCUSSION

This study investigated whether pyruvate oxidase (SpxB) plays a role in competence of S. pneumoniae. Such a role was suggested by the coincidence between competence and peak expression of spxB gene during the early growth phase as shown in this and earlier reports (18). Deletion of the spxB gene in strain D39 influenced three competence-associated parameters: transformation frequency, expression of the comC and recA genes, and DNA release. Therefore, a connection between spxB and the competence machinery seems to exist, but the mechanisms are still unknown.

Supplementation with CSP restored spontaneous transformability in the spxB mutant. Therefore, deletion of spxB did not affect the response to CSP, DNA uptake, or recombination. Others have shown that expression of spxB is not controlled by CSP (25, 26).

It may be that the role of SpxB in competence is energy supply and is similar to the role of NADH oxidase. Competence in S. pneumoniae depends on the availability of oxygen (1) and high levels of ATP (6). Energy supply is required for pneumococci to enter competence and for the uptake of transforming DNA (6). Most of the ATP supply in S. pneumoniae is derived from the glycolytic breakdown of glucose, which is more efficient in the presence of NADH oxidase. NADH accumulated during glycolysis is reoxidized by the NADH oxidase using O₂ (1, 4, 10, 11). SpxB also utilizes O₂ and produces the energy-rich acetyl phosphate, a potential source of ATP (24). In addition, knockout of the spxB gene reduces the ATP level, as shown by Pericone et al. (24). However, addition of CSP restored transformability, which speaks against the energy hypothesis for the effect of SpxB on competence.

SpxB generates H₂O₂ and acetyl phosphate. Therefore, we investigated whether either of the two compounds may mediate the observed effects of spxB deletion on DNA release. Supplementation of the spxB mutant with H₂O₂ and/or acetate (restores the acetyl phosphate level [29]) also revealed no effect on DNA release. Therefore, we found no indication of an intracellular effect for either H₂O₂ or acetate. We cannot exclude the possibility that more sophisticated ways of supplementation would have shown an effect. The hypothesis seemed attractive, because low intracellular concentrations of H₂O₂ may not be toxic but may induce oxidative stress and trigger competence (7, 28). Also, Kim et al. demonstrated that acetyl phosphate can act as an intracellular messenger (17), and Spellerberg et al. (29) showed that reduced adherence in an spxB mutant can be restored by addition of 0.1% sodium acetate.

Lastly, we investigated whether a direct toxic effect of H₂O₂ may have contributed to DNA release (3). Supplementation of the parent strain with catalase did not reduce DNA release significantly. The limited effect of catalase supplementation on DNA release was not due to an insufficient amount of catalase addition, since the H₂O₂ concentration in the supernatant was at an undetectable level during the whole experiment (data not shown). However, catalase supplementation may not be appropriate to study the effect of endogenously produced H₂O₂, since catalase cannot penetrate cell membranes.

In conclusion, this study provides evidence for a role of the strain D39 spxB gene in competence; however, the mechanism remains elusive.