Casein phosphopeptide acts synergistically with �uoride to inhibit initial adhesion of Streptococcus mutans to the saliva-coated hydroxyapatite disc

Background Recent preventive strategies for dental caries focus on targeting the mechanisms underlying bio�lm formation, including inhibiting bacterial adhesion. A promising approach to prevent bacterial adhesion is to modify the composition of acquired salivary pellicle. This in vitro study investigated the effect and possible underlying mechanism of pellicle modi�cation by casein phosphopeptide (CPP) on Streptococcus mutans (S. mutans) initial adhesion, and the impact of �uoride on the e�cacy of CPP. Methods The salivary pellicle-coated hydroxyapatite (s-HA) disc was modi�ed by 2.5% CPP or 2.5% CPP supplemented with 900 ppm �uoride solutions. After cultivation of S. mutans, the adherent bacteria were visualized by scanning electron microscopy (SEM) and quantitatively evaluated using the plate count method. Confocal laser scanning microscopy (CLSM) was used to evaluate the proportions of total and dead S. mutans. The concentrations of total, free, and bound calcium and �uoride in CPP and uoride-doped CPP solutions were determined. The water contact angle and zeta potential of s-HA with and without modi�cation were measured. The data were statistically analyzed using one-way ANOVA followed by a Turkey post hoc multiple comparison test. Results Initial adhesion of S. mutans to s-HA was inhibited in response to pellicle modi�cation by CPP and uoride-doped CPP, and the latter was more e�cient. CLSM analysis showed that the proportion of dead S. mutans did not differ between the groups. Water contact angle and zeta potential decreased after pellicle modi�cation, and both were lowest in the uoride-doped CPP group. Conclusions Pellicle modi�cation by CPP inhibited S. mutans initial adhesion to s-HA, possibly by reducing hydrophobicity and negative charge on the s-HA surface, and incorporating �uoride into CPP enhanced its anti-adhesion effect.


Background
Dental caries is a multi-factor lifestyle-related disease caused by the effect of metabolic end products of dental bio lm on the tooth surface [1].A dental bio lm is a community of microorganisms that are embedded in their extracellular polymeric substances [2].Bio lm formation is initiated by bacterial adhesion to the acquired salivary pellicle on the surface of dental hard tissues [3].Many microorganisms participate in the development of caries.Streptococcus mutans (S. mutans) is considered the primary etiologic agent because of its excellent acidogenicity, aciduricity and glucan synthesis ability [4].Pellicle formation starts immediately after toothbrushing through the selective adsorption of salivary components onto the tooth surface.The pellicle has a protective effect against dental caries by functioning as a semi-permeable barrier and a calcium and phosphate reservoir to inhibit demineralization and promote remineralization [5].However, speci c receptors in the pellicle, such as amylase and proline-rich proteins, can be recognized and bound by lectin-like bacterial adhesins, thereby facilitating bacterial adhesion [6].Conventional treatment of bio lm-related diseases aims at mechanically removing the bio lms.However, recent therapeutic strategies focus on targeting the mechanisms underlying bio lm formation, including inhibiting bacterial adhesion [7].Bacterial adhesion can be disrupted when the molecular composition and physicochemical properties of the pellicle are altered [8].Modi cation of the pellicle by plant extracts or ingredients of natural products, such as tannic acid and propolis, is a safe and cost-effective approach to prevent bacterial adhesion [9,10].
Casein phosphopeptide (CPP) are a subgroup of milk-derived phosphorylated peptides that can be naturally present in milk or obtained by proteolytic digestion of casein [11][12][13].CPP contains the cluster sequence -Ser(P)-Ser(P)-Ser(P)-Glu-Glu-, and this highly polar acidic domain has high binding a nity for divalent metal ions [14].CPP can e ciently bind and solubilize calcium, thereby enhancing the passive absorption of calcium in the gastrointestinal tract, improving remineralization of the tooth surface, and buffering the pH of dental bio lm [15].The inhibitory effect of CPP on dental enamel demineralization has been con rmed in vitro and in vivo [16,17].A randomized controlled trial showed that toothpaste containing 2% CPP or 1190 mg/kg uoride similarily reduced the incidence of caries [18].The cariespreventive activity of CPP is attributed to its ability to stabilize high levels of amorphous calcium phosphate (ACP) on the tooth surface [19], and the CPP-amorphous calcium phosphate (CPP-ACP) complex has been developed and patented by the Reynolds group [20].However, limited information is available on the effect of CPP on bacterial activity.0.5% CPP solution had no signi cant effect of on the established S. mutans bio lm [21].A previous study indicated that CPP inhibited S. mutans and Streptococcus sobrinus (S. sobrinus) initial adhesion to the pellicle-coated hydroxyapatite surface [22].
However, the exact anti-adhesion mechanism of CPP remains unknown.
Fluoride has been recognized as an anti-caries agent for more than seven decades; it can inhibit demineralization and promote remineralization of dental hard tissues [23].It also affects bacterial viability by acting as a glycolytic enzyme inhibitor as well as a transmembrane proton carrier, which leads to cytoplasmic acidi cation [24].It has been proposed that, incorporating uoride as a phosphatase inhibitor into the CPP-containing dental care products can enhance the effectiveness of CPP during application [25].Although several studies showed that incorporating uoride into CPP-ACP improves the remineralization [26,27], there is also evidence indicating that uoride supplementation did not improve the demineralization-suppressing effect of CPP or CPP-ACP [16, 19,28,29].The impact of uoride on the e cacy of CPP remains unclear.
The aim of this study was to investigate the effect and possible underlying mechanism of CPP on pellicle modi cation and S. mutans initial adhesion, as well as the impact of uoride on the e cacy of CPP.

Methods
Casein phosphopeptide and uoride CPP powder (Fuji lm Wako Pure Chemical Corporation, Osaka, Japan) and sodium uoride (Sigma, Saint Louis, MO, USA) used in this study were reagent-grade.

Test solutions
Two-fold serially diluted CPP solutions in the range of 0.15625-10% (w/v) and 2.5% CPP solution supplemented with 900 ppm uoride were prepared.The pH of solutions was adjusted to 7.0 using NaOH, and the solutions were sterilized through a 0.22 μm polyethersulfone membrane (Merck Millipore Ltd., Tullagreen, Carrigtwohill, Co. Cork, IRL).Heat-inactivated 2.5% CPP solution was prepared by autoclaving at 121°C for 15 min.All solutions were prepared freshly and used on the same day.

Hydroxyapatite disc
Hydroxyapatite (HA) disc (9.7 mm in diameter and 1.5 mm in thickness, Clarkson Chromatography Products, South Williamsport, PA, USA) was used as the substrate in this study.To visualize the live and dead bacteria cells on the substrate surface, the HA disc was polished using a precision lapping/polishing machine (Unipol-1502, Kejing Auto-Instrument Co., LTD, Shenyang, China) with #360, #600, and #1000 silicon carbide sandpapers under constant water cooling, until the disc reached a thickness of 0.3 mm.After polishing, the disc was sonicated in ultrapure water for 10 min to remove the debris.Before experiments, the HA disc was autoclaved at 121°C for 15 min.Bacterial strain and culture conditions S. mutans UA159 (ATCC 700610) purchased from Guangdong Microbial Culture Collection Center was used in this study.To prepare the inoculum, S. mutans was rstly recovered on a brain heart infusion (BHI, Difco, Detroit, MI, USA) agar plate supplemented with 5% sterile de brinated sheep blood for 48 h at 37 °C under anaerobic conditions (5% CO 2 , 10% H 2 , 85% N 2 ).A single colony was selected and inoculated into 10 mL of BHI broth and incubated anaerobically at 37 °C overnight.Bacteria was harvested by centrifugation (3000 rpm, 4 °C, 5 min), washed twice with sterile phosphate buffered saline (PBS), and nally re-suspended in BHI broth.The optical density at 600 nm (OD 600nm ) was adjusted to 0.2 (corresponding to a concentration of approximately 2.0 × 10 8 cells/mL).

Saliva collection and preparation
Human whole unstimulated saliva was collected from healthy volunteers with informed consent, and the procedure was approved by the Ethical Review Committee of Sun Yat-Sen University (No. ERC-[2017]-24).Saliva samples were collected and prepared according to a published method with some modi cation [30].Brie y, saliva was collected from eight volunteers who had not taken in anything but water for at least 2 h before saliva collection, were never smokers, and had not consumed any antibiotics or mouthwashes for at least 3 months.The saliva was spitted into a sterile and iced centrifuge tube until 10 mL of saliva was obtained, and this process usually took about 50-60 minutes.The saliva from eight donors was pooled.After centrifugation at 4000 rpm and 4 °C for 20 min, the supernatant was collected and ltered using a 0.22 μm polyethersulfone membrane.Clari ed saliva samples were frozen quickly in liquid nitrogen, and stored at -80 °C.Saliva collection from the same donors was carried out with an interval of at least 30 days, until the present study was completed.Saliva samples were used within 6 months and thawed at room temperature prior to experiments.

Adhesive inhibitory concentration assay
The HA discs were individually placed into a 24-well plate (Costar, Corning, NY, USA).A volume of 1 mL of the clari ed saliva was added into each well and incubated at 37 °C for 2 h.After rinsing twice with sterile PBS to obtain salivary pellicle-coated HA (s-HA) discs, the discs were randomly divided into eight groups (n = 6 per group): control group (PBS), and seven treated groups containing serially diluted concentrations of CPP.The s-HA disc was incubated in 1 mL of the corresponding solutions for 2 h at 37 °C, followed by two rinses with PBS.A volume of 1 mL of the bacteria suspension was seeded and incubated for 30 min and 2 h in a humidi ed atmosphere of 5% CO 2 at 37 °C (three discs per group and incubation time).After rinsing twice with PBS to remove non-attached and loosely bound bacteria, the HA disc was transferred into an eppendorf tube containing 2 mL of PBS.Adherent S. mutans was detached by sonication for 10 min, followed by vortexing for a further 60 seconds.The sonicated and vortexed S. mutans suspension was serially diluted and 100 μL of the diluted solution was spread over a BHI agar plate and incubated for 48 h at 37 °C in a humidi ed atmosphere of 5% CO 2 .The number of adherent S.
mutans was expressed as colony-forming units (CFU) per disc.The experiment was repeated three times in triplicate.The adhesion reduction percentage was calculated as follows: (CFU counts of the control group-CFU counts of the treated group)/ CFU counts of the control group× 100%.

Bacterial adhesion assay
CPP at 2.5% (w/v) was used for further experiments based on the ability to reduce the adhesion of S. mutans by approximately 50% [31].For the S. mutans adherence assay, s-HA disc was prepared as described above.The s-HA discs were randomly divided into four groups (n = 10 per group): (1) blank control (PBS); (2) negative control (heat-inactivated 2.5% CPP); (3) CPP (2.5% CPP); and (4) CPP + F (2.5% CPP supplemented with 900 ppm uoride).Pellicle modi cation and inoculation of S. mutans were performed as described above (three discs per group and incubation time).The number of adherent S. mutans was expressed as CFU per disc.The experiment was repeated three times in triplicate.
Scanning electron microscopy (SEM) was used to visualize S. mutans adhesion on the HA surface.Two samples from each group and incubation time were pre xed in 2.5% glutaraldehyde at room temperature for at least 3 h.After washing four to six times using ultrapure water, samples were dehydrated using gradient concentrations of ethanol (30%, 50%, 70%, 80%, 85%, 90%, 95%, and 100%) for 15 min, and then substituted by tert butyl alcohol three times, freeze-dried, sputter-coated with gold, and examined by SEM (JSM-6330F, JEOL, Japan).
Bacteria LIVE/DEAD staining Adhesion of S. mutans was also measured using a LIVE/DEAD BackLight Bacterial Viability Kit (L7012, Thermo Scienti c, USA).The polished HA discs with a thickness of 0.3 mm were used.Pellicle formation and modi cation as well as cultivation of S. mutans were performed as described above (three HA discs per group and incubation time).After incubation for 30 min and 2 h, the HA disc was washed twice with 0.9% NaCl and then stained with 1 mL of LIVE/DEAD ® BacLight™ solution at room temperature in the dark for 15 min.A volume of 1 mL of LIVE/DEAD ® BacLight™ solution contained 997 μL of ultrapure water, 1.5 μL of propidium iodide (PI), and 1.5 μL of SYTO 9, which was prepared according to the manufacturer's instructions.Dead bacterial cells with damaged membrane were nally stained in red color by PI, while live bacterial cells with intact cell membrane were stained in green color by SYTO 9.
Samples were observed by a confocal laser scanning microscopy (CLSM, LSM 780, Zeiss, Oberkochen, BW, Germany) with a 20× water-immersion objective lens.Dual-channel scanning observations were performed through a green channel for SYTO 9 (excitation wavelength: 488 nm) and a red channel for PI (excitation wavelength: 543 nm).Four elds were randomly selected on each sample for scanning.Image analysis was performed with ImageJ software.The ratio of the area occupied by green or red uorescence to the whole area on each visual eld was measured.For each specimen, the ratio (%) of the area covered by total uorescence of each color to the whole area was calculated.Subsequently, the proportion (%) of red to total uorescence was measured.The experiment was replicated three times.

Ion quanti ed analysis
The total and free calcium and uoride concentrations in 2.5% CPP and 2.5% CPP supplemented with 900 ppm uoride solutions were quanti ed, and the corresponding bound ion concentrations were calculated.Before detecting the total ion concentrations, 1 mL of the initial solution was diluted with 19 mL of 1.0 M HNO 3 and reacted for 24 h, followed by centrifugation at 1000g for 15 min at room temperature; the supernatant was collected to detect the total ion concentration [32].Before detection of free ion concentrations, macromolecular CPP with a molecular weight of approximately 3 kDa was ltered using a magnetically stirred ultra ltration device (Amicon, Model 8200, 200 mL, Millipore) equipped with an ultra ltration disc (1kDa NMWL, PLAC06210, Ultracel ® regenerated cellulose, Millipore).
According to the instructions, 50 mL of the initial solution was added into the device; equipped with a magnetic stirrer at 300 rpm, the ultra ltration separation was driven using high-purity N 2 (99.999%) and the pressure was maintained within 2.4 atm.CPP was ltered and the ltrate was collected to detect the free ion concentrations.Fluoride concentration was detected by ion chromatography (IC-1100, Thermo Fisher Scienti c, USA), and calcium concentration was measured using inductively coupled plasmaoptical emission spectroscopy (ICP-OES, ICP-OES 730, Agilent, USA).Calcium and uoride concentrations were expressed as mmol/L (mM).

Hydrophobicity determination
The hydrophobicity of the s-HA surface with and without modi cation was determined by measuring the water contact angle with a Contact Angle Meter (DMo-501, Kyowa Kogyo CO., LTD, Japan).Deionized water was used as the medium (2 μL per drop), and the sessile drop technique was employed.The right and left water contact angles for each droplet were measured at room temperature and averaged.For each sample, the measurement was repeated at three randomly selected regions.Finally, the result was expressed as degree (°).

Zeta potential analysis
The electrophoretic mobilities of pellicle coated ceramic hydroxyapatite powder (Clarkson Chromatography Products, Williamsport, PA, USA) with and without modi cation were measured by a Zeta Potential Analyzer (Zetasizer Nano ZS90; Malvern Instruments, Malvern, UK) according to a published method [33].The ceramic hydroxyapatite powder is the raw material for the HA disc that was used in the current study.Both the ceramic hydroxyapatite powder and HA disc were purchased from the same company.Pellicle formation and modi cation were carried out as described above in an incubator shaker (50 rpm) at 37 °C.Five measurements of electrophoretic mobilities for each sample were averaged, and the zeta potential was calculated according to the Helmholtz-Smoluchowski formula.

Statistical analysis
For all statistical analyses, SPSS v.19.0 software (IBM, Armonk, NY, USA) was used.All values were expressed as the mean ± standard deviation (SD).The inter-group differences were estimated by one-way analysis of variance (ANOVA) followed by a Turkey post hoc multiple comparison test.The level of signi cance was set at p < 0.05.

Effect of serially diluted CPP on S. mutans initial adhesion
As shown in Fig. 1, CPP inhibited the initial adhesion of S. mutans in a dose-dependent manner.At 30 min of incubation, compared with the control group, the reduction in the adhesion of S. mutans was 12.6%, 36.8%,49.6%, 76.2%, and 82.9% for CPP at concentrations of 0.63% to 10%, successively and respectively (0.63% CPP: p < 0.01; others: p < 0.0001).At 2 h of incubation, the adhesion of S. mutans was reduced by 32.9%, 51.5%, 78.5 %, and 83.8% for CPP at concentrations of 1.25% to 10%, successively and respectively (all: p < 0.0001).Although the adhesion reduction in the 0.63% CPP group was 8.9% at 2 h of incubation, the difference was not signi cant (p = 0.094).There were no signi cant differences between 0.16% CPP, 0.31% CPP, and the control group at 30 min or 2 h.

Effect of CPP and uoride-doped CPP on S. mutans initial adhesion to s-HA
The effect of 2.5% CPP and 2.5% CPP supplemented with 900 ppm uoride on the initial adhesion of S. mutans to s-HA was visualized by SEM (Fig. 2a) and quantitatively evaluated by the plate count method (Fig. 2b).
SEM images showed the distribution of S. mutans on HA discs with different coatings.The images showed a lower distribution of S. mutans in the CPP and CPP + F groups than that in the control groups, both at 30 min and 2 h, and the CPP + F group showed the lowest distribution.Quantitative analysis using the plate count method showed no signi cant difference between the blank control and the negative control at 30 min (p = 0.409) and 2 h (p = 0.877).At 30 min of incubation, the CFU counts were 50.8% and 74.0%lower in the CPP and CPP + F groups, respectively, than in the blank control (both: p < 0.0001).The amount of S. mutans in the CPP + F group was reduced by 47.2% compared with that in the CPP group (p < 0.0001).Similar results were obtained at 2 h of incubation: the CFU counts were 48.4% and 74.5% lower in the CPP and CPP + F groups, respectively, than in the blank control (both: p < 0.0001), and 50.6% lower in the CPP + F group than in the CPP group (p < 0.0001).

Live and dead analysis
Live and dead analyses were performed to con rm the anti-adhesion effect of CPP and uoride-doped CPP, and to estimate their antimicrobial properties.As shown in Fig. 3a, uorescence was lower in the CPP and CPP + F groups than that in the controls both at 30 min and 2 h.The ratio of the area occupied by total uorescence to the whole area for the blank control, negative control, CPP, and CPP + F groups was 11.19%, 11.17%, 5.43%, and 2.73% at 30 min of incubation, and 20.00%, 19.84%, 9.71%, and 4.59% at 2 h of incubation, respectively (Fig. 3b).The differences between groups were consistent with the results of the plate count method.The area (%) occupied by S. mutans in the CPP and CPP + F groups was 51.43% and 75.64% lower at 30 min and was 51.47% and 77.06% lower at 2 h than that in the blank control, respectively (all: p < 0.0001).The area occupied by S. mutans in the CPP + F group was 49.84% lower at 30 min and 52.43% lower at 2 h than that in the CPP group (both: p < 0.05).There was no signi cant difference between the blank and negative controls at 30 min (p = 0.873) or 2 h (p = 0.714).

Calcium and uoride concentrations in experimental solutions
The concentrations of total, free and bound calcium and uoride in 2.5% CPP and 2.5% CPP supplemented with 900 ppm uoride solutions are shown in Table 1

Water contact angle measurement
The hydrophobicity of the s-HA surface with and without modi cation was estimated by measuring the water contact angle.Representative images are shown in Fig. 4, and the results are shown in Table 2.
Water contact angle signi cantly decreased from 42.0° (blank control) to 24.1° (CPP) and 17.4° (CPP + F) (both: p < 0.0001), and was smaller in the CPP + F group than in the CPP group (p < 0.05).The water contact angle was higher in the negative control group (55.9°) than in the other groups (all: p < 0.0001).

Zeta analysis
All s-HA surfaces with or without modi cation exhibited negative zeta potential values, and the results are shown in Table 2.The zeta potential value for the blank control was -9.8 mV, which decreased to -11.5, -16.4,and -19.4 mV for the negative control, CPP, and CPP + F groups, respectively (all: p < 0.0001).The value for the CPP + F group was signi cantly lower than that for the CPP group (p < 0.0001).

Discussion
A promising approach to prevent dental caries is to reduce the abundance of S. mutans in dental bio lm by interfering with its initial adhesion [34,35].Pellicle modi cation is recognized as an alternative option for improving the anti-adhesion properties of the tooth surface [36].The present study provides evidence of the anti-adhesion effect of CPP and uoride-doped CPP on S. mutans.In addition, this is the rst study to show that the net negative charge and hydrophilicity of the s-HA surface increase after pellicle modi cation by CPP, and uoride further enhances the e cacy of CPP.
An in vitro model was established to simulate pellicle formation and modi cation as well as initial adhesion of S. mutans.We did not perform experiments on natural enamel because individual enamel specimens had different chemical compositions and unclear previous history [23].Hence, the HA disc, which is the major component of enamel, was used as the model substrate in the present study.Our results demonstrated that CPP inhibited the initial adhesion of S. mutans in a dose-dependent manner; a concentration of 2.5% CPP reduced the adhesion of S. mutans by approximately 50%, and this CPP concentration was selected for the current study [31].Fluoride is commonly added to various oral care products.To investigate the effect of uoride on the e cacy of CPP, a commonly used concentration of uoride (900 ppm) was selected for incorporation into CPP [23].
In the present study, the effect of pellicle modi cation by CPP and uoride-doped CPP on S. mutans initial adhesion was evaluated using the plate count method, SEM and live-dead staining.And the incubation times of 30 min and 2 h were chosen to better understand the nonspeci c and speci c interactions between S. mutans and the substrate [8].The results showed an apparent reduction of adherent S. mutans on the modi ed s-HA surface.The inhibitory effect was most obvious in the uoride-doped CPPmodi ed pellicle group.Heat-inactivated CPP was used as the negative control to con rm the effect of CPP in the active form.There was no difference between the blank control (non-modi ed pellicle) and negative control groups, indicating that denaturing CPP eliminated the anti-adhesion activity.As shown in the live-dead staining analysis, the proportion of red uorescence (dead S. mutans) to the total uorescence was similar among the four groups.This suggested that the anti-adhesion effect of CPP and uoride-doped CPP was non-bactericidal, namely S. mutans was prevented from attaching to the modi ed s-HA surface rather than being killed after adhesion.This non-bactericidal therapeutic approach targeting at bacterial adhesion can avoid the selective pressure on microorganisms, which may induce resistance to treatment or proliferation of opportunistic pathogens [37].
Calcium is commonly used as an aggregation-inducing agent during the production of CPP [25].In the present study, we performed ion quanti cation to measure the free and bound calcium and uoride concentrations in the CPP and uoride-doped CPP solutions.The results showed that the 2.5% CPP solution contained 30 mM calcium.Moreover, 29 mM calcium and 641 ppm uoride were bound by CPP in the 2.5% CPP supplemented with 900 ppm uoride solution.This suggests that CPP can bind and stabilize calcium and uoride to prevent the formation of CaF 2 precipitates.
CPP adsorption and bacterial adhesion to the pellicle are both driven by speci c and non-speci c interactions, such as van der Waals, electrostatic, and hydrophobic interactions [3,38].To investigate the effect of pellicle modi cation on surface properties, the zeta potential and water contact angle were measured to evaluate the charge and hydrophobicity of the s-HA surface before and after pellicle modi cation.Surface charge and hydrophobicity of the substrate are two important determinants that in uence bacterial adhesion [39].Decreased surface hydrophobicity is correlated with decreased bacterial adhesion [40], because initial bacterial adhesion involves the interaction between hydrophobic components on the surface of bacteria and those on the substrate [41].Increasing the net negative charge of the tooth surface reduces Streptococcal adhesion [42], because Streptococcal species (e.g., S. mutans) typically have a negatively charged surface [43].The present study showed that both zeta potential and water contact angle decreased after pellicle modi cation by CPP.This suggested that the formation of a more hydrophilic and negatively charged surface increased the repulsion between S.
mutans and the substrate.CPP is an amphiphiles and negatively charged peptide [25,44].An alternative mechanism mediating adherence of CPP to the pellicle is the localization of hydrophobic sites to the pellicle surface and the orientation of hydrophilic sites toward the outer environment [15].As shown in the present study, CPP increases the hydrophilicity of the s-HA surface.A previous study suggested that the calcium maximally bound by CPP is adequate to neutralize the intrinsic negative charge of CPP [44].Although 29 mM calcium was bound by CPP in this study, the zeta potential of s-HA decreased after pellicle modi cation by CPP.We assumed that the amount of calcium bound by CPP was not saturated, or calcium was partially released after CPP absorption onto the pellicle, or other anions bound by CPP partially eliminated the effect of calcium.This is the rst study demonstrating that CPP can increase the hydrophilicity and negative charge of the s-HA surface, and this synergistic effect partially led to the reduction of S. mutans initial adhesion.A recent study indicated that CPP selectively binds to certain pellicle components, including salivary acidic proline-rich phosphoprotein 1/2 precursor and mucin 7 [38], which are also binding sites for S. mutans [45].Therefore, it is reasonable to speculate that the number of binding sites for S.mutans decreased on the CPP-modi ed pellicle, which decreased the a nity of S.mutans for the CPP-modi ed s-HA surface.Although the caries-preventive activity of CPP is known to be mediated by its ability to stabilize high concentrations of ACP on the tooth surface, the anti-adhesion activity of CPP demonstrated in the present study may be partially responsible for its caries-preventive effect.
Our results showed that the anti-adhesion effect was most e cient in the uoride-doped CPP-modi ed pellicle group, which showed the lowest water contact angle and zeta potential values.Fluoride inhibits the adhesion of Streptococcus sanguis to the s-HA surface even at low concentrations [46].The considerable amount of free uoride (258.4 ppm) in the uoride-doped CPP solution may play a role in inhibiting S. mutans adhesion by further decreasing the negative charge and hydrophobicity of the s-HA surface [23,47].The bound uoride (641 ppm) may have played a role in the present study as well.On one hand, uoride may have increased the net negative charge of CPP, and uoride-doped CPP modi ed the s-HA surface.On the other hand, uoride may be partially released because of degradation of CPP by peptidases and/or phosphatases from S. mutans [48], and/or interaction between CPP and speci c pellicle components [38].The gradually released uoride could continually affect the s-HA surface.
Although the antimicrobial property of uoride against cariogenic bacteria has been documented [24], the anti-adhesion effect of uoride-doped CPP on S. mutans was non-bactericidal, as demonstrated in this study.The results suggested that the bound uoride molecules did not act on bacteria directly.Free uoride may have been incorporated onto the pellicle only during the pellicle modifying process and was washed away or diluted during subsequent steps.In general, the bound and free uoride might function synergistically to enhance the anti-adhesion effect of CPP.
The present study is limited in that a single bacterial strain was investigated, and the impact of other strains was not assessed.The effects on a greater number of bacterial species should be assessed in future studies.

Figures Figure 1
Figures

Figure 2 Effect
Figure 2

Figure 3 Live
Figure 3

Table 1
Concentrations of total, free, and bound calcium and fluoride in the experimental solutions.
CPP: 2.5% CPP solution; CPP + F: 2.5% CPP supplemented with 900 ppm uoride solution; Ca: Calcium; F: Fluoride.The concentrations of total and free calcium and uoride in solutions were measured, and the corresponding bound ion concentrations were calculated.Data are expressed as the mean ± SD.Percent of ion bound by CPP is indicated in parentheses.

Table 2
Water contact angle and zeta potential of the HA with different coatings.