of 5, 5.5, 6, 6.5, 7, 7.5, and 8 were achieved.
The biofilms were formed on 96-well plates. First the surface was coated with a proteinaceous solution (25% human serum, 0.27% mucin; both Sigma-Aldrich) for 1 h, before 250 µL/well of the bacteria/nutrient broth suspension was added. Thereafter, the plates were incubated at 37 °C with 10% CO2 or anaerobically (periodontitis biofilm) for different lengths of time. The “healthy” biofilm was analyzed after 2 and 6 h, the “cariogenic” biofilm after 6 and 24 h, and the “periodontal” biofilm after 24 and 48 h. For the “periodontal” biofilm, 10 µL of microbial suspension consisting of one part each of T. denticola, T. forsythia, and P. gingivalis were added again after 24 h. Each of the three different plates were used in one experimental setting and per time point.
At the respective times, the nutrient broth was removed and the biofilms were careful washed once with 0.9% w/v NaCl. From the first plate, the biofilms were scraped from the surface and suspended in 0.9% w/v NaCl. After intensive mixing by pipetting and vortexing, a serial 10-fold dilution series was made. Each 25 µL were spread on tryptic-soy agar plates with 5% sheep blood (and 10 mg/L NAM). After incubation at 37 °C with 10% CO2 or anaerobically (the “periodontitis” biofilm) for about 7 days, the total bacterial counts (log10 colony-forming units; CFU) as well as the percent of the different bacteria used were determined. The identification was based on the colony morphology. As this was in part difficult or impossible (T. denticola does not grow on the agar plates), P. gingivalis, T. forsythia, and T. denticola, were counted using real-time PCR as described previously [6].
The second 96-well plate was used for the determination of metabolic activity with the use of Alamar blue reagent as a redox indicator [7]. A total of 5 µL of Alamar blue (alamarBlue®, Thermo Fisher Scientific Inc., Waltham, MA, USA) was mixed with 100 µL of the nutrient media and added to the biofilm. After extensive mixing with the biofilm and an incubation for 1 h at 37 °C, absorbances were measured at 570 nm against 600 nm.
The biofilm mass was quantified from the third 96-well plate according to recently published protocols [8]. First, the biofilms were fixed at 60 °C for 60 min. Thereafter, each 50 µL of 0.06% crystal violet (Sigma-Aldrich) was added. The plate was incubated at 37 °C for 10 min, then washed 3 times with 200 µL of dH2O. Finally, the plate was read at 600 nm.
Each experiment was performed in quadruplicate in two independent series, resulting in at least eight single values each. ANOVA with post hoc Bonferroni was used for the statistical analysis. The level of significance was set to p = 0.05 and SPSS v.24.0 software (IBM SPSS Statistics, Chicago, IL, USA) was used.
Results
We present the total bacterial counts, the percentages of the respective bacteria (or groups), the metabolic activity, and the biofilm masses. Post hoc statistical analysis was restricted to the differences from pH 7.
Biofilm Representing Oral Health
Total counts of bacteria within each biofilm (log10 CFU) differed significantly between the initial pH values at 2 and 6 h (each p < 0.001; Fig. 1a). At 2 h, the lowest values were counted at pH 5 and 5.5 with approximately 5.8 log10 CFU, while the highest counts were detected at pH 7 with a mean of 6.6 log10 CFU. Post hoc analysis showed differences between pH 5, 5.5, 7.5, and 8 on the one hand, and pH 7 on the other (each p < 0.01). At 6 h, the lowest CFU counts remained at 5.7 log10 when the initial pH was 5, the highest value increased to 7.3 log10 (pH 7.5). Also at 6 h, the CFU count differences were each statistically significant between pH 5, 5.5, and 7 (each p < 0.01).
Fig. 1. Total bacterial counts (mean and SD; a), bacterial composition (b), metabolic activity (mean and SD; c), and biofilm mass (mean and SD; d) of the two-species biofilm representing oral health in relation to the initial pH. Differences versus pH 7 are presented: * p < 0.05, ** p < 0.01 (the asterisk colors correspond with the composition of biofilms in b).
At 2 h, the quantity of the two species did not differ significantly between the different pH levels. At 6 h, there was a significantly higher percentage of A. naeslundii and a lower percentage of S. gordonii at pH 5 in comparison with pH 7 (each p < 0.001; Fig. 1b).
The metabolic activity differed between the different pH values at 2 and 6 h (each p < 0.001; Fig. 1c). At both time points, the lowest activities were at pH 5 and 5.5, each with a statistically significant difference to pH 7 at 2 h (each p < 0.05) and at 6 h (each p < 0.01).
Regarding biofilm mass, no statistically significant difference from the initial pH values was found, either at 2 or 6 h (Fig. 1d). The pH measurements performed at 6 h still showed a gradient from pH 5.15 (initially pH 5) to pH 7.05 (initially pH 8; Table 2).
Table 2. Initial pH values and values (mean ± SD) in the surrounding media after 6 h of formation of the two-species biofilm representing “oral health”
Fig. 2. Total bacterial counts (mean and SD; a), bacterial composition (b), metabolic activity (mean and SD; c), and biofilm mass (mean and SD; d) of the five-species biofilm representing caries in relation to the initial pH. Differences versus pH 7 are presented: * p < 0.05, ** p < 0.01 (the asterisk colors correspond with the composition of biofilms in b).
Biofilm Representing Caries
In the biofilms representing caries, total counts of bacteria within each biofilm (log10 CFU) differed significantly between the initial pH values at 6 and 24 h (each p < 0.001; Fig. 2a). At 6 h, the lowest values were counted at pH 5 with approximately 5.8 log10 CFU, while the highest counts were recorded at pH 6 with a mean of 7.7 log10 CFU. Post hoc analysis showed differences between each pH 5 and 5.5 on the one hand, and pH 7 on the other (each p < 0.01). At 24 h, the lowest CFU counts were 6.0 log10 when the initial pH was 5; the highest value was 9.7 log10 (pH 7 and 8). At 24 h, the CFU count differences were each statistically significant between pH 5, 5.5, 6, and 7 (each p < 0.01).
At 6 h, there were no statistically significant differences in the percentage of S. gordonii, A. naeslundii, and S. mutans/S. sobrinus between the pH levels. The percentage of L. acidophilus was higher at pH 5 and 5.5 versus pH 7 (p < 0.001, p = 0.017). At 24 h, the percentage of L. acidophilus remained high at pH 5 (p < 0.001 vs. pH 7). Furthermore, there was a significantly higher percentage of S. mutans/S. sobrinus at pH 5 (p < 0.001) and pH 6 (p = 0.016), and a lower percentage of S. gordonii at pH 5, 5.5, and 6 (each p < 0.001) versus pH 7 (Fig. 2b).
The metabolic activity differed between the different pH values at 6 and 24 h (each p < 0.001; Fig. 2c).