Green Tea-Derived Catechins Suppress the Acid Productions of Streptococcus mutans and Enhance the Efficiency of Fluoride

Green tea is one of the most popular everyday beverages around the world, and it is considered to possess oral health-promoting properties. Because of the rich content of tea polyphenols, green tea is considered to have caries-preventive potential through the inhibition of caries-associated microbial activity [Hamilton-Miller, 2001]. The bioactive components of green tea are reported to influence the caries formation process at several different stages by inhibiting the glucose uptake and enhancing bacterial aggregation of streptococci, interfering with bacterial adhesion and biofilm formation, and suppressing cariogenic virulence factors [Otake et al., 1991; Xu et al., 2011; Han et al., 2021; Schneider-Rayman et al., 2021; Aragão et al., 2022]. These biological properties have been ascribed to green tea-derived catechins, which have the potential as clinical therapies due to their biological safety, multiple bioactivities, and synergetic effects with a wide range of antimicrobial molecules [Wu and Brown, 2021; Mehmood et al., 2022; Shao et al., 2022]. Evidence supports the idea that the preventative and therapeutic effects of green tea catechins on various diseases are based on their antioxidant capacity, anti-inflammatory activity, and antimicrobial properties [Zaveri, 2006]. Another antimicrobial component in green tea is fluoride, which is used for caries prevention worldwide. The main function of fluoride is to promote tooth surface remineralization and the subsequent formation of acid-resistant fluorapatite [tenCate, 1999], but fluoride is also known to inhibit microbial sugar metabolism and the subsequent acid production [Hata et al., 1990; Maehara et al., 2005; Takahashi and Washio, 2011]. Although green tea infusions have been reported to contain 0.26–4.09 ppm F (0.01–0.22 mm F) [Satou et al., 2021], several studies have concluded that catechins, rather than fluoride, are the most important factor in the caries-preventive effects of green tea [Onisi et al., 1981; Yu et al., 1992, 1995].

Green tea-derived catechins are chemically characterized by two benzene rings, which are referred to as the A-ring (resorcinol moiety) and B-ring (catechol moiety), and a dihydropyran heterocycle (the C-ring) with a hydroxyl group at carbon 3 (Fig. 1a) [Truong and Jeong, 2021]. They can be classified into non-galloylated catechins (epigallocatechin [EGC], epicatechin [EC], and catechin [C]) and galloylated catechins (EGCgallate [EGCG] and epicatechin gallate [ECG]) based on the group present at the R2 position (Fig. 1a). Their chemical structures, including the total number of hydroxyl groups they possess and the presence/absence of a galloyl moiety, are considered to be closely related to their bioactivities. However, the detailed effects of different structures on the antimicrobial properties of catechins are still unclear.

Although the direct antibacterial activity of non-galloylated catechins is weak, they exert antibiotic effects by sensitizing bacteria to antibiotics [Wu and Brown, 2021]. For example, galloylated catechins, including ECG and EGCG, were shown to reduce oxacillin resistance in mecA-containing strains of Staphylococcus aureus. The binding of these catechins to staphylococcal cells may be enhanced by the non-galloylated catechins EC and EGC, which significantly increased the capacity of galloylated catechins to reduce levels of staphylococcal oxacillin resistance [Stapleton et al., 2006]. Moreover, green tea extracts and catechins (EGC and EGCG) potentiated the antimicrobial effects of gentamicin against some clinical isolates of S. aureus and standard Pseudomonas aeruginosa strains [Fazly Bazzaz et al., 2016]. Therefore, non-galloylated catechins can still play an antimicrobial role through their combined effects with antibiotics, which may be a promising approach to combating microbial resistance.

EGCG, which possesses the highest number of phenolic hydroxyl groups of all catechins, is the most abundant catechin in green tea and exhibits higher antioxidant activity and bioactivity than the other catechins. In our previous study, we found that EGCG reduced the glycolytic metabolism of Streptococcus mutans by inhibiting glucose uptake-related enzymes (the phosphoenolpyruvate-dependent phosphotransferase system, PEP-PTS) and inducing bacterial aggregation [Han et al., 2021]. However, the differences between the antibacterial effects of galloylated catechins and non-galloylated catechins on caries-associated bacteria, and their combined effects with fluoride are still unknown. Consequently, we intended to compare the antimicrobial effects of different catechins and their combined effects with fluoride on the acid production and aggregation of S. mutans, and to further explore the mechanisms underlying these effects through in silico analysis. This study provided further insights into the antimicrobial effects of tea polyphenols and fluoride, and hence, revealed new perspectives regarding the caries control potential of green tea and the possible application form of green tea-derived catechins.

Five kinds of catechins that are present in green tea, including EGCG, ECG, EGC, EC, and C, were purchased from Funakoshi (Funakoshi, Tokyo, Japan) (Fig. 1a). The concentration of 1 mg/mL was used in the present study to compare the effect of different catechins, based on the IC₅₀ (50% inhibition concentration) of EGCG for S. mutans acid production being approximately 0.5–1.0 mg/mL [Han et al., 2021]. After 2 days grown on blood agar plates at 37°C in an anaerobic glove box with N₂, 80%; H₂, 10%; CO₂, 10% (NHC-type; Hirasawa Works, Tokyo, Japan), S. mutans (NCTC 10449) were maintained under anaerobic conditions at 4°C for further use, as described previously [Manome et al., 2019]. S. mutans were pre-cultured anaerobically overnight at 37°C in TYG medium, which contains 1.7% tryptone (Becton Dickinson, Franklin Lakes, NJ, USA), 0.3% yeast extract (Becton Dickinson), 0.5% NaCl, 50 mm potassium phosphate buffer solution (PPB, pH 7.0), and 0.5% glucose. Then, bacterial culture was transferred to TYG medium (final concentration: 5%) and grown under the same condition. In the logarithmic growth phase (optical density [OD] at 660 nm: 0.8–0.9), bacterial cells were harvested by centrifugation (21,000g, 7 min, 4°C) in double-sealed centrifuge tubes to maintain anaerobic conditions. Then, they were washed with washing buffer (2 mm PPB containing 150 mm KCl and 5 mm MgCl₂, pH 7.0) before being used. The washing and preservation of the cells were performed in the anaerobic glove box (N₂, 90%; H₂, 10%; NH-type; Hirasawa Works, Tokyo, Japan).

Overnight cultures of S. mutans (0.1 mL) with/without catechins (final concentration of 1 mg/L) were incubated in 2 mL TYG medium anaerobically for 14 h at 37°C. Growth curves of each group were recorded by detection of OD at 660 nm [Han et al., 2021].

Cell suspensions of S. mutans (1.5 mL, OD at 660 nm = 2.0) were mixed with catechin solution in washing buffer (1.5 mL) in a test tube and incubated at 37°C for 2 h. The OD at 660 nm of the mixture was examined at 10-min intervals. The bacterial aggregation rate was calculated using the following equation as previously described: aggregation rate (%) = (OD_before - OD_after)/OD_before × 100 [Han et al., 2021].

S. mutans (OD at 660 nm = 0.5, 1 × 10⁸ CFU/mL) were suspended in washing buffer. To measure the combined effect of catechins and fluoride, 2 mm potassium fluoride (KF) (pH 7.0) or 0.3 mm KF (pH 5.5) was added with/without 1 mg/mL catechins in the reaction mixtures after pre-incubation of the cell suspension at 37°C for 5 min. The concentration of fluoride was chosen based on the previous study [Maehara et al., 2005]. Then, glucose was added to the reaction mixture at a final concentration of 10 mm. We monitored 10 min acid production using a pH-stat system (AUTO pH-Stat, model AUT-211S; TOA Electronics, Tokyo, Japan). The experiments were carried out at pH 7.0 and pH 5.5. When the pH of the reaction mixture started to decrease, 0.025 m KOH was automatically added to the mixture to maintain the settled pH. Acid production was estimated by the amount of KOH added, as described in previous studies [Kawashima et al., 2013; Manome et al., 2019]. All of the experiments were performed in the NH-type anaerobic glove box. The inhibition rate of S. mutans acid production was calculated using the following equation: reduction rate (%) = (A–B)/A × 100, where A is the amount of titrated KOH in the control, and B is the amount of titrated KOH in the catechin with/or KF groups.

We treated S. mutans cell suspension (OD at 660 nm = 0.5) with/without 1 mg/mL catechins for 15 min and then washed twice before centrifugation. To measure PEP-PTS activity, the catechins/washing buffer-treated cells were suspended in 40 mm PPB (pH 7.0) containing 5 mm MgCl₂. The suspension was then treated with a 0.05 volume of toluene-acetone (1:4) and mixed vigorously to permeabilize the cells. The assay mixture contained 10 mm MgCl₂, 1 mm NADP, 1 mm phosphoenolpyruvate, 1 mm glucose, 2.1 U/mL glucose-6-phosphate dehydrogenase (from yeast; Roche Diagnostics, Mannheim, Germany), and permeabilized cells in 100 mm Tris-HCl buffer (pH 7.6). The increase in the concentration of NADPH was monitored spectrophotometrically at 340 nm and used to estimate PEP-PTS activity [Iwami and Yamada, 1985].

The bacterial PEP-PTS utilizes PEP in a group translocation process to phosphorylate incoming sugars via a phosphoryl-transfer involving general energy coupling, non-sugar-specific proteins, enzyme I (EI), and histidine protein, and subsequently, a sugar-specific, membrane-bound enzyme II complex (EIIC), which catalyzes the transportation and phosphorylation of the relevant carbohydrate [Vadeboncoeur and Pelletier, 1997]. Exogenous catechins are probably able to access the EIIC due to its location. The 3D structure of the EIIC (PDB CODE: 5iws) was obtained from the Protein Data Bank database [McCoy et al., 2016]. Its protein structure was prepared for modeling by adding hydrogen atoms and removing water molecules. The 3D structures of C (Compound CID: 9064), EC (Compound CID: 72276), EGC (Compound CID: 72277), ECG (Compound CID: 107905), EGCG (Compound CID: 65064), and d-glucose (Compound CID: 5793) were obtained from the PubChem database. The whole structure of EIIC was covered by the docking box and every possible functional pocket was included in the docking stimulation. The AutoDock Vina (v1.1.2) software (Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA, USA) was applied for molecular docking. Based on affinity values and the number of hydrogen bonds formed between catechins and EIIC, we chose the most practical conformation based on the lowest binding energy and visualized complexes of the conformation in PyMOL (DeLano Scientific Limited Liability Company, South San Francisco, USA).

The significant differences between the control and the other groups were evaluated using the Dunnett’s Test. p values of <0.05 were considered statistically significant. Data are presented as mean ± standard deviation values.

One mg/mL of EGC, EC, or C had no significant effect on the growth of S. mutans, but EGCG and ECG were found to have significantly inhibited the growth of S. mutans (Fig. 1b). To investigate the effects of catechins on bacterial aggregation, we examined the changes in the turbidity of bacterial cell suspensions. Fig. 1c shows that galloylated catechins induced bacterial aggregation; i.e., they caused a 61.4–81.0% reduction in the OD at 660 nm of S. mutans cell suspensions. The three non-galloylated catechins did not have any such effect. In addition, neither fluoride itself nor the addition of fluoride to catechins had any effect on bacterial aggregation (data not shown).

Based on the titration volume of KOH solution, bacterial acid production was monitored in real-time using the pH-stat assay (Fig. 2a, b). The rate of glucose-induced acid production was markedly inhibited by 1 mg/mL galloylated catechins (46.6–51.5% reduction at pH 7.0 and 45.0–51.9% reduction at pH 5.5). The three non-galloylated catechins reduced glucose-induced acid production slightly by 2.5–9.3%, but there was no significant difference from the control.

The PEP-PTS is a series of cell membrane-associated enzymes involved in sugar uptake and phosphorylation. After the addition of 1 mg/mL EGCG or ECG, the PEP-PTS activity of S. mutans decreased by 50.3% and 42.4%, respectively. On the contrary, EGC, EC, and C had weak inhibitory effects on the PEP-PTS activity of S. mutans, i.e., they caused it to decrease by <13% (Fig. 2c).

The molecular docking simulations showed that all of the examined catechins were able to bind to the transport domain of the EIIC, with affinities ranging from −8.2 kcal/mol to −9.9 kcal/mol, which are lower than the binding energy of glucose (−5.7 kcal/mol) (Fig. 3). In the best binding mode, the flavan-3-ol structures of the five catechins formed π-π interactions or π-donor hydrogen bonds with the 391-Phe on periplasmic helices 2 (HP2), which is a conserved amino acid of the EIIC used for substrate binding and contributes to stacking interactions with substrates (Fig. 3k ∼ o). EGCG was able to bind to the EIIC glucose transport domain with a free binding energy of −9.9 kcal/mol and to form hydrogen bonds with conserved amino acid residues in the EIIC, such as 239-His, 353-Glu, and 395-Glu (Fig. 3e, j, o). ECG showed a similar affinity for the EIIC, with a binding energy of −9.8 kcal/mol, and the galloylated benzene rings of both galloylated catechins interacted with the EIIC. Although the non-galloylated catechins, which contained fewer hydroxyl groups than the galloylated catechins, exhibited lower affinity for the EIIC (from −8.2 to −9.3 kcal/mol) than the galloylated catechins, they still showed the ability to bind to the EIIC. C formed the fewest hydrogen bonds with the EIIC and demonstrated the lowest binding affinity of −8.2 kcal/mol. The binding energy of catechins is consistent with their effects on PEP-PTS activity, which exhibited the following tendency: EGCG > ECG >> EGC > EC >> C >> glucose.

The combination of galloylated catechins and fluoride significantly inhibited acid production by S. mutans at both pH 7.0 (52.0–52.8%) and pH 5.5 (75.2–86.5%) (Fig. 4). The combination of non-galloylated catechins with fluoride did not reduce the acid production of S. mutans at pH 7.0 but had an inhibitory effect at pH 5.5, causing a 47.5–72.3% reduction in acid production.

Green tea-derived catechins are considered to have potential health benefits in the biomedicine field. Their antioxidant effects are known to be related to their chemical structures, including the presence/absence of galloylation and the total number of hydroxyl groups they possess [Vidal et al., 2014; Dos Santos et al., 2020]. The physicochemical structures of tea polyphenols also make a marked contribution to their antibacterial properties. At the concentrations used in this study, only the galloylated catechins (ECG and EGCG) inhibited the growth of S. mutans (Fig. 1b). The galloylated catechins also significantly inhibited bacterial acid production, while the non-galloylated catechins (C, EC, and EGC) did not show similar effects at either pH 7.0 or pH 5.5 (Fig. 2a, b). As was shown for EGCG in our previous study [Han et al., 2021], the present study demonstrated that EGC and EGCG inhibited bacterial glucose uptake by blocking PEP-PTS activity (Fig. 2c), and thus, acid production. These findings indicate that galloylated catechins inhibit glucose metabolism, which results in retarded energy production, and consequently, reduced bacterial growth.

Our in silico molecular docking analysis suggested that the effects of catechins are due to their interaction with the cell membrane-bound glucose transporter EIIC, a component of the PEP-PTS (Fig. 5). According to crystal structure analysis of the EIIC, the main functional region of the EIIC for glucose transport is the elevator structure formed by transmembrane helices (TM7) and periplasmic helices (PH1 and PH2) in the C-terminal substrate-binding domain [McCoy et al., 2016], which are colored red in Figures 3a∼e. Unlike the non-galloylated catechins, the galloyl structures of EGCG and ECG allowed them to bind tightly to the functional domain of the EIIC, and the ester moiety found in these molecules contains a rotatable carbon-oxygen bond, which provides a flexible molecular configuration for interacting with the transport domain of the EIIC, and thus, contributes to higher binding energies (EGCG and ECG: −9.9 and −9.8 kcal/mol, respectively). Similarly, the bacterial aggregation-inducing effect of the galloylated catechins was stronger than that of the non-galloylated catechins (Fig. 1c). This may also have been due to the characteristic intermolecular activity of the galloyl group, but further studies are needed to determine which bacterial surface molecules react with the galloyl group and how bacterial aggregation is induced. These results suggest that galloylation is a key pharmacophore for the antimicrobial activity of monomeric catechins against S. mutans. Early studies on the biofilm formation of Eikenella corrodens, a periodontopathogenic bacterium, reported a similar difference between galloylated catechins and non-galloylated catechins with a pyrogallol-type B-ring [Matsunaga et al., 2010].

The molecular docking analysis showed that the non-galloylated catechins (EGC, EC, and C) also exhibited a higher binding affinity for the EIIC than glucose and that the flavan-3-ol structures of the five catechins share the same binding site (391-Phe) on the EIIC (Fig. 3). This may explain why the non-galloylated catechins also had weak inhibitory effects on PEP-PTS activity and weak bacterial aggregation-inducing effects. The number of phenolic hydroxyl groups may also determine the binding affinity of catechins for bacterial proteins by influencing the formation of hydrogen bonds, enhancing intermolecular interactions, and allowing catechins to affect protein conformations and functions [Sugita-Konishi et al., 1999; Ingólfsson et al., 2011]. EGC, which possesses one more hydroxyl group, showed a slightly higher binding affinity for the EIIC and stronger aggregation-inducing effects than EC and C. Similarly, active catechins containing more phenolic hydroxyl groups were reported to inhibit leukotoxin A (LtxA) derived from Aggregatibacter actinomycetemcomitans, suggesting that hydrogen bonds are critical for such protein interactions [Chang et al., 2019]. Consequently, our results indicate that both galloyl structures and hydroxyl groups are involved in the inhibition of bacterial components by plant phenols. However, gallic acid, which is present in tea; the alkaloid caffeine; and theobromine were found to be inactive against Bacillus cereus in a previous study [Friedman et al., 2006]. Thus, it appears that the basic structure of catechins is also required for their antimicrobial activity.

Many studies have demonstrated that tea polyphenols make bacteria more sensitive to antibiotics; therefore, harnessing the synergistic effects of antibiotics and natural compounds may be a promising approach to reducing microbial antibiotic resistance [Sudano Roccaro et al., 2004; Bernal et al., 2010]. Interestingly, while galloylated catechins may reduce the oxacillin resistance of S. aureus, non-galloylated catechins significantly increased the staphylococcal binding capacity of galloylated catechins, which resulted in the bacteria becoming less resistant to oxacillin [Stapleton et al., 2006]. In our study, in addition to its individual bioactivity, fluoride, another major functional element in green tea, also had synergistic antimicrobial effects when co-administered with catechins. In the presence of fluoride, both non-galloylated catechins, which showed no ability to inhibit acid production individually, and galloylated catechins, exerted significantly enhanced inhibitory effects on S. mutans acid production in acidic conditions.

One mechanism through which catechins reduce multi-drug resistance is the inhibition of efflux pumps that actively expel drugs from cells, resulting in the accumulation of drugs inside cells and toxic effects [Sudano Roccaro et al., 2004; Stapleton et al., 2006]. The Fluc F⁻ channel is a highly selective cell membrane ion channel, which is used by microorganisms to resist fluoride toxicity [Stockbridge et al., 2013]. Since intracellular F⁻ will accumulate within bacteria according to the pH gradient, normally functioning Fluc channels, which can pump out excess F⁻, were found to be indispensable for maintaining microbe survivability in a mildly acidic growth medium [Ji et al., 2014]. We supposed that in acidic environments catechins may inactivate Fluc channel proteins, leading to intracellular F⁻ accumulation, and hence, reduce the fluorine tolerance of S. mutans (Fig. 5). This underlying mechanism still needs to be confirmed and may partially explain how the inhibitory effects of catechins on acid production under lower pH conditions are enhanced by the coadministration of fluoride. Although the non-galloylated catechins had no significant effect on bacterial growth or acid production or aggregation, their synergistic effects with fluoride under acidic conditions suggest that both the galloylated and non-galloylated catechins in green tea may play a role in manipulating bacterial metabolism.

The present study (1) revealed the antimicrobial activity of catechins alone and their synergistic effects with fluoride and (2) suggested that these antimicrobial activities of catechins are not only due to the presence of galloyl structures and phenolic hydroxyl residues but are also related to the flavanol skeleton. These results may facilitate the efficient application of catechins and fluoride for caries control. Further investigation is necessary to deepen our understanding of the molecular mechanisms and structure-activity relationships of natural catechins and their interactions with fluoride or Fluc channels. Additionally, exploration of the antimicrobial effects of catechins on other oral microbes besides S. mutans in biofilm and in vivo conditions, as well as the selective toxicity of long-term catechin use is also required.

An ethics statement was not required for this study type, no human or animal subjects or materials were used.

The authors declare no conflict of interest.

This study was supported in part by Grants-in-Aid for Scientific Research B (21H03151) and Grants-in-Aid for Scientific Research C (23K09475, 20K10241) from the Japan Society for the Promotion of Science, and Sichuan Science and Technology Program (2022NSFSC0614).

All authors contributed to the present study. Sili Han contributed to design, acquisition and analysis, and drafted the manuscript. Jumpei Washio and Yuki Abiko contributed to design and acquisition and analysis. Linglin Zhang contributed to conception. Nobuhiro Takahashi contributed to conception and design, acquisition, analysis and interpretation, and critically reviewed the manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.