Pneumococcal gene complex involved in resistance to extracellular oxidative stress

Abstract

Streptococcus pneumoniae is a gram-positive bacterium which is a member of the normal human nasopharyngeal flora but can also cause serious disease such as pneumonia, bacteremia, and meningitis. Throughout its life cycle, S. pneumoniae is exposed to significant oxidative stress derived from endogenously produced hydrogen peroxide (H(2)O(2)) and from the host through the oxidative burst. How S. pneumoniae, an aerotolerant anaerobic bacterium that lacks catalase, protects itself against hydrogen peroxide stress is still unclear. Bioinformatic analysis of its genome identified a hypothetical open reading frame belonging to the thiol-specific antioxidant (TlpA/TSA) family, located in an operon consisting of three open reading frames. For all four strains tested, deletion of the gene resulted in an approximately 10-fold reduction in survival when strains were exposed to external peroxide stress. However, no role for this gene in survival of internal superoxide stress was observed. Mutagenesis and complementation analysis demonstrated that all three genes are necessary and sufficient for protection against oxidative stress. Interestingly, in a competitive index mouse pneumonia model, deletion of the operon had no impact shortly after infection but was detrimental during the later stages of disease. Thus, we have identified a gene complex involved in the protection of S. pneumoniae against external oxidative stress, which plays an important role during invasive disease.

Fig 1

TlpA has no role in protection against intracellular oxidative stress. After growth to mid-log phase, wild-type D39 nisRK (■) and D39 nisRK ΔtlpA (□) were incubated with or without 30 mM paraquat for 2 h, and the percent survival was determined. Error bars indicate standard deviations from the mean. Data are the average of three independent assays.

Fig 2

Deletion of tlpA impaired hydrogen peroxide resistance of S. pneumoniae. Survival of exposure to 20 mM H₂O₂ for 30 min of parent (filled bars) and mutant (striped bars) strains in the background of strains G54, 670-6B, D39 nisRK (D39 RK), R6 nisRK (R6 RK), and TIGR4 (T4). The percent survival was calculated by dividing the number of CFU of cultures after exposure to H₂O₂ by the number of CFU of the control cultures without H₂O₂. Data are averages of at least three independent assays. Error bars indicate standard deviations from the mean. *, P < 0.05.

Fig 3

The operon spd0571-0573 is involved in resistance of S. pneumoniae against extracellular oxidative stress. (A) Survival of exposure to 20 mM H₂O₂ of D39 nisRK (D39 RK), D39 ΔtlpA nisRK (D39 RK ΔtlpA), D39 ΔtlpA nisRK pNZ8048::tlpA (D39 RK ΔtlpA P tlpA), D39 Δspd0571-0573 nisRK (D39 RK Δoperon), D39 Δspd0571-0573 nisRK pNZ8048 (D39 RK Δoperon P), and D39 Δspd0571-0573 nisRK pNZ8048:: spd0571-0573 (D39 RK Δoperon P operon). (B) Expression of the operon increased the resistance of S. pneumoniae against external oxidative stress. D39 nisRK (D39 RK) D39 nisRK pNZ8048 (D39 RK P), and D39 nisRK pNZ8048::spd571-573 (D39 RK P operon). (C) All three proteins cooperate with each other to protect the bacteria against oxidative stress. Survival of exposure to 20 mM H₂O₂ of D39 nisRK pNZ8048 (D39 RK P), D39 nisRK Δspd0571-0573 pNZ8048 (D39 RK Δoperon P), D39 nisRK Δspd0571-0573 pNZ spd0571-0572 (D39 RK Δoperon P spd0571-0572), D39 nisRK Δspd0571-0573 pNZ spd0572-0573 (D39 RK Δoperon P spd572-573), D39 nisRK Δspd0571-0573 pNZ spd0573 (D39 RK Δoperon P spd573), D39 Δspd0571-0573 nisRK pNZ8048:: spd0571-0573 (D39 RK Δoperon P operon). All strains were induced during early mid-log phase by nisin (3 ng/ml) before being exposed to H₂O₂. (D) Complementation in TIGR4 strain. TIGR4 nisRK (T4 RK), TIGR4 Δspd0571-0573 nisRK (T4 RK Δoperon), TIGR4 Δspd0571-0573 nisRK pNZ8048 (T4 RK Δoperon P), TIGR4 Δspd0571-0573 nisRK pNZ8048::spd0571-0573 (T4 RK Δoperon P operon). Error bars indicate standard deviations from the mean. Data are the average of three independent assays. *, P < 0.05.

Fig 4

The three genes form an operon as determined by RT-PCR. (A) Schematic representation of genome organization and location of primers and expected PCR products. (B) RT-PCR analysis of spd0571-0573. Results shown are representative for two independent experiments. (C) Table of primers used and expected product sizes. cDNA was prepared from RNA isolated from D39 harvested at exponential growth and subjected to PCR with the primer pairs as detailed. no RT, cDNA prepared in the absence of reverse transcriptase.

Fig 5

Expression of the operon increased the resistance of L. lactis against external oxidative stress. Survival rates of H₂O₂ are shown for L. lactis NZ9000 containing pNZ8048 (no insert) induced with nisin (white bar), pNZ with the operon and no nisin induction (gray bar), and pNZ with the operon and with nisin (black bar). Error bars indicate standard deviations from the mean. Data are the average of three independent assays. *, P < 0.05.

Fig 6

TlpA is a membrane protein, and its amount responds to oxidative stress. (A) Western blot analysis of strains D39 nisRK and the tlpA mutant determined the specificity of the antiserum and showed the presence of specific anti-TlpA antibodies. (B). Western blot analysis was performed on the supernatant, cytoplasm, and membrane fractions of strain D39 grown in GM17 with the anti-TlpA antibodies. A specific band around 24 kDa was exclusively present in the membrane fraction. As a control for accurate separation, the same fractions were also probed with anti-S. pneumoniae MgtA and anti-S. aureus TrxA antibodies. They are established as cytoplasmic and membrane proteins, respectively, in Gram-positive bacteria. (C) The amount of TlpA is increased under oxidative stress conditions in strain G54, whereas there was no significant increase in strain D39. When bacteria were grown under aerobic conditions and exposed to oxidative stress, the amount of TlpA was increased 2-fold in strain G54 compared to growth without hydrogen peroxide exposure but not in strain D39. (D) In the presence of H₂O₂, G54 has a 1.5-fold larger amount of TlpA than D39. Cells were exposed to H₂O₂ or mock exposed and harvested after 30 min. Protein samples derived from cells that were exposed to H₂O₂ are indicated by an asterisk. Based on CFU counts, total protein extracts of the same amount of cells were loaded on an SDS-PAGE gel and used for Western blot analysis. Samples were equalized based on CFU numbers determined after the oxidative stress procedure under both conditions before loading. In addition, the nonspecific signal around 7 kDa in the antiserum also indicated that similar amounts of protein were used. Signals of Western blotting were quantified using ImageJ.

Fig 7

Deletion of the operon reduces virulence of S. pneumoniae. The mutant has a survival deficiency in the nasopharynx (A), lungs (B), and blood (C). Mice were infected intranasally with TIGR4 and TIGR4 Δsp0658-0660 at a 1:1 ratio; results shown are from two combined experiments. A significant decrease in the amount of the mutant was observed at day 3 in the nasopharynx (A) and at day 5 in the lungs (B) and blood (C). Each circle represents one mouse. The decrease in number of mice over time is a result of mortality. Statistical analysis was done using a nonparametric Mann-Whitney rank sum test testing for a ratio deviating from 1 versus a ratio of 1.

Fig 8

Complementation of the operon mutant enhances virulence of S. pneumoniae. Mice were infected intranasally with TIGR4 Δsp0658-0660 nisRK complemented with pNZ8048 or pNZ8048::spd0571-0573. The results are compiled from three individual experiments. The complemented mutant outcompeted the empty vector control in the nasopharynx on all days tested (A) and was present in greater numbers in the blood (B) from day 3 onward. Each circle represents one mouse. The decrease in number of mice over time is a result of complete bacterial clearance. Statistical analysis was done using a nonparametric Mann-Whitney rank sum test testing for a ratio deviating from 1 versus a ratio of 1.