The consenting patient with caries/poor oral health, as identified by the dentist in RAK College of Dental Sciences (RAKCODS) by having at least two active dental caries and a fair or poor oral health, was asked to chew an orthodontic elastic for 2 minutes. This was placed in his mouth by the principal investigator or one of the co-investigators. It was necessary that the patient was 20–55 years old and had not eaten or drunk for 2 hours prior; it was also required that they were previously and currently clear of diabetes, cancer, and systemic infections. Similarly, a healthy volunteer, as reported by the dentist, underwent the same procedure and a healthy sample was procured from him. The participant was then asked to spit into a chilled sterile container provided. Afterward, the stimulated whole saliva samples were packaged with care to not allow contamination, then delivered within 30–60 minutes to the laboratory.
We opted to collect stimulated saliva samples because they exhibit a more even and true representation of the oral ecosystem, as reported by Vercher et al. . Although the genus Streptococcus is by far the most abundant in unstimulated saliva samples, it appears in higher quantities yet in lower proportions in stimulated saliva samples, probably because of plaque removal during chewing of paraffin in stimulated saliva sample collection.
The unhealthy saliva samples (from patients with caries/poor oral health) were cultured on TYCSB (tryptone yeast extract cystine with sucrose and bacitracin) agar to allow the growth of S. mutans from those samples. The culture plates were then kept in the incubator in a candle jar (5% CO2 atmosphere) for 48 hours at 37 °C. Afterward, they were sub-cultured on BHI (brain heart infusion broth) in test tubes. These tubes were covered with parafilm and kept in the incubator in a candle jar overnight at 37 °C. TYCSB agar was chosen in this study because it was determined to be the most sensitive and selective media for culture of S. mutans for laboratory and clinical studies, as the recovery of the laboratory S. mutans strain was highest in TYCSB agar when compared to other commercially available selective media; and ratios of mutans to non-mutans bacteria were also highest in TYCSB .
Meanwhile, saliva samples from healthy volunteers were centrifuged after collection at 4,000 × g for 10 minutes to remove cellular debris. They were then subjected to filter sterilization through a 0.22-µm acrodisc filter. The clarified saliva at the bottom was then stored in the freezer at 0 °C until they were used to coat the wells of the microtiter plate.
Biofilm and growth assays preparation
Biofilm formation by S. mutans was assayed using the component crystal violet staining method described by Ahn et al. with some modifications . This method remains the most frequently used quantitation technique in microplate assays, despite its limitations and the development of more advanced methods . The assays were done using polystyrene 96-well flat-bottom microtiter plates in quadruplicates, in which four different samples from patients with caries were processed at a time. Two microtiter plates were used for each set of samples (Fig. 1). The first plate was used to assay the inhibitory effect of vitamin C on biofilm formation by S. mutans (MBIC) and its ability to detach previously formed biofilm; the second plate was used to assay the inhibitory effect of vitamin C on bacterial growth (minimum inhibitory concentration or MIC). Accordingly, conditions were made more favorable for biofilm formation in the former plate, as exemplified by saliva-coating of the wells and addition of 3% sucrose to the broth, and less favorable in the latter by eliminating those two factors. Filtered salivary preparations from healthy volunteers were used to coat the wells of the first microtiter plate, before adding the cell suspensions. This method is thought to promote bacterial adherence and, therefore, biofilm formation by providing binding receptors to bacterial proteins. To achieve this, each well was conditioned with 100 µl of filtered saliva. The plate was incubated at 37 °C for 2 hours with gentle shaking and then washed three times with PBS (phosphate-buffered saline). The wells were then allowed to air-dry for 30 minutes.
Serial dilutions of vitamin C
The plates were processed as follows: 200 µL of vitamin C at a concentration of 25 mg/mL was added to the wells of the first column; 100 µL of broth (BHI) was also added to the next nine wells along the same row. Serial two-fold dilutions were then carried out from the first to seventh wells, such that vitamin C concentration was 25 mg/ml in the first well and 0.391 mg/ml in the seventh well. As for the eighth well, it was kept as a control for biofilm formation wherein Staphylococcus aureus ATCC (American type culture collection) 25923, a bacterium with a known capacity of biofilm formation, was grown. Besides, the ninth and tenth wells were kept as positive and negative controls, respectively. Ten µL of S. mutans subculture from the respective sample and cultured Staphylococcus aureus ATCC at 0.5 McFarland standard were added into wells one to nine (except eight) and eight, respectively. Ultimately, each well (except the tenth) contained 110 µL and each saliva sample was represented by a row on the microtiter plate; with columns one through seven being serial dilutions of vitamin C onto cell suspensions, column eight being the control of biofilm formation, nine being the positive control, and ten being the negative control (Fig. 1).
A set of wells in the first plate was conditioned to test vitamin C-induced detachment. After coating the wells with saliva, columns one through seven were inoculated with 100 µL of broth and 10 µL of bacteria at 0.5 McFarland standard of the respective sample, keeping wells eight, nine, and ten controls as before. Both plates were subsequently placed in the incubator at 37 °C in the candle jar for 16–18 hours. Growth on the first plate was then visually analyzed, and 100 µL of vitamin C was added to the wells that were used to study the vitamin C-induced detachment of the biofilm. The concentration of the added vitamin C was 25 mg/mL at the very first well, and it was diluted two-fold up to the last well before the controls. The microtiter plates were again placed in the incubator at 37 °C in the candle jar for 24 hours.
The growth on the second microtiter plate was interpreted visually to determine the minimum inhibitory concentration (MIC) using the aforesaid microtiter plate broth dilution or microdilution method, wherein the vitamin C concentration in the last well from columns one to seven in each row visually lacking growth was identified as MIC. The overall average MIC was calculated by averaging the MICs of all the samples we tested, each representing a row on the second microtiter plate (Fig. 1). Following this, on a single TYCSB or blood agar, each row with its first 10 wells was sub-cultured by transferring 10 µL into separate zones of the agar plate to determine the minimum bactericidal concentration (MBC). The agar plate was then kept in the incubator at 37 °C in a candle jar for 24 hours. Afterward, MBC was visually interpreted as the vitamin C concentration in the last zone of the agar not showing growth. The overall average MBC was calculated by averaging the MBCs of all the samples we tested in our four quadruplicate experiments. A similar setup was used by Moradian et al.  to determine the MIC and MBC of certain compounds against S. mutans.
Biofilm and detachment assays
On the same day, biofilm formation assay was performed for both plates by crystal violet staining method. Firstly, the contents of the well, comprising planktonic or free-floating bacteria, were removed with care; leaving behind the growth at the bottom of the well. Each well was then gently washed twice with 200 µL of sterile distilled water and dried for 10 minutes. Fifty µL of 0.1% crystal violet was then added to the wells, and the plates were kept for 15 minutes at room temperature to stain the remaining adherent bacteria. After the wells were rinsed twice with 200 µl of distilled water, the bound dye was extracted from the stained cells using 200 µl of 99% ethanol, which was left for 5 minutes. The biofilm was then quantitated by measuring the absorbance of the solution at 560 nm in the GloMax microplate ELISA reader (GM3500, Promega, Madison, WI, USA). The readings from the first plate were then used to determine the MBIC and were used to interpret the detachment assay, whereas those from the second plate were merely used to confirm that the conditions we placed this plate in were not favorable for biofilm formation. In other words, biofilm is not expected to form in the second plate, and, therefore, the readings were anticipated to be entirely infra-threshold.
According to Thieme et al. , MBIC should be assessed by comparing post-treatment biofilms to their baseline. This can be obtained by measuring the pre-treatment biofilm or a standardized control of biofilm formation, such as the S. aureus ATCC that we included in our study as the positive control for biofilm formation. A similar set up was used by Saputo et al.  to measure the MBIC of different vitamin D compounds against S. mutans. Consequently, we averaged the absorbance of S. aureus biofilm, from the four quadruplicate experiments. Afterward, this value was used as a cutoff for determining MBIC, wherein the vitamin C concentration at which the absorbance was just below the cutoff was set as MBIC. Since the MBIC value we determined was in between the serial dilutions that we have done (see results and discussion), we calculated it using the interpolation method. This method uses a discrete set of known data points, representing the absorbance values at the six 2-fold serial dilutions of vitamin C, to construct a new data point within their range, which was the MBIC in our case.
With regards to the detachment assay, the appearance of the wells at 18 and 48 hours was visually compared, and the readings were used to confirm our findings. It was thought that if vitamin C can detach previously formed biofilm, there will be a visual difference in appearance, although bacterial sedimentation can interfere with those findings. Therefore, to ascertain the presence of any difference between the biofilm before and after the addition of vitamin C, the readings from those wells were compared to our positive controls, which merely represent the same samples if vitamin C were not added. The method we used for detachment is similar to the setup used by Isela et al.  and embodies a simplified, yet less accurate method to the that demonstrated by Sanchez et al .
To establish the statistical significance of our findings, we compared two sets of results in a dependent observation using the two-sample (paired) t-test and estimated the range of difference between their means by measuring the confidence interval. We used SPSS software (statistical package for the social sciences, version 26, International Business Machines Corporation, Armonk, NY, USA) to carry-out five statistical tests, all of which were two-sample t-tests, and the significance level was determined at p < 0.05. All our data sets were determined to be normally distributed by the Shapiro-Wilk test. First, we compared the average absorbance value of the samples when not exposed to vitamin C (positive controls), and that of the same strains when exposed to a supra-MBIC yet infra-MBC (6.25 mg/ml) concentration of vitamin C. Second and third, we compared the average absorbance value of the positive controls with that of two sub-MBIC concentrations (around ½ and ¼ MBIC). Fourth and fifth, we compared the average absorbance value of the positive controls with that of the detachment assays at 25 and 12.5 mg/ml of vitamin C. To the best of our abilities, similar conditions were maintained across all comparisons.
The kinetic study is essentially a turbidimetric assay that allows the evaluation of microbial growth throughout incubation time. Selected samples were used in this study. Overnight cultured samples were conditioned on a microtiter plate with two supra-MBC concentrations and their absorbance was measured over a period of 24 hours. Six wells were used for each sample: the first being the positive control; the second and third containing bacteria/broth with vitamin C at 25 mg/ml and 12.5 mg/ml, respectively; the fourth and fifth containing bacteria/broth with gentamicin at 20 mg/ml and 10 mg/ml, respectively; and the sixth being the negative control. The plate was appropriately covered and incubated at 37 ºC and 5% CO2 between the readings. The results were then plotted on a graph of time against absorbance. In this study, it was expected that if growth occurs, the turbidity will increase and, hence, the absorbance values will gradually grow over time. On the contrary, if the agent, such as vitamin C or gentamicin, interacted with S. mutans sufficiently to alter its growth kinetics; an altered response, as epitomized by the turbidity and hence the absorbance after exposure, will be seen [28, 35].
Comparison with gentamicin and study of synergy
Selected samples were used to gauge the difference between the inhibitory capacity of vitamin C and gentamicin against S. mutans, and to study the presence of any synergistic effects between them. Samples were cultured on blood agar and the zones of inhibition of each were compared when alone and when combined together. As for studying any possible synergy or antagonism between them, the fractional inhibitory concentration index (FICI) was used. It can be calculated by the following equation: FICI = , where A and B are the two drugs tested alone, or in combination (AB). The MIC for both vitamin C and gentamicin was measured using the same strains under the same conditions. FICI was interpreted according to standard definitions, wherein a FICI of ≤ 0.5 was determined as a synergistic effect, a FICI of > 4.0 was determined as an antagonistic effect, and a FICI between 0.51 and 3.99 was determined as no interaction [33, 36].