Extracellular DNA and lipoteichoic acids interact with exopolysaccharides in the extracellular matrix of Streptococcus mutans biofilms

Abstract

Streptococcus mutans-derived exopolysaccharides are virulence determinants in the matrix of biofilms that cause caries. Extracellular DNA (eDNA) and lipoteichoic acid (LTA) are found in cariogenic biofilms, but their functions are unclear. Therefore, strains of S. mutans carrying single deletions that would modulate matrix components were used: eDNA - ∆lytS and ∆lytT; LTA - ∆dltA and ∆dltD; and insoluble exopolysaccharide - ΔgtfB. Single-species (parental strain S. mutans UA159 or individual mutant strains) and mixed-species (UA159 or mutant strain, Actinomyces naeslundii and Streptococcus gordonii) biofilms were evaluated. Distinct amounts of matrix components were detected, depending on the inactivated gene. eDNA was found to be cooperative with exopolysaccharide in early phases, while LTA played a larger role in the later phases of biofilm development. The architecture of mutant strains biofilms was distinct (vs UA159), demonstrating that eDNA and LTA influence exopolysaccharide distribution and microcolony organization. Thus, eDNA and LTA may shape exopolysaccharide structure, affecting strategies for controlling pathogenic biofilms.

Keywords: Biofilms; Streptococcus mutans; eDNA; exopolysaccharides; extracellular matrix; lipoteichoic acids.

Figure 1

pH measurements from single- and mixed-species biofilm cultures. Notes: pH values for (A) single- and (B) mixed-species biofilms. Biofilms were evaluated at 19, 29, 43, 53, 67, 77, 91, 101 and 115 h by measuring the pH values of the spent medium. Statistically significant differences in pH were observed at 43, 67, 91 and 115 h for the S. mutans parental strain and all mutant strains of S. mutans, tested in both single- and mixed-species biofilms (* two-way ANOVA using as factors biofilm type (single- vs mixed-species) and biofilm developmental phase (time), followed by Tukey’s test; p < 0.05). The data shown are averages and error bars indicate the SDs (n = 6 for all time points except for 67 h, where n = 10).

Figure 2

Population dynamics of S. mutans, S. gordonii and A. naeslundii species during the formation of mixed-species biofilms. Notes: The data shown are average proportions of all species at 29, 43, 53, 67, 77, 91, 101, and 115 h post-inoculation of the biofilm cultures. Data are represented as the percentage total population for each strain: S. mutans (blue line), S. gordonii (orange line), and A. naeslundii (gray line), where in each panel the S. mutans strain is: (A) UA159 parental strain; (B) ΔgtfB; (C) ΔlytT; (D) ΔlytS; (E) ΔdltA; and (F) ΔdltD. The data shown are averages (n = 6 for all time points, except for 67 h, where n = 10).

Figure 3

Bacterial population in single- and mixed-species biofilms at 67 h and 115 h development. Notes: CFU detected in mixed-species biofilms after incubation for 67 h (Panel A) and 115 h (Panel B). No statistical differences in CFU values between strains were detected in the 67 h biofilm cultures (Panel A) (p > 0.05; two-way ANOVA using bacterial species and S. mutans strains as factors, followed by Tukey’s test). In Panel B, aged 115 h, CFU values are shown for the S. mutans strains and S. gordonii from mixed-species biofilms, as A. naeslundii was not detected. S. mutans was statistically the major inhabitant at 115 h in all mixed biofilms tested (p > 0.05; two-way ANOVA using bacteria species and S. mutans strains as factors, followed by Tukey’s test). The S. mutans strain ΔgtfB was detected in significantly higher quantity, compared to the other strains tested (p < 0.05). All biofilms contained significantly increased CFU values for S. mutans at 115 h, compared to biofilms at 67 h (Panel B vs Panel A) (p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Sidak’s test). Elevated numbers of S. gordonii were observed only in biofilms formed with the S. mutans strains UA159, ΔlytS and ΔlytT (^† p < 0.05; two-way ANOVA with developmental phase and S. mutans strains as factors, followed by Sidak’s test). (Panel C) S. mutans CFU values from single-species biofilm cultures, where the symbols indicate significantly different CFU values in pairwise comparisons (p < 0.05). The ΔlytT strain was the only S. mutans strain tested in single-species biofilms that differed in CFU recovered at 115 h, compared to 67 h (p = 0.0205; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). The data shown are averages and error bars indicate the SD (n = 6 for single-species biofilms at 67 and 115 h, and for mixed-species biofilms at 115 h; n = 10 for mixed-species biofilms at 67 h).

Figure 4

Insoluble biomass (dry-weight) and proteins from single- and mixed-species biofilms. Notes: Insoluble biomass was measured from single- (Panel A) and mixed-species biofilms (Panel B) aged 67 h (black bars) and 115 h (white bars). A significant increase in insoluble biomass was observed for all strains between 67 and 115 h for both single- and mixed-species biofilms (p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Sidak’s test). Total protein was determined from single- (Panel C) and mixed-species biofilms (Panel D) at 67 (black bars) and 115 h (white bars). At 67 h, the protein content for both single- and mixed-species biofilms was not significantly different between tested strains. However, at 115 h, ΔgtfB displayed a lower amount of protein in single- (vs ΔlytT, ΔdltD and ΔdltA) and mixed-species biofilms (vs ΔdltD and ΔdltA) (p < 0.05, two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). Moreover, there was a significant increase in protein content from 67 to 115 h for ΔdltD and ΔdltA in single-species biofilms (p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Sidak’s test). The plotted data are averages, and error bars indicate SDs (n = 6 for single-species biofilms at 67 and 115 h, and for mixed-species biofilms at 115 h; n = 10 for mixed-species biofilms at 67 h).

Figure 5

eDNA and LTA in the ECM of single- and mixed-species biofilms. Notes: Total amounts of eDNA were measured from single- (panel A) and mixed-species (panel B) biofilms at 67 h (black bars) and 115 h (white bars). In single-species biofilms, at 67 h, the parental strain UA159 yielded similar amounts of eDNA to those observed from the ΔlytT and ΔlytS strains (^γp > 0.05), and quantities were greater than those measured from the other strains tested; while at 115 h, ΔlytT and ΔlytS biofilms contained a greater amount of eDNA. Significant statistical differences are shown in comparisons of like symbols (*p < 0.0001; ^§p ≤ 0.0001; ^¤p ≤ 0.0141; ^#p < 0.0001; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). For mixed-species biofilms (Panel B), the amount of eDNA detected at 67 h was similar for all strains (p > 0.05), while at 115 h, a significantly higher amount of eDNA was measured in biofilms with ΔlytT and ΔlytS compared to all other strains (*p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). In addition, total amounts of extracellular LTA were measured from single- (panel C) and mixed-species (panel D) biofilms at 67 h (black bars) and 115 h (white bars). Overall, the amount of LTA measured at 67 and 115 h increased as the biofilms aged; however, the values were statistically significant only in single-species biofilms of S. mutans UA159 (Panel C), and in mixed-species biofilms containing UA159, or the deletion strains ΔlytT and ΔdltA (Panel D) (*p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Sidak’s test). In mixed-species biofilms, the amounts of LTA (Panel D) were significantly different between the ΔlytT and ΔdltD strains at 67 h (#); while at 115 h, ΔdltA contained more LTA than the ΔlytT and ΔlytS strains (^§p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). The plotted data are averages, and error bars indicate the SDs (n = 6 for single-species biofilms at 67 and 115 h, and for mixed-species biofilms at 115 h; n = 10 for mixed-species biofilms at 67 h).

Figure 6

Water-soluble and -insoluble exopolysaccharides in the ECM of single- and mixed- species biofilms. Notes: Total amounts of extracellular water-soluble exopolysaccharides (WSP) were measured from single- (Panel A) and mixed-species (Panel B) biofilms at 67 h (black bars) and 115 h (white bars). Total amounts of extracellular water-insoluble exopolysaccharides (ASP) were measured from single- (Panel C) and mixed-species (Panel D) biofilms at 67 h (black bars) and 115 h (white bars). For both single- and mixed-species biofilms, there was a significant increase in the amount of WSP and ASP in the ECM between 67 to 115 h (p < 0.05; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Sidak’s test – symbols representing statistical analysis outcome are not depicted in the graphs). In single-species biofilms (Panel A), ΔgtfB contained less WSP in the ECM at both 67 and 115 h, compared to all other strains tested, except ΔlytT, at 67 h (*p ≤ 0.03; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). The biofilm derived from the ΔlytT strain was composed of less WSP at the 67 h time point (vs strains ΔdltA and ΔdltD; #p ≤ 0.0278), and at 115 h (vs parental strain; ¤p = 0.0099; two-way ANOVA, followed by Tukey’s test). In mixed-species biofilms (Panel B), there was no significant variation in WSP between strains at 67 h or at 115 h (p > 0.05). ASP measurements were similar for all strains at 67 h in single-species biofilms (Panel C); while at 115 h, the ΔgtfB strain contained less ASP (vs ΔlytT and ΔlytS strains; ^§p ≤ 0.02; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). In mixed-species biofilms, there were no differences between strains for ASP at 67 h (Panel D); however, at 115 h, there was less ASP for the ΔgtfB strain (vs parental strain UA159, ΔdltA and ΔdltD; ^γp ≤ 0.02; two-way ANOVA using developmental phase and S. mutans strains as factors, followed by Tukey’s test). The plotted data are averages, and error bars indicate the SDs (n = 6 for single-species biofilms at 67 and 115 h, and for mixed-species biofilms at 115 h; n = 10 for mixed-species biofilms at 67 h).

Figure 7

3D architecture of the single- and mixed species biofilms. Notes: Representative 3-D renderings of single- (upper panel) and mixed-species (lower panel) biofilms formed by S. mutans parental (UA159) and mutant strains (ΔgtfB, ΔlytT, ΔlytS, ΔdltA, and ΔdltD) at 115 h. Biofilms were labeled as described in Materials and methods. Nucleic acids were stained green with SYTO^® 9 green fluorescent nucleic acid stain and the extracellular polysaccharides are stained red with Alexa Fluor^® 647-labeled dextran conjugate. The imaging was performed using a Zeiss LSM 780 microscope equipped with a 20× objective lens. The larger image in each set represents the overlaid images of the red and green channels. The lower image is a cross-section of the biofilm with overlaid images. Scale bar: 25 μm.

Figure 8

Biovolume of bacteria and EPS in single- and mixed species biofilms. Notes: Biovolume is represented as biomass (μm³ μm⁻²) of bacteria (black bars) and EPS (white bars) for single- (Panel A) and mixed-species (Panel B) biofilms at 115 h. (Panel A) For single-species biofilms, two-way ANOVA demonstrated no interaction between distinct strains and biofilm component (bacteria vs EPS). Sidak’s multiple comparison test demonstrated significant differences for all mutants compared to parental strains (p < 0.05), except for ΔlytT. Moreover, no differences were found by comparing all mutants against each other for both bacteria and EPS biomass. The plotted data are averages, and error bars indicate the SDs (n = 4). (Panel B) For mixed-species biofilms, two-way ANOVA demonstrated no interaction between distinct strains and biofilm component, but distinct strains factor and biofilm component factor (bacteria vs EPS) were significant (p < 0.0001). The Tukey’s multiple comparison test demonstrated that bacteria and EPS biomass of biofilm with parental strain UA159 was significantly different (higher) than all deletion strains tested (p < 0.05). The plotted data are averages, and error bars indicate the SDs (n = 4 for single-species biofilms; n = 8 for mixed-species biofilms).

Figure 9

Profile of the distribution of bacteria and EPS in each of the single-species biofilms. Note: The data shown are the mean percentage coverage per area from the interface substratum/biofilm (HA disk) to the top (outer layer) of each biofilm at 115 h (n = 12 images per biofilm).

Figure 10

Profile of the distribution of bacteria and EPS in each of the mixed-species biofilms. Note: The data shown are the mean percentage coverage per area from the interface substratum/biofilm (HA disk) to the top (outer layer) of each biofilm at 115 h (n = 15 images per biofilm).