Streptococcus mutans UA159 was used for colonization study and ELISA. Actinomyces naeslundii X600 was used for ELISA as control oral bacteria. All bacteria were grown in an atmosphere of H₂ and CO₂ (GasPack, Becton/Dickinson, Sparks, MD) in Brain Heart Infusion broth (BHI, Difco Laboratory, Detroit, MI) at 37°C.

NOD/LtJ mice naturally develop IDDM, SS, and dry mouth; and were the parent strain to develop NOD/SCID.e2f1^−/− mice. They were used as the control to compare S. mutans susceptibility to NOD back ground E2F-1^−/− mice (NOD.e2f1^−/−) and NOD/SCID back ground E2F-1 heterogeneous (NOD/SCID.e2f1^+/−) and homogeneous deficient NOD/SCID mice (NOD/SCID.e2f1^−/−) [33]. NOD/SCID mice were the parental lines to produce NOD/SCID.e2f1^−/− mice [33] and were used as control mice in bacterial inoculation experiments. Heterozygous NOD/SCID.e2f1^+/− mice were bred to generate NOD/SCID.e2f1^−/− mice. Three types (+/+, +/− and −/− of e2f1) of NOD/SCID mice were screened using PCR [33]. All strains were female, 4 months of age and were maintained in accordance with the guidelines for the Care and Use of Laboratory Animals from the National Institute of Infectious Diseases. Experimental protocols (#209125, 210110, and 21124) were approved by the National Institute of Infectious Diseases Animal Resource Committee.

Saliva samples were collected from volunteers with good oral health, after stimulation by chewing paraffin gum. The volunteers refrained from eating, drinking, and brushing for at least 2 h prior to collection. The saliva was placed into ice-chilled sterile bottles for 5 min; then centrifuged at 10,000 g for 10 min to remove cellular debris. For the inoculation assay and the enzyme-linked immunosorbent assay (ELISA), the clarified saliva was used after filter sterilization through a 0.22 µm Acrodisc filter (Pall Corporation, Ann Arbor, MI). After filtration, they were pooled and stored at −20°C until used.

Lyophilized secretory Immunoglobulin A (sIgA) from human colostrum, α-amylase from human saliva, and mucin from bovine submaxillary glands (Sigma-Aldrich, St. Louis, MO) were mixed in PBS and adjusted to similar physiological concentrations as in human-saliva: 0.25, 0.4 and 2.7 mg/ml, respectively. These reagents were stored at −20°C until used.

Bacterial inoculation, sampling and CFU estimates were performed using procedures and conditions described previously [31], [34], [35]. All oral streptococci were cultured in BHI broth overnight and then washed twice with sterile phosphate-buffered saline (PBS). Our previous study demonstrated that colony counts of S. mutans were significantly higher than that of other streptococci (i.e. S. sanguis, S. sobrinus, and S. salivarius) in mice that ingested 1% sucrose in water one day before inoculation [31]. Thus, mice were given drinking water containing 1% sucrose (less than the usual concentration in juice) one day prior to S. mutans inoculation to reproduce the early adherence of S. mutans in conditions resembling a natural state. Chlorhexidine (0.2%) soaked sterile cotton swabs were used to disinfect the oral cavities of the mice including the maxillary incisor teeth. The cavity was immediately washed with sterile PBS. Four or 6 mice were treated with 100 µl of human saliva or salivary components for 2.5 min with the aid of micropipette. Casein was used as a control as a non-salivary component for the treatment. Five min after treatment, mice were washed with 100 µl of PBS. S. mutans solutions were introduced to the oral cavities of all females at 4 months of age at a final concentration of 7×10⁹ CFU in 250 µl of PBS during 2.5 min. Mice were separated into four groups based on the feeding conditions 24 h after inoculation. During the 24 h, one group was fed food with distilled water compared to another fed food with 1% sucrose-water; and the other set was food-deprived with 1% sucrose water or distilled water. Following inoculation, samples were collected from the labial surfaces of the maxillary incisor teeth with a sterile cotton ball and then dipped in 2 ml of PBS. To evaluate NOD/SCID.e2f1^−/− mice as compared with previous results and to obtain stable data, samples collected from incisors were tested as parameters used in previous studies [31], [35]. The samples in sterile PBS were sonicated using ultrasonic dispersion (power output, 60 W) for 10 s, and then poured onto Mitis-Salivarius agar plates containing 0.02 M bacitracin (MSB). CFUs were determined by counting rough-surface colonies on MSB plates after 48 h using anaerobic incubation at 37°C.

To determine if sIgA reacts with S. mutans in vitro and if sIgA is absorbed on the tooth surface after treatment with human saliva, ELISA was performed with some modifications as described previously [33]. 96-well microtiter H-plates (Sumitomo Bakelite, Tokyo, Japan) were coated overnight at 4°C with a culture of S. mutans or A. naeslundii (1 µl/ml) in Na₂CO₃ coating buffer at pH 9.6 and incubated at 4°C overnight. In the sandwich assay to detect absorbed sIgA, we used 1/1,000 mouse anti-human immunoglobulin A antibody (Sigma-Aldrich, St. Louis, MO). The bacteria and antibody were diluted in Na₂CO₃ coating buffer at pH 9.6 and incubated at 4°C overnight. The plates were washed with PBS containing 0.1% (v/v) Tween 20 (PBST); and blocked with 1% (wt/vol) skim milk in PBST for 1 h at 37°C. Excess skim milk was removed by washing three times with PBST. To determine the presence of sIgA on the tooth surface, the tooth surface was swabbed using a sterile cotton ball after treatment with human saliva, and the swabbed ball was soaked in 2 ml coating buffer and shaken for 1 min. A 100 µl aliquot of 0.25 mg/ml sIgA, human saliva, or the soaked sample was added to the wells and the plates were incubated for 1 h at 37°C. The wells were washed three times with PBST; and further incubated for 1 h at 37°C with 100 µl 1/1,000 alkaline phosphatase conjugated goat anti-human immunoglobulin A antibodies (Zymed Laboratories, South San Francisco, CA). After three washings with PBST, the bound antibodies were detected after the addition of 50 µl of 3 mg/ml para-nitrophenyl phosphate as a substrate and incubated for 30 min at 37°C. Absorbance at 405 nm (A₄₀₅) was measured using a microplate reader (Multiskan Bichromatic; Laboratory Japan, Tokyo, Japan). The mean value for each sample was used to calculate the ELISA value: Abs₄₀₅ × 100/t (t: time of reaction). Triplicate measurements were performed and means calculated with standard error.

To determine if specific sIgA is employed for S. mutans colonization on the tooth surface, an absorption procedure was performed to remove specific antibody against S. mutans. Solutions of 1 mg/ml sIgA in PBS were absorbed with 0.5 mg (dry weight)/ml whole cells of lyophilized S. mutans UA159 at 37°C for 1 h and then overnight at 4°C. The mixture was centrifuged at 8,000 rpm for 10 min to remove S. mutans-IgA complex. Protein concentrations in the sIgA sample were measured using the Bio-Rad Protein Assay kit (Bio-Rad Laboratory, Hercules, CA) based on the method of Bradford and measured at 595 nm. The concentration of sIgA was adjusted to 0.25 mg/ml after the absorption procedure.

To determine if the animal model could be used for the analysis of inhibitors for colonization and biofilm formation of S. mutans on the tooth surface, fructanase (FruA), a candidate inhibitor for biofilm formation of S. mutans [36], was used in the in vivo assay. The inhibiting activity of FruA at 1.25 units/ml was assayed in 96 well microtiter plates coated with human saliva [36]. FruA at 1.25 units/ml was also added within a 1% sucrose solution in drinking water (DW). FruA does not digest sucrose at 20∼25°C in 1% sucrose drinking water and does at 37°C in the oral cavity after mice drink the water [36]. After pre-treatment of sIgA following bacterial inoculation, all NOD/SCID.e2f1^−/− mice were fed and supplied 1% sucrose water containing or not containing FruA. After 24 h inoculation, samples were collected and the CFU was counted as described above.

The CFU and ELISA data were expressed as means ± standard deviations. GraphPad Prism version 5.0 d for Mac OS X (GraphPad Software, San Diego, CA) was used to perform tests of significance. The statistical significance of differences between two groups was determined using the unpaired t-test. For comparison between multiple groups, one-way analysis of variance (ANOVA) and Tukey-Kramer tests were used. P-values less than 0.001, 0.01 or 0.05 were considered statistically significant using two-tailed comparisons. All experiments were repeated and analyzed independently.

Human saliva is thought to play a significant role in the attachment of S. mutans to the tooth surface. We evaluated human saliva in bacterial colonization of NOD/SCID wild type, NOD/SCID.e2f1^+/− mice, and NOD/SCID.e2f1^−/− mice. S. mutans colonization in each mouse was significantly increased at all time points after the inoculation when they were treated with human saliva (Fig. 1 A, B and C). Bacterial numbers on the tooth surfaces were significantly higher in NOD/SCID.e2f1^−/− mice than those in NOD/SCID wild type or NOD/SCID.e2f1^+/− mice after 90 and 120–180 min post inoculation (Fig. 1 D). Colony numbers of S. mutans gradually decreased from 30 min to 90 min; however, after the colonization phase, the CFU gradually increased from 90 to 180 min in human saliva-treated NOD/SCID.e2f1^−/− mice; whereas the other mice did not show a difference comparing time points.

To determine if salivary components induce colonization of S. mutans on the tooth surface using the in vivo model, α-amylase, mucin and sIgA, receptors for S. mutans adhesins, were used to treat teeth before bacterial inoculation. CFUs were lower within non-treated mice compared to NOD/SCID.e2f1^+/− and ^−/− 18 mice treated with all components other than casein treatment (control; non-salivary component) in NOD/SCID.e2f1^+/+ mice (data not shown). NOD/SCID.e2f1^−/− mice had a higher colonization than NOD/SCID.e2f1^+/− and NOD/SCID.e2f1^+/+ in each pre-treatment using the salivary components (Fig. 2 A, B and C). Bacterial colonization on teeth treated with 0.25 mg/ml sIgA at physiological concentrations was increased significantly in NOD/SCID.e2f1^−/− mice (13,992±6,423); however, there was no significant difference in treating with saliva compared to sIgA (Fig. 2 C). In NOD/SCID.e2f1^+/+ and NOD/SCID.e2f1^+/− mice, treatment with 0.25 mg/ml sIgA did not show greater colonization (Fig. 2 A, B). Further, higher concentrations of sIgA (0.4 mg/ml) did not result in higher colonization by S. mutans in comparison with BSA and casein in NOD/SCID.e2f1^+/− and NOD/SCID.e2f1^−/− mice (Fig. 2 B and C). Treatment in NOD/SCID.e2f1^−/− mice with mucin (at 0.4 and 2.7 mg/ml) or with BSA did not result in increased levels of S. mutans colonization; these pre-treatments yielded significantly lower CFU counts compared to treatment with 0.25 mg/ml sIgA and considerably higher counts compared to treatment with 0.4 mg/ml amylase. Treatment with amylase at 0.1 mg/ml showed significantly higher colonization than at 0.4 mg/ml in NOD/SCID.e2f1^+/−; whereas there was no significant difference using NOD/SCID.e2f1^−/− mice.

SIgA was taken from human colostrum, and therefore may include various antibodies to pathogens. To confirm whether sIgA reacts with S. mutans, ELISA was performed using S. mutans-coated 96 well microtiter plates. A. naeslunidii was also used for coating as another oral bacterium. 0.25 mg/ml sIgA reacted strongly with S. mutans but not A. naeslundii (Fig. 3 A). The specificity of sIgA was observed by absorption of specific antibody to S. mutans in pre-incubation using S. mutans whole cells within sIgA. The absorbed sIgA was used for the ELISA assay and showed no significant reactivity to S. mutans (Fig. 3 A). Using human saliva, specific antibody to S. mutans was also observed using the ELISA assay (Fig. 3 A). The 0.25 mg/ml absorbed sIgA was used for the colonization assay in NOD/SCID wild type, NOD/SCID.e2f1^+/− and NOD/SCID.e2f1^−/− mice, and the effect of absorbed sIgA was compared with 0.25 mg/ml non-absorbed sIgA in all mice. The absorbed sIgA did not increase colonization of S. mutans in comparison with non-absorbed sIgA at 120 min after inoculation of S. mutans (Fig. 3 B). Therefore, increased colonization of S. mutans was dependent on specific antibody to S. mutans in sIgA and human saliva using this animal model. To determine whether sIgA remained on the tooth surface after treatment with human saliva, the surface was swabbed using a sterilized cotton ball at 120 min after the treatment in mice; and sIgA in the swabbed sample was measured using ELISA. The level of human-IgA that remained on the teeth for 120 min was significantly higher in NOD/SCID.e2f1^−/− mice as compared to the other two strains (Fig. 3 C). This shows that specific sIgA antibody to S. mutans remains on the tooth surface after treatments with sIgA and human saliva in mice having decreased saliva and lack of IgA and IgG, the NOD/SCID.e2f1^−/− mice. To determine whether a lack of IgA, by inserting the SCID type in NOD.e2f1^−/− mice, promoted the colonization of S. mutans, the parent strain (NOD.e2f1^−/− mice) and previous the parent strain (NOD mice) to NOD.e2f1^−/− mice were used for the colonization assay after pre-treatment with 0.25 mg/ml sIgA and compared with NOD/SCID.e2f1^−/− mice. We found that the colonization at 120 min after inoculation was significantly lower in NOD and NOD.e2f1^−/− mice than NOD/SCID.e2f1^−/− mice (Fig. 3 D). Therefore, lack of IgA and decreased saliva allowed specific IgA to remain on the tooth surface and to promote colonization of S. mutans in NOD/SCID.e2f1^−/− mice.

Long-term colonization is necessary in a mouse model to study several agents for the prevention to oral diseases. We observed that after inoculation, the colonization of S. mutans was slight at 24 hours in NOD/SCID, NOD/SCID.e2f1 ^+/− and NOD/SCID.e2f1^−/− mice pre-treated with human saliva (Fig. 4 A). Drinking water and diet including sucrose helped biofilm formation in other studies [14], [31]. A low concentration of 1% sucrose water was selected and supplied as drinking water with the usual animal diet for mice to establish an animal model that avoided high sucrose concentration-dependent colonization. The significant colonization was not observed in only the 1% sucrose water group as compared to that in non-sucrose water and non-diet group (Fig. 4 A, B). However, the group supplied with the combination of 1% sucrose-water and diet showed the most CFU/ml of S. mutans; colony numbers in NOD/SCID.e2f1^−/− (693±500 CFU/ml) and in NOD/SCID.e2f1^+/− (193±190) were significantly higher than those in wild type mice (17±32) (Fig. 4 D). The colonization was significantly higher in 1% sucrose-water and diet than non-sucrose water and diet in NOD/SCID.e2f1^−/− mice (Fig. 4 C, D).

In our previous report, purified and commercial fructanase (FruA) from Aspergillus niger completely inhibited S. mutans GS-5 biofilm formation on saliva-coated polystyrene and hydroxyapatite surfaces [36]. Therefore, we examined inhibition using FruA in S. mutans colonization in the established mouse model system. The bacterial load in NOD/SCID.e2f1^−/− mice pre-treated with sIgA and supplied sucrose-water containing FruA (13±20 CFU/ml) decreased as compared to that without FruA (104±159); however, there was no significant difference (P = 0.088).