Identification of the Streptococcus mutans LytST two-component regulon reveals its contribution to oxidative stress tolerance

Abstract

Background: The S. mutans LrgA/B holin-like proteins have been shown to affect biofilm formation and oxidative stress tolerance, and are regulated by oxygenation, glucose levels, and by the LytST two-component system. In this study, we sought to determine if LytST was involved in regulating lrgAB expression in response to glucose and oxygenation in S. mutans.

Results: Real-time PCR revealed that growth phase-dependent regulation of lrgAB expression in response to glucose metabolism is mediated by LytST under low-oxygen conditions. However, the effect of LytST on lrgAB expression was less pronounced when cells were grown with aeration. RNA expression profiles in the wild-type and lytS mutant strains were compared using microarrays in early exponential and late exponential phase cells. The expression of 40 and 136 genes in early-exponential and late exponential phase, respectively, was altered in the lytS mutant. Although expression of comYB, encoding a DNA binding-uptake protein, was substantially increased in the lytS mutant, this did not translate to an effect on competence. However, a lrgA mutant displayed a substantial decrease in transformation efficiency, suggestive of a previously-unknown link between LrgA and S. mutans competence development. Finally, increased expression of genes encoding antioxidant and DNA recombination/repair enzymes was observed in the lytS mutant, suggesting that the mutant may be subjected to increased oxidative stress during normal growth. Although the intracellular levels of reaction oxygen species (ROS) appeared similar between wild-type and lytS mutant strains after overnight growth, challenge of these strains with hydrogen peroxide (H2O2) resulted in increased intracellular ROS in the lytS mutant.

Conclusions: Overall, these results: (1) Reinforce the importance of LytST in governing lrgAB expression in response to glucose and oxygen, (2) Define a new role for LytST in global gene regulation and resistance to H2O2, and (3) Uncover a potential link between LrgAB and competence development in S. mutans.

Figure 1

LytS-dependent expression of lrgAB in S.mutans. Overnight cultures were diluted in THYE, containing either 11 mM (A) or 45 mM glucose (B) to an OD₆₀₀ = 0.02 and grown at 37°C as static cultures at 5% CO₂ (“low-O₂”) or as aerobic shaking cultures at 250 RPM (“aerobic”). RNA was harvested at exponential (EP) and stationary phase (SP). Reverse-transcription, real-time PCR reactions, and determination of copy number were performed as described previously using lrgA and 16S-specific primers [37,77]. Fold-change expression of lrgAB and 16S under each growth condition was calculated by dividing the gene copy number of each test sample by the average gene copy number of UA159 EP. Data was then normalized by dividing each lrgAB fold-change value by its corresponding 16S fold-change expression value. Data represent the average of 3 biological replicates. Dark grey bars represent UA159 and light grey bars represent lytS mutant. Error Bars represent the standard error (SEM).

Figure 2

Distribution of functions of genes affected by loss of LytS at late exponential phase. Statistical analysis was carried out with BRB array tools (http://linus.nci.nih.gov/BRB-ArrayTools.html/) with a cutoff P value of 0.001. The 136 genes differentially expressed at P ≤0.001 are grouped by functional classification according to the Los Alamos S. mutans genome database (http://www.oralgen.lanl.gov/).

Figure 3

Transformation efficiencies of UA159 and isogenic lytS and lrg mutants. To compare the ability of UA159 and its isogenic lytS, lrgA, lrgB, and lrgAB mutants to take up exogenously-added plasmid DNA, a quantitative competence assay was performed on n = 4-6 biological replicates of each strain as described in Methods [83]. Plasmid pAT28 [encoding spectinomycin resistance; [84] was used to assess transformation efficiency in UA159, lytS, lrgB, and lrgAB mutants. Because the lrgA mutant was generated with a spectinomycin-resistance cassette [37], plasmid pORi23 [encoding erythromycin resistance; [85]] was used to assess transformation efficiency in UA159 and lrgA mutant. Transformation efficiencies (Y axis) in the presence (grey bars) and absence (white bars) of CSP are expressed as the percentage of transformants (CFU/ml on BHI + selective antibiotic) among total viable cells (CFU/ml on BHI). Error bars represent SEM. Brackets with P values denote statistically-significant differences between two samples (Mann–Whitney Rank Sum Test).

Figure 4

H₂O₂challenge assay of UA159, lytS and lrgAB mutants. Cultures of UA159, lytS, and lrgAB mutants (n = 6 biological replicates per strain) were grown in the presence (open symbols) and absence (filled symbols) of 1.0 mM H₂O₂ for 20 h at 37°C (aerobic atmosphere) in a Biotek microplate reader. OD₆₀₀ measurements of each well were recorded at 2 h intervals. Black circles represent UA159, red triangles represent lytS mutant, blue squares represent lrgAB mutant. Error bars represent SEM.

Figure 5

Measurement of intracellular ROS in UA159 and lytS mutant by CM-H₂DCFDA staining. Cells were harvested from 20 h BHI cultures of UA159 and isogenic lytS mutant grown at 37°C 5% CO₂ (n = 3-6 biological replicates each), resuspended in HBSS containing 5 μM CM-H₂DCFDA, and incubated at 37°C to load the cells with stain. After 60 min incubation, cell suspensions were centrifuged, washed once in HBSS buffer, and then resuspended in HBSS buffer alone (light grey bars) or in HBSS containing 5 mM H₂O₂ (dark grey bars). Each suspension was transferred to wells of an optically-clear 96 well plate, and incubated at 37°C in a microplate reader. Cell fluorescence (as measured by relative fluorescence units; RFU) and the OD₆₀₀ of each well was recorded after 30 min incubation. RFU measurements are expressed per OD₆₀₀ of each well to account for any subtle variations in cell density. Error bars represent SEM. Brackets with P values denote statistically-significant differences between two samples (Mann–Whitney Rank Sum Test).