A Highly Arginolytic Streptococcus Species That Potently Antagonizes Streptococcus mutans

Abstract

The ability of certain oral biofilm bacteria to moderate pH through arginine metabolism by the arginine deiminase system (ADS) is a deterrent to the development of dental caries. Here, we characterize a novel Streptococcus strain, designated strain A12, isolated from supragingival dental plaque of a caries-free individual. A12 not only expressed the ADS pathway at high levels under a variety of conditions but also effectively inhibited growth and two intercellular signaling pathways of the dental caries pathogen Streptococcus mutans. A12 produced copious amounts of H2O2 via the pyruvate oxidase enzyme that were sufficient to arrest the growth of S. mutans. A12 also produced a protease similar to challisin (Sgc) of Streptococcus gordonii that was able to block the competence-stimulating peptide (CSP)-ComDE signaling system, which is essential for bacteriocin production by S. mutans. Wild-type A12, but not an sgc mutant derivative, could protect the sensitive indicator strain Streptococcus sanguinis SK150 from killing by the bacteriocins of S. mutans. A12, but not S. gordonii, could also block the XIP (comX-inducing peptide) signaling pathway, which is the proximal regulator of genetic competence in S. mutans, but Sgc was not required for this activity. The complete genome sequence of A12 was determined, and phylogenomic analyses compared A12 to streptococcal reference genomes. A12 was most similar to Streptococcus australis and Streptococcus parasanguinis but sufficiently different that it may represent a new species. A12-like organisms may play crucial roles in the promotion of stable, health-associated oral biofilm communities by moderating plaque pH and interfering with the growth and virulence of caries pathogens.

FIG 1

(A) Arginine deiminase enzyme activity in S. gordonii DL1 and A12 wild-type (WT) and ΔarcR mutant strains grown in TY medium with or without supplemental arginine. (B) Arginine deiminase activity in wild-type and ΔarcR mutant strains of S. gordonii grown in chemically defined FMC medium containing the indicated amounts of arginine. A12 and S. gordonii DL1 were not able to grow in FMC medium unless some arginine was added. In panel A, asterisks indicate statistical significance between the wild-type and arcR mutant strains under both growth conditions using an alpha value of 0.05. In panel B, comparisons are between the arcR mutants of A12 and S. gordonii.

FIG 2

(A) Plate-based growth inhibition assays of arginolytic strains versus S. mutans UA159. In all cases, cultures grown overnight were collected by centrifugation and resuspended at an OD₆₀₀ of 0.5 in PBS. In the left two columns, the arginolytic strain S. gordonii DL1 or A12 was spotted onto plates and incubated for 24 h in a 5% CO₂ aerobic atmosphere, an equivalent amount of S. mutans UA159 was then spotted adjacent to S. gordonii or A12, and the plates were incubated an additional 24 h. In the two columns on the right, A12 or S. gordonii DL1 was spotted at the same time as S. mutans UA159, and the plates were incubated for a total of 24 h. The addition of catalase to the medium effectively eliminated the inhibition of S. mutans by the commensal, but inclusion of proteinase K did not have the same effect. (B) H₂O₂ production by S. gordonii or A12 is dependent on carbohydrate source and nutrient availability. Cells were grown under aerobic conditions in the indicated media, and the concentration of H₂O₂ in the supernatants was measured as described in Materials and Methods. Asterisks indicate statistical significance between growth on glucose and growth on galactose for each organism (alpha = 0.05). (C) Comparison of pyruvate, lactate, l-amino acid, and NADH oxidase enzyme levels in S. gordonii DL1 and A12 (see the text for methodology). Pyruvate oxidase (Pox) was the dominant H₂O₂-producing oxidase measured in both organisms under the conditions tested, and A12 produced significantly higher levels of Pox enzyme activity than did S. gordonii. Asterisks signify statistical significance between pyruvate oxidase levels in A12 and those in S. gordonii.

FIG 3

(A and B) Plate-based mutacin assays using S. sanguinis SK150 as the indicator strain. The indicated strains were mixed in equal proportions and spotted onto BHI agar plates, and the plates were incubated overnight under aerobic (A) or anaerobic (B) conditions. Subsequently, 3 ml of a soft BHI agar overlay containing 10⁷ S. sanguinis SK150 bacteria was gently and evenly overlaid onto the plate, and incubation was continued for 24 h prior to measurement of zones of inhibition. Mutants of A12 or S. gordonii lacking the Sgc protease were unable to block mutacin production to protect the indicator strain. (C) An agar plug was obtained from the inoculation spot after the experiment, cells were dispersed, and the proportions of bacteria of S. mutans UA159 and the indicated S. gordonii or A12 strains were enumerated by dilution and plating (see Materials and Methods for more details). Asterisks indicate statistically significant differences in the proportions of S. mutans bacteria cocultivated with the indicated mutant strains compared to cocultivation of wild-type A12 with S. mutans.

FIG 4

Effects of treatment of sCSP with supernatants from cultures of the indicated strains grown overnight. A strain of S. mutans containing a lacZ fusion to the cipB promoter was grown to an OD₆₀₀ of 0.12 in BHI medium (optimal responsiveness to sCSP). Supernatants from cultures of the indicated strains grown overnight were obtained after centrifugation, adjusted to pH 7.0 (also optimal for sCSP signaling), and filter sterilized. sCSP was added to the supernatants, and the mixtures were incubated for 2 h. The S. mutans cipB-lacZ reporter strain was then resuspended in the indicated supernatants for 2 h prior to measurement of β-galactosidase (LacZ) activity. The Sgc protease was necessary for inhibition of sCSP signaling. The results shown are from a minimum of three biological replicates, each performed in triplicate. Values are averages, and error bars indicate standard deviations. Asterisks indicate statistically significant differences compared with wild-type A12.

FIG 5

Effects of treatment of XIP with supernatants from cultures of the indicated strains grown overnight. Supernatant fluids from cultures of A12 and its derivatives, S. gordonii, or S. mutans UA159 that had been grown in FMC medium overnight were adjusted to pH 7.0, supplemental glucose was added to increase the glucose concentration by 25 mM, and the supernatants were filter sterilized. sXIP was added to the supernatants, and the mixtures were incubated overnight at 37°C. Cultures of S. mutans UA159::PcomX-lacZ that had been grown in FMC medium to an OD₆₀₀ of 0.12 were pelleted by centrifugation and resuspended in the supernatants. The cultures were incubated for 2 h, and LacZ assays were performed to detect XIP-dependent activation of the comX promoter. The results shown are from a minimum of three biological replicates, each performed in triplicate. Values are averages, and error bars indicate standard deviations. Data from treatment with 400 or 1,000 nM sXIP for DL1 and S. mutans are not shown because neither supernatant influenced sXIP signaling any differently than the positive control. Asterisks indicate statistical significance in comparisons between wild-type A12 with 200 nM XIP and DL1 with 200 nM XIP and between wild-type A12 with 1,000 XIP and Spx-deficient A12 with 1,000 nM XIP.

FIG 6

Effects of growth conditions on interference of XIP signaling by supernatants from cultures of the indicated strains grown overnight. All experiments were performed as detailed in the legend of Fig. 5 except that cells were grown in different atmospheres or supernatants were treated with catalase. Supernatants from A12 cells grown anaerobically had an apparently enhanced capacity to interfere with sXIP signaling, and the effects were not impacted by the inclusion of catalase. S. gordonii again was unable to interfere with sXIP signaling, and data for treatment with 400 or 1,000 nM sXIP for DL1 are not shown because there was no difference from the positive control. Catalase itself also had no effect on signaling (FMC 5 μg/ml catalase bars). The results shown are from a minimum of three biological replicates, each performed in triplicate. Values are averages, and error bars indicate standard deviations. Asterisks indicate statistical significance between A12 grown in CO₂ and A12 grown anaerobically and between A12 grown in CO₂ and A12 grown with catalase treatment.

FIG 7

Maximum likelihood phylogeny showing relationships among 14 Streptococcus species and A12. Numbers on the branches show the proportions of gene phylogenies that support the grouping.