Influence of caries activity and number of saliva donors: mineral and microbiological responses in a microcosm biofilm model

Abstract Objective this study evaluated the mineral and microbiological response of biofilms originating from different types of saliva inoculum with distinct levels of caries activity. Methodology the biofilms grown over enamel specimens originated from saliva collected from a single donor or five donors with two distinct levels of caries activity (caries-active and caries-free) or from pooling saliva from ten donors (five caries-active and five caries-free). The percentage surface hardness change (%SHC) and microbiological counts served as outcome variables. Results the caries activity of donors did not affect the %SHC values. Inoculum from five donors compared to a single donor showed higher %SHC values (p=0.019). Higher lactobacilli counts were observed when saliva from caries-active donors was used as the inoculum (p=0.017). Pooled saliva from both caries activity levels showed higher mutans streptococci counts (p<0.017). Conclusion Overall, pooled saliva increased the mineral response of the derived biofilms, but all the inoculum conditions formed cariogenic biofilms and caries lesions independently of caries activity.


Introduction
Dental caries is a sugar-dependent disease of polymicrobial origin and has been described as one of the most prevalent human diseases. [1][2][3] Carious lesions are formed as a consequence of complex interactions over time between an undisturbed microbial biofilm growth producing acid on the tooth surface due to a sucrose-rich diet. [4][5][6] The continuum acid production, creating a low pH environment, is able to drive the selection of cariogenic bacteria (any microbial species capable to survival in this acidic environment) leading to a disease state and the development of caries lesion. 4,7 The complexity of dental caries and the ethical issues related to its investigation in humans have led to the development of laboratory models to simulate the clinical condition under well-controlled circumstances. Several in vitro biofilm models have been employed to produce caries-like lesions. 8,9. Despite the high variability, the microcosm biofilms originating from saliva or dental plaque inoculum have a similar capability of producing caries-like lesions and reproducing more closely the complexity of microbial changes related to cariogenic biofilm development. 10,11 Saliva remains the most used inoculum source for microcosm biofilm formation due to its easy collection and handling. A recent systematic review on studies using biofilm models to develop dental caries showed that the number of donors and the caries activity profile from the donors (caries-active or caries-free) vary considerably among the studies when saliva is used as the inoculum source. 8 The use of inoculum by mixing saliva samples from different donors may lack inherent stability, leading to unrepresentative outcomes. 12 Moreover, when saliva inoculum from single donors is used, the biofilm variability decreases and inter-individual differences in caries-like lesions and biofilm development could be observed 13. When sucrose regimen is applied consistently on these microcosm models, the artificial caries lesions produced tends to develop in a similar pattern for all the caries activity profiles evaluated. 10,14,15 Azevedo, et al. 14 (2011) and Azevedo, et al. 15 (2014) Figure 1 shows an overview of the experimental design. Microcosm biofilms were formed individually on bovine enamel specimens (n=10 per condition) using saliva from a single donor or pooled saliva from five donors with two distinct caries activity profiles: caries-active and caries-free. Moreover, another condition consisted of pooling saliva from ten donors (five caries-active and five caries-free).

Experimental design
In 24-well plates, the biofilms were grown over the specimens using Defined Medium with Mucin (DMM) and subjected to cariogenic challenges (DMM plus 1%

Enamel specimen preparation
A minimum of 50 freshly extracted sound bovine incisors were prepared and used in this microcosm biofilm model. 16 Enamel-dentin discs were cut from the buccal surface from each incisor using a watercooled trephine drill. Both the dentin and the enamel surfaces were ground with 600, 1200 and 2500 grit SiC abrasive papers, respectively, to obtain plan-parallel surfaces. All procedures during disc preparation were performed under distilled water-cooling. The side and bottom surfaces of the discs were coated with nail varnish, leaving only the buccal surfaces exposed.
The baseline enamel surface hardness (SH1) was assessed by three indentations placed at the center of the enamel surface and spaced 100 μm apart using a Knoop diamond indenter loaded with a 50 g weight for 5 s (Micro Hardness Tester, FM 700, Future-Tech Corp., Tokyo, Japan). The baseline hardness of the selected enamel discs was 261.02±33.87 kgf mm 2 (Knoop hardness number). The maximum of % variability accepted for baseline enamel surface hardness to include the enamel specimens in the study was set at 22%.

Enamel specimen sterilization
All specimens were sterilized by gamma radiation in the Regional Center of Oncology/Radiotherapy Service (Faculty of Medicine, Pelotas, RS, Brazil). The specimens were kept moist in distilled water inside microtubes and placed 2 cm from the radiation source.
They were sterilized with gamma radiation 17 from a cobalt-60 source using particle energies of 1.25 MeV and subjected to 609.25 Gy/min. The total dose was 4.08 kGy. Saliva collection and inoculum saliva groups As a general requirement, volunteers should have good general health and not have undergone antibiotic therapy in the last 6 months. In total, ten donors were selected, five caries-free and five caries-active.
Volunteers included in the caries-active group should present at least two active caries lesions (cavitated or not), whereas caries-free volunteers should be free of any caries lesions. Fresh whole stimulated saliva by paraffin film chewing (6 ml) was collected from each volunteer that abstained from oral hygiene for 24 h and from food ingestion for 2 h prior to collection. No filter method was performed for the saliva collected.
Saliva samples were kept in ice during collection and inoculation procedures.
Immediately after saliva collection, 1-ml aliquot of saliva from each of the five caries-free volunteers (aged 20-25 years; mean age=22.6 years) was mixed and vortexed for 1 min to create the "cariesfree pooled saliva from five donors" group. The same process was performed with saliva collected from the five caries-active volunteers (aged 10-13 years; mean age=11.4 years) to create the "caries-active pooled saliva from five donors" group. One volunteer from the caries-free group and another from the cariesactive group were randomly selected to represent the caries-free (aged 23 years) and caries-active singledonor (aged 13 years) inoculum, respectively. A 4-ml aliquot of saliva was collected from each one of these volunteers. Finally, 1 ml of saliva from all volunteers selected was used and pooled in a falcon tube to create the pooled saliva group with both cariogenic profile conditions (caries-active and caries-free).

Saliva microbiological baseline measurements
After the collection procedure, an aliquot of each fresh whole saliva sample was dispersed by vortexing for 2 s, serially diluted (10 0 -10 −7 v/v) in sterile saline solution and cultivated in duplicate on the following culture media: blood agar enriched with 5% sheep/ horse blood (total microorganism counts), brain heart infusion agar adjusted to pH 4.8 by few hydrochloric acid drops (total aciduric counts), mitis salivarius agar supplemented with 0.2 U ml −1 of bacitracin (mutans streptococci) and Rogosa agar (lactobacilli). Mutans streptococci, total acidurics and lactobacilli were used as caries markers microorganisms. All agar plates were incubated under 5-10% CO 2 , <1% O 2 for 96 h at 37°C. The CFU were counted and expressed as CFU ml -1 and used as the baseline microbiological data of the inoculum (Table 1). supplemented with sucrose (DMM+S) and without (DMM) was followed. 10,14,15 The pH was daily recorded as a control procedure to monitor the experiment.

Microcosm biofilm model
The pH measurements were recorded from the supernatants (randomly from 3 wells of each group) immediately after the enamel discs were transferred to a new plate containing new medium. The supernatant pH of the DMM+S was 4.53±0.18, and 7.10±0.06 for the pure DMM.

Microbiological analysis
For biofilm microbiological analysis, the enamel specimens were individually removed from each well plate and the biofilms were collected from the enamel surface with a sterile microbrush and placed in preweighed microtubes. The wet weight of each biofilm sample was determined, and 1 ml of sterile saline solution was added to each microtube. The biofilms were dispersed by vortexing for 1 min and sonicated for 30 s at 20 w. The biofilm suspension was serially diluted (10 0 -10 7 ) and inoculated in duplicate onto the above-mentioned agar media. CFUs were counted by a trained operator blind to the study and the results were expressed as CFU mg -1 of biofilm (wet weight).

Measurements of %SHC
After biofilm collection, the enamel specimens were  Table 1 shows the baseline microbiological counts of the inoculum groups. In general, total acidurics and mutans streptococci baseline counts were lower in caries-free groups than those in carie-active donors.  According to the ecological plaque theory, when biofilms are exposed to ecological pressure, such as frequent sucrose exposure, specific bacteria are  , et al. 12 (2005) argued that inoculum by mixing saliva samples from different donors could lack inherent stability, leading to unrepresentative outcomes including less biofilm carcinogenicity when compared with single inoculum.
On the other hand, when more donors are used to compose one inoculum, microbial interactions may occur, such as antagonism, synergism or mutualism, and this relationship may lead to differences in the demineralization pattern compared to single-donor inoculum 22 and could explain the results of this study.
In addition, demineralization response for single donor inoculum showed higher variability when compared with a different saliva inoculum (e.g., 10 saliva donors). This finding could be explained by specific In this study, the presence of higher counts of total acidurics and mutans streptocci associated with pooled saliva inoculum compared to single-donor inoculum could also be explained by microbial interactions theory. In a simulated higher cariogenic challenge, as in this study (six hour of sucrose exposition during day), acidurics and acidogenic bacteria may be more able to proliferate in response to the ecological pressure compared to a more stable microcosm represented by single donors. This is shown by comparing the baseline microbial counts of total acidurics from the caries-free group (Table 1)  In general, microbiological counts were not affected by the caries activity level but lactobacilli were highly expressed when using saliva from a caries-active donor. This bacterial group is often found in plaque from caries-active people and represents a marker for caries activity. Although these bacteria were not expressed in higher counts in the baseline saliva of caries-active compared to caries-free groups, they could increase considerably when exposed to a sucrose regimen.
Notwithstanding that more demineralization for the pooled saliva groups (with five or ten donors) was noticed in this model, all inocula evaluated were able to produce caries-like lesions and cariogenic biofilms in this microplate microcosm biofilm model. That is, depending on the purpose of the study, the researchers could choose the group/inoculum that best fits their study by considering the speed of results, convenience of collection, operational costs and human resources involved. Although the microcosm biofilms were formed individually over each enamel samples, it is important consider that biofilm samples corresponding to the single donor came from one volunteer (caries free or caries active profile). In this study, the statistical unit was the biofilm formed over the samples individually. role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Declaration of interests
The authors declare no conflicts of interest with respect to the authorship and/or publication of this article.