2.5.2. MDFR

The bioreactor used in the present study was a modification of a commercially available drip-flow reactor (DFR 110, BioSurface Technologies; Bozeman, MT, USA) according to previously described [21]. The modified design (MDFR) allowed the placement of customized PTFE trays on the bottom of the flow cells to submerge the specimen surfaces in the flowing medium. Specimens from each group were randomly divided into the eight flow cells of two identical MDFRs. After sterilization (STERRAD), the MDFR was assembled inside a sterile hood, and a salivary pellicle was obtained on the specimen surface after incubation with sterile saliva at 37 ◦C for 24 h. Saliva was then removed, each flow-cell was inoculated with 10 mL of either *S. mutans* suspension or mixed oral flora suspension to allow bacterial adherence. After 4 h, a peristaltic pump (RP-1k; Rainin, Emeryville, CA, USA) provided the constant flow of sterile modified artificial saliva medium through the flow cells (9.6 mL/h). For each microbiological inoculum, the amount of viable adherent biomass was evaluated after either 12 h or 24 h of further incubation in aerobic conditions (Figure 1).

#### *2.6. Viable Biomass Assessment*

Viable and metabolically active biomass adherent to the specimen surface was assessed using a tetrazolium-based assay as described previously [21]. In brief, a tetrazolium salt stock solution was prepared by dissolving 5 mg/mL 3-(4,5)-dimethylthiazol-2-yl-2,5-diphenyltetrazolium bromide (MTT) in sterile PBS; a phenazinium salt stock solution was prepared by dissolving 0.3 mg/mL of N-methylphenazinium methyl sulphate (PMS) in sterile PBS. The solutions were stored at 2 ◦C in light-proof vials until the day of the experiment when a fresh measurement solution (FMS) was prepared by diluting 1:10 v/v of MTT stock solution and 1:10 v/v of PMS stock solution in sterile PBS. A lysing solution (LS) was prepared by dissolving 10% v/v of sodium dodecyl sulphate (SDS) and 50% v/v dimethylformamide in distilled water and stored at 2 ◦C until the day of the experiment when it was warmed at 37 ◦C for 2 h before use. After the specified incubation times, the orbital shaker and the MDFR medium flow was halted, flow cells and plates were opened, the specimens were carefully removed and immediately placed into Petri plates containing sterile PBS at 37 ◦C. They were gently washed three times with sterile PBS to remove non-adherent cells and finally placed inside the wells of 48-well plates containing 300 μL of FMS each. The plates were incubated at 37 ◦C under light-proof conditions for 3 h. During incubation, electron transport across the microbial plasma membrane and, to a lesser extent, microbial redox systems, converted the yellow salt to insoluble purple formazan crystals. The conversion at the cell membrane level was facilitated by the intermediate electron acceptor (PMS). The unreacted FMS was gently removed by aspiration, and the formazan crystals were dissolved by adding 300 μL of LS to each well. The plates were stored for an additional 1 h under light-proof conditions at room temperature; 100 μL of the solution was then transferred into

96-well plates. The absorbance of the solution was measured using a spectrophotometer (Genesys 10-S, Thermo Spectronic, Rochester, NY, USA) at a wavelength of 550 nm; results were expressed as relative absorbance in optical density (OD) units corresponding to the amount of adherent, viable and metabolically active biomass.

#### *2.7. Confocal Laser-Scanning Microscopy (CLSM)*

A total of three specimens for each material and treatment group (Figure 1) were prepared for CLSM analysis. They were incubated using the MDFR test setup with the artificial oral microcosm model for 12 h. Then, they were gently removed from the flow cells, rinsed twice with sterile PBS, stained using the FilmTracer™ LIVE/DEAD® Biofilm Viability Kit (Invitrogen Ltd., Paisley, UK), and analyzed using confocal laser-scanning microscopy (CLSM; Eclipse Ti2 inverted CLSM, Nikon, Tokyo, Japan). Three randomly selected image stack sections were recorded for each specimen. Confocal images were obtained using a dry objective (20<sup>×</sup>; NA = 0.5) at a resolution of 2048 × 2048 pixels, with a zoom factor of 1.0 and a scan speed of 400 Hz, and digitalized using the Nikon Imaging Software (NIS)—Elements Viewer, v. 4.50. Three channels were acquired in parallel scanning; the first used an excitation at 405 nm and emission at 420–470 nm in order to subtract potential autofluorescence digitally. The other two channels had an excitation wavelength of 488 nm, and emission was acquired at 500–570 nm (green channel, live bacteria) and 610–760 nm (red channel, dead bacteria). For each image stack section, 3D-rendering reconstructions were obtained using Drishti (Ajay Limaye, Australian National University, CAN, AUS, http://sf.anu.edu.au/Vizlab/drishti/).

#### *2.8. Fluoride Release Measurements*

Measurements of fluoride release from the fluoride-hydroxyapatite-containing toothpaste (β) were performed to correlate it with the hypothesized antimicrobial activity of the toothpaste. A total of 48 enamel disks and 48 RBC disks were additionally prepared and brushed as previously specified (cf. Sections 2.1 and 2.2): one half (n = 24) was brushed with slurry from toothpaste β, and the other half was brushed with distilled water. Immediate fluoride release (t = 0) was evaluated from eight disks for each toothpaste. These disks were placed in 48-well plates containing 300 μL of sterile saliva in each well for 24 h. Disks were then discarded, 1:50 dilution of total ionic strength solution adjustment buffer, TISAB III (Orion Research Inc, Boston, MA, USA) was added, and the solution adjusted to pH = 5.5. The fluoride content of the solution was determined using the ion-selective electrode micro-method. The remaining disks were weighted with an analytical balance, then sterilized, and a monospecies *S. mutans* biofilm was allowed to develop on their surfaces after salivary pellicle formation using the MDFR as described in Section 2.5.2 for the same incubation times (12 h and 24 h). A total of eight disks for each toothpaste and time group were weighed again, and the weight of the adherent wet biomass was calculated. After that, each disk was inserted into one well of 48-well plates containing 300 μL of 10 vol% SDS in distilled water and 1:50 TISAB III (pH = 5.5), and the plates were sonicated for 5 min to allow dispersion of the biomass. Then, disks were discarded, and the fluoride content of the solution was determined.

The amount of released fluoride was calculated and displayed in parts per billion (ppb) after normalizing the fluoride readings by the disks' surfaces. The amount of fluoride incorporated in the biofilm structures was calculated and displayed in parts per million (ppm) after normalizing the fluoride readings by the amount of adherent biomass. All analyses were performed using the ion-selective electrode micro-method, as described previously [22]. In brief, a stock solution with a fluoride concentration of 1000 ppm was appropriately diluted with 10 vol% SDS in distilled water at pH = 5.5 to obtain fluoride standards with fluoride concentrations ranging from 0.0019 to 64 ppm. A calibration curve was obtained using a digital pH/mV meter (SA-720, Orion Research Inc, Boston, MA, USA). A 1:50 final dilution of TISAB III as an ionic strength adjustor was added to each standard before the analyses. A negative reference standard (0 ppm fluoride) was prepared by adding 1:50 TISAB III to the

solution containing 10 vol% SDS in distilled water; this solution was also used to rinse the electrodes between measurements.

## *2.9. Statistical Analysis*

All statistical analyses were performed using statistical software (JMP 12.0, SAS Institute, Inc., Cary, NC, USA). The OD and F data were reported as means and standard errors calculated from the natural values. The normality of distributions was preliminarily checked using Shapiro-Wilk test, and homogeneity of variances was preliminarily checked using Bartlett's test. A multi-way ANOVA model was used on OD dataset considering the factors: toothpaste ( α, β, control), adherence surface (enamel, RBC), test setup (shaking multiwell plate, MDFR), microbiological model (*S. mutans*, oral microcosm), and incubation time (12 h, 24 h). A multi-way ANOVA model was used on fluoride dataset considering the factors: toothpaste (β, control), adherence surface (enamel, RBC), and incubation time (12 h, 24 h). Student-Newman-Keuls post-hoc test was used to highlight significant di fferences (*p* < 0.05).
