Introduction

Teeth consist of a mineral matrix composed mainly of calcium phosphate and serve as sites where diverse microorganisms in the oral cavity can easily reside. While teeth themselves are not nutritional sources for microbes, their surfaces are coated with various substances derived from saliva and food. The bacteria inhabiting the oral cavity are predominantly facultative anaerobes, such as Streptococcus mutans and Lactobacillus species, along with a small number of aerobes. S. mutans, first isolated from the tooth surface by Clarke [1] in 1924, is a Gram-positive facultative anaerobe further classified into S. mutans, Streptococcus sobrinus, Streptococcus ferus, and Streptococcus rattus. Experimental studies by Gibbons et al. [2] demonstrated that these strains can induce dental caries in animals.

When a clean tooth surface is exposed to saliva, salivary glycoproteins coat the enamel surface to form an acellular protein film known as the acquired pellicle [3]. S. mutans in the oral cavity attaches to the pellicle through pellicle-binding proteins. Once attached, S. mutans secretes glucosyltransferase, which reacts with exogenous sucrose to synthesize insoluble glucose polymers such as mutan and dextran. S. mutans utilizes mutan to adhere firmly to the tooth surface, enhancing the cohesion between bacterial cells growing on the enamel. As bacterial adhesion to the pellicle proceeds, additional salivary proteins and extracellular polysaccharides accumulate, leading to the maturation of dental plaque.

Dental caries is, therefore, an infectious disease caused by bacteria, but its progression is influenced by the types of food consumed and the frequency of intake. Meanwhile, the intrinsic resistance of tooth structure to demineralization and the cleansing action of saliva from salivary glands play crucial preventive roles. Although the resistance of enamel and the washing action of saliva are thought to inhibit the early stages of caries formation, the occurrence of the disease suggests that multiple factors interact synergistically to promote caries development.

Traditionally, mechanical removal of plaque by toothbrushing has been the most fundamental method for caries prevention. However, brushing may be difficult for young children and individuals with physical disabilities, and it cannot effectively eliminate bacteria residing in pits and fissures. Consequently, various supplementary preventive approaches have been explored, including fluoride application, immunization, antibiotic therapy, and the use of natural extracts.

Among these, fluoride therapy is the most widely employed. Fluoride not only inhibits demineralization and promotes remineralization but also exerts antimicrobial activity, thus demonstrating significant preventive effects against dental caries [4-6]. However, epidemiological studies have reported that excessive fluoride exposure may lead to dental fluorosis, characterized by white enamel spots, raising ongoing debate over the optimal concentration of fluoride for clinical use [7,8].

Because dental caries is a bacterial disease, vaccination strategies targeting the causative organism have also been investigated. Experimental studies demonstrated that immunization with S. mutans or its glucosyltransferase reduces caries incidence in animals, indicating the potential for immunological prevention. Nevertheless, since S. mutans shares antigenic structures with human cardiac tissues, antibodies produced against the bacterium may cross-react with heart tissue, posing risks of autoimmune damage. Thus, further refinement is required before immunization can be safely applied in humans [9,10].

Although antibiotic therapy has been tested for caries prevention, the potential harm to the host limits its use. Recently, attention has shifted toward naturally derived substances that inhibit bacterial growth without adverse side effects. Various natural products, including traditional herbal medicines, green tea, oolong tea, bamboo salt, and grapefruit seed extract, have been reported to exhibit non-specific antibacterial activity with potential anti-cariogenic effects [11,12].

Recently, it has become possible to isolate and purify lytic enzymes produced by soil microorganisms. These enzymes target bacterial cell walls―structures absent in human cells― making them less likely to cause adverse effects compared with fluoride, immunotherapy, or antibiotics [13]. Preliminary analyses revealed that a particular lytic enzyme, designated L27, exhibited the strongest antibacterial activity against S. mutans, while showing more than 90% activity against Streptococcus sanguinis and Streptococcus salivarius, and less than 30% activity against Lactobacillus species. This finding suggests that enzyme L27 acts selectively against bacteria primarily responsible for dental caries.

Accordingly, this study aimed to evaluate the anti-cariogenic effects of the lytic enzyme L27―isolated from soil samples collected throughout Korea―by assessing its ability to inhibit bacterial adhesion to enamel surfaces, modulate pH and calcium ion release in an artificial oral environment, and influence the surface hardness of artificially induced enamel lesions.

Materials and Methods

1. Bacterial culture and preparation of lytic enzyme

S. mutans American Type Culture Collection 25175 was obtained from the Genetic Engineering Research Center at the Korea Advanced Institute of Science and Technology. The strain was inoculated into brain heart infusion broth and subcultured at 37℃ under 5% CO₂ in an incubator before use in experiments.

From soil samples collected across various regions in Korea, approximately 1,500 microorganisms were isolated using the plate dilution method on a neutral bacterial medium. These isolates were inoculated onto agar plates containing a suspension of S. mutans . Colonies that formed distinct transparent halos were selected as candidates producing lytic enzymes. Among them, the strain exhibiting the highest activity was identified as Bacillus licheniformis YL-1005 and its protein was designated as L27. The culture supernatant containing this enzyme was separated and purified using methanol and subsequently employed in the present study.

2. Preparation of saliva

Stimulated saliva was first collected from healthy male volunteers in their twenties and thirties by chewing paraffin wax. The saliva was cooled on ice, pooled, and heated at 60℃ for 30 minutes to inactivate enzymes and sterilize the sample, then stored at –20℃ until use.

3. Radiolabeling of S. mutans with [³H]-thymidine

S. mutans was inoculated into the culture medium containing [³H]-thymidine (specific activity, 20 Ci/mmol) at a final concentration of 2 μCi per mL. The culture was incubated anaerobically for 24 hours in a gas-pack jar. The labeled bacteria were washed three times with 2 mM potassium phosphate buffer (pH 6.0) containing 5 mM KCl and 1 mM CaCl₂, and suspended in KCl buffer to an optical density (OD) of 0.5 at 660 nm (approximately 9 × 10⁹ cells/mL). One milliliter aliquots were used for the adhesion assays.

4. Evaluation of cytotoxicity of the lytic enzyme

The MTT assay was used to assess cell viability. In this assay, mitochondrial dehydrogenase in viable cells reduces MTT to blue formazan crystals, which are then dissolved and quantified spectrophotometrically to determine the number of living cells.

Gingival fibroblasts were suspended at a concentration of 5 × 10³ cells per 0.16 mL of culture medium and seeded into each well (0.16 mL per well). After 24 hours of incubation at 37℃ under 5% CO₂ to allow cell attachment, the lytic enzyme was added to the wells. Following incubation of 24 hours, 0.05 mL of MTT solution (5 mg/mL) was added to each well and allowed to react for 4 hours. The medium was then discarded, and 0.05 mL of dimethyl sulfoxide was added to dissolve the formazan crystals. Absorbance was measured at 570 nm using an enzyme-linked immunosorbent assay reader, and cytotoxicity was evaluated based on the OD values.

5. Measurement of Ca²⁺ concentration and pH in the artificial mouth model

Because the oral cavity is a semi-anaerobic environment that continuously receives saliva and nutrients, the artificial mouth model was designed to allow continuous gas and fluid flow (artificial saliva, sucrose-containing nutrients, and the lytic enzyme). The model was maintained in a thermostatic water bath at 37℃. Artificial saliva solution was prepared with the composition of 5.5 mM NaH₂PO₄, 2.5 mM NaHCO₃, 20.7 mM KCl, 0.3 mM MgCl₂, and 0.1 mM CaCl₂.

To measure Ca²⁺ release and pH changes, spheroidal hydroxyapatite (HA) beads (0.08–0.20 mm in diameter; Merck) were used in the artificial mouth model. The concentrations of Ca²⁺ and pH were continuously recorded using ion-sensitive electrodes (Ca²⁺ electrode and pH electrode; ORION model 920A) connected to a physiograph (Biopac converter) and a personal computer.

Because the measured values were expressed in millivolts (mV), they were converted to Ca²⁺ concentrations (mM) using a calibration curve. Standard solutions containing 0.1, 0.5, 1, 2, 5, and 10 mM Ca²⁺ in artificial saliva were prepared, and a linear regression equation was derived to calculate Ca²⁺ concentration from the recorded mV values.

6. Effect of the lytic enzyme on demineralization induced by S. mutans

S. mutans cells were centrifuged at 5,000 rpm for 15 minutes and washed twice with artificial saliva. In the oral model, 30 mL of artificial saliva, 30 mg of HA beads, and the lytic enzyme were combined and allowed to equilibrate for 20 minutes. After equilibrium was established, S. mutans was introduced into the chamber. Ca²⁺ release was first measured in the presence of S. mutans alone, then compared with that observed after addition of the lytic enzyme.

7. Adhesion assay of S. mutans to HA beads

Thirty milligrams of HA beads were washed five times with distilled water to remove fine particles and suspended in 800 μL of artificial saliva. Then, 100 μL of radiolabeled S. mutans suspension and 100 μL of the lytic enzyme solution were added. The mixture was incubated at 37℃ with gentle rotation (10 rpm) to allow bacterial adhesion. After incubation, the HA beads were washed three times with KCl buffer, dried overnight at 37℃, and transferred to 5 mL of scintillation fluid. The radioactivity (cpm) was measured to determine the effect of the lytic enzyme on S. mutans adhesion.

Results

1. Cytotoxicity of the lytic enzyme

To determine the cytotoxic potential of the lytic enzyme toward human gingival fibroblasts, cells were cultured in media containing various concentrations of the enzyme and analyzed by the MTT assay. As shown in Fig. 1, OD values remained nearly unchanged with increasing enzyme concentrations, indicating minimal cytotoxicity. However, at concentrations exceeding 250 μg/mL, the proliferation of gingival fibroblasts was slightly inhibited. Therefore, subsequent experiments were conducted using enzyme concentrations below this threshold.

2. Effect of the lytic enzyme on pH changes induced by S. mutans

When 250 μg/mL of lytic enzyme L27 was added to artificial saliva containing HA beads, no immediate change in pH was observed. However, after approximately 10 minutes, the pH began to decrease sharply and then declined more gradually after 40 minutes. The overall pattern of pH reduction was similar to that in samples without enzyme addition.

In contrast, in the absence of S. mutans, the pH remained nearly constant throughout the experiment. When 5 mM NaF (positive control) was added, a slight pH decrease was observed during the first 20 minutes, after which the pH stabilized and remained constant (Fig. 2).

3. Effect of the lytic enzyme on Ca²⁺ release induced by S. mutans

When 250 μg/mL of lytic enzyme L27 was introduced into the artificial saliva containing HA beads, Ca²⁺ release was negligible during the first 20 minutes but gradually increased over time, continuing up to 100 minutes. In the absence of the enzyme, Ca²⁺ release was initially similar for the first 20 minutes but subsequently increased sharply. Thus, compared with the control, the enzyme-treated group showed a marked suppression of Ca²⁺ release (Fig. 3).

4. Effect of the lytic enzyme on the adhesion of S. mutans to HA beads

To evaluate the influence of the lytic enzyme on the adherence of S. mutans to HA beads, bacterial cells labeled with [³H]-thymidine were incubated with HA beads in the presence of varying enzyme concentrations. As shown in Fig. 4, bacterial adhesion decreased sharply at enzyme concentrations below 10 μg/mL and continued to decline up to 100 μg/mL. At 50 μg/mL and 100 μg/mL, adhesion of S. mutans to HA beads was inhibited by 74 and 80%, respectively. These results indicate that the lytic enzyme significantly reduces bacterial adherence in a dose-dependent manner.

Discussion

Dental caries, a progressive infectious disease involving the destruction of hard tissues, is a multifactorial disorder resulting from the complex interaction among dental plaque bacteria, dietary carbohydrates, and saliva. Among the bacterial species present in dental plaque, S. mutans is recognized as the primary etiological agent. This microorganism attaches to the tooth surface, proliferates, and synthesizes extracellular polysaccharides via glucosyltransferase, producing insoluble glucose polymers such as mutan from sucrose. These polymers enhance bacterial cohesion, allowing S. mutans to form localized, adherent bacterial plaques on specific tooth sites, which initiate the carious process.

Cariogenic plaques contain a large number of bacteria (approximately 2 × 10⁸), and their metabolic activity results in the hydrolysis of sucrose into glucose and fructose, thereby lowering the pH below 5.5―the critical level for enamel demineralization. Repeated cycles of acid production led to microstructural breakdown of the enamel and formation of carious lesions.

Although the development of dental caries is strongly associated with bacterial metabolism, it is also influenced by host factors such as the intrinsic resistance of enamel to acid dissolution and the protective actions of saliva. Saliva not only buffers acids but also removes food debris and bacteria, thereby playing an important role in preventing the progression of dental caries.

In the present study, when S. mutans was introduced into artificial saliva containing HA beads, a gradual decrease in pH and a concurrent increase in Ca²⁺ concentration was observed, confirming that microbial acid production and enamel demineralization are closely related. However, although the rate of HA demineralization increased over time, the pH decreases gradually slowed, indicating that the relationship between pH decline and demineralization rate was not perfectly proportional. This discrepancy may be attributed to the neutralizing effect of Ca(OH)₂ released during demineralization, which partially counteracts the acid produced by S. mutans.

Many studies have sought to prevent or suppress dental caries. Since caries development depends on the interaction among host, microbial, and environmental factors, prevention can be achieved by eliminating or modifying one or more of these factors. Clinically, fluoride supplementation and dietary control―such as reducing intake of fermentable carbohydrates ―are widely used methods. In recent years, research has increasingly focused on natural products with minimal side effects and prolonged activity.

Attempts to suppress S. mutans as a means of preventing caries have been numerous. However, it has been difficult to maintain a sustained reduction of oral bacterial populations in vivo, and long-term equilibrium among oral microorganisms is challenging to preserve. Moreover, because caries progression occurs from the external enamel surface inward through acid erosion, relying solely on microbial control, antimicrobial agents, or immunological mechanisms to prevent lesion formation may be unrealistic. Consequently, efforts to eliminate cariogenic microorganisms have shown limited clinical success.

During our investigations into more direct approaches to caries control, we confirmed that certain soil-dwelling Bacillus species secrete substances capable of lysing S. mutans. Among these, lytic enzymes that attack bacterial cell walls have attracted particular attention [14,15]. Such enzymes may reduce bacterial adhesion, inhibit the formation of new dental plaques, selectively suppress specific pathogenic cells, and downregulate virulence factors.

The lytic enzyme designated L27 demonstrated the strongest bacteriolytic activity against S. mutans among the oral microbial flora. This study evaluated its potential anti-cariogenic effects. Because lytic enzymes act by degrading bacterial cell walls―structures absent in human cells―they are unlikely to damage host tissues. However, cytotoxicity testing was essential to confirm their biosafety. Our MTT assay results revealed no inhibition of cell proliferation at concentrations up to 250 μg/mL, establishing this as the maximum safe level for further experimentation.

In the artificial mouth model, voltage increased proportionally with Ca²⁺ concentration, confirming that the detection system was operating correctly. When S. mutans was added, the initial pH corresponded closely to physiological oral conditions, indicating that the model reliably reproduced the oral environment.

Although the addition of the lytic enzyme did not significantly alter the acidogenic activity of S. mutans, as indicated by similar pH reduction patterns, it markedly decreased Ca²⁺ release from HA beads. This observation may seem contradictory, since demineralization is typically driven by acid production. However, as noted earlier, Ca(OH)₂ released from HA beads could have neutralized the acid, thereby stabilizing pH even as demineralization slowed. Alternatively, the observed inhibition of Ca²⁺ release may result from specific properties of the enzyme itself, such as interference with bacterial adhesion or biofilm integrity.

Indeed, the enzyme exhibited a pronounced inhibitory effect on bacterial adhesion, even at relatively low concentrations. This property may represent the most significant aspect of its anti-cariogenic potential. Although adhesion inhibition was assessed using artificial saliva alone, real human saliva contains a variety of enzymes and buffering proteins that could enhance this effect. Therefore, it is reasonable to assume that coating HA beads with natural saliva before enzyme treatment―or varying the sequence of saliva and enzyme application―could further clarify the mechanism and optimize its preventive efficacy.

The precise mechanism underlying the inhibition of S. mutans adhesion remains unclear. It is hypothesized that the enzyme may form a thin film over the HA surface or release compounds that interfere with bacterial binding sites, thereby preventing adherence.

In summary, lytic enzyme L27 did not suppress acid production by S. mutans, but it effectively inhibited bacterial adhesion to HA beads and reduced Ca²⁺ release. Moreover, it enhanced the surface hardness of artificially demineralized enamel specimens. These findings suggest that, in addition to its bacteriolytic activity, enzyme L27 may exert anti-cariogenic effects by preventing S. mutans attachment, inhibiting demineralization, and promoting remineralization.

If further studies confirm its in vivo biocompatibility, safety, and stability, lytic enzyme L27 could provide a fundamental and effective approach for controlling dental caries―a condition that remains a major challenge in dentistry―and may also offer therapeutic potential for other bacterial infections.