G. mangostana pericarp and C. nutans leaf extraction: Powdered herbal materials were obtained from Vejpongosot Holding Co., Ltd., Thailand, and Don Bong Herbal Group, Prachinburi Province, Thailand. Each plant material was macerated in 95% ethanol (Labscan, Thailand) at a ratio of 1:10 (w/v) for seven days, with gentle agitation twice daily. The mixture was then filtered using a vacuum pump with Whatman No. 1 filter paper, and the ethanol was evaporated using a rotary evaporator (Heidolph Basis Hei-VAP ML Rotary Evaporator, Germany).

C. asiatica leaves extraction: Leaves obtained from Vejpongosot Holding Co., Ltd., Thailand, were subjected to Soxhlet extraction with 95% ethanol (Labscan, Thailand) for 3 days. The solvent was removed using a rotary evaporator (Heidolph Basis Hei-VAP ML Rotary Evaporator, Germany).

A. vera leaves extraction: A. vera leaves were collected from Prachuap Khiri Khan Province, Thailand. The inner gel was manually separated from the leaf, washed, homogenized, and freeze-dried using a freeze dryer (Labconco, Missouri, USA).

The extracted compounds were subjected to experimental design using Minitab (version 18, Minitab Inc., USA) to determine the optimal combination for antimicrobial efficacy against S. mutans and S. oralis (Arminian and Ozgur 2020). The experimental procedure is illustrated in Fig. 1.

Figure 1. 

Flowchart of the experimental procedure.

The individual and combined herbal extracts (G. mangostana, C. nutans, C. asiatica, A. vera) were optimized using a design of experiments (DoE) approach facilitated by Minitab (version 18, Minitab Inc., USA). A factorial design was selected to systematically evaluate various combinations and interactions of these extracts to determine their optimal antimicrobial efficacy (Chamutpong et al. 2021). The primary factors were the four herbal extracts in different combinations, with each extract either present or absent, coded as 1 or 0, respectively. This resulted in 15 runs, encompassing all possible extract combinations for evaluation, as shown in Table 2. The antimicrobial efficacy of each extract combination was evaluated by testing against oral pathogens S. mutans and S. oralis (Feng et al. 2024). This study assessed the antibacterial activity of herbal extracts against S. mutans JCM 5705 and S. oralis JCM 12997. Bacteria were grown on brain heart infusion agar (BHIA), adjusted to 0.5 McFarland (∼1 × 10⁸ CFU/mL), and spread onto BHIA plates. Extracts (10 µL) were applied to 6 mm sterile discs and placed on the plates, incubated anaerobically at 37 °C for 24 h. DMSO and tetracycline were used as negative and positive controls. Inhibition zones were measured in millimeters. All experiments were performed in duplicate.

The optimal extract combination was selected for minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) determination. Bacterial suspensions (0.1 mL, ∼1 × 10⁶ CFU/mL) were added to 10 mL brain heart infusion broth (BHIB). Extracts (10 mg/mL) were prepared in DMSO, diluted in BHIB, and tested in a 96-well plate using a two-fold serial dilution. DMSO and tetracycline served as negative and positive controls, respectively. A volume of 50 µL of standardized bacterial suspension was added to 50 µL of the prepared optimal extract formula or tetracycline in a 96-well plate. After adding bacterial inoculum, plates were incubated anaerobically at 37 °C for 24 h. Bacterial growth was initially assessed by visual inspection, followed by the addition of 10% Alamar Blue. The MIC was defined as the lowest concentration at which no visible growth was observed. To determine the MBC, 10 µL from each well showing no visible growth was streaked onto BHIA plates and incubated anaerobically at 37 °C for 24 h. The MBC was defined as the lowest concentration that prevented colony growth on agar plates. These methods were systematically applied to the effective concentrations required to inhibit specific oral pathogens (Thongyim et al. 2024).

Mucoadhesive films were formulated using the casting method. A full factorial design was applied in duplicate, varying the formulation for 36 runs (Minitab, version 18). The full factorial design included three factors:

1. Silicon dioxide (SiO ₂) (1, 1.5, 2 % w/w)
2. 3% of Chitosan (10, 15, 20% w/w)
3. Carboxymethyl cellulose (CMC) (1, 1.5% w/w).

All formulations are illustrated in Table 1.

SiO₂ was dissolved in 60–80 g PEG600 in a water bath. The 16-fold MIC extract from the previous experiment was mixed with 5 g of glycerin and 1.5 g of sorbitol (Phase I) (Limboo and Singh 2024). Carboxymethylcellulose and chitosan solution were then added using a homogenizer at 6000 rpm for 10 min (Phase II). Phases I and II were mixed until homogeneous and dried in a hot-air oven at 60 °C for 12 h.

Chemical properties of mucoadhesive film:

• Thickness and weight

The weight of the film was measured using a digital analytical balance (4 decimals, Mettler Toledo, Switzerland). Thickness was measured using a vernier micrometer (Digital Offset Caliper Vernier, Mitutoyo, Japan).

• Radial swelling index

A mucoadhesive film, 2 cm in diameter, was allowed to swell on the surface of an agar plate and then placed in an incubator maintained at 37 °C. The diameter of the swollen film was measured at predetermined time intervals for 90 min. The radial swelling index was calculated using the following equation:

Radial swelling index (%) = [(Dt-D0/D0) × 100]

where Dt is the diameter of the swollen film after time t, and D0 is the original film diameter at time zero (Nafee et al. 2003).

• Mucoadhesive film surface pH

The surface pH of the mucoadhesive film was determined using a pH meter (pH meter S400, Mettler Toledo, Switzerland). The film was placed on a Petri dish containing 5 mL of deionized water and allowed to swell for 1 h at room temperature. The pH was measured by bringing a combined glass electrode into contact with the surface of the swollen film and allowing it to equilibrate for 1 min. The average of three readings was recorded (Chaiprateep et al. 2018).

• Puncture resistance test

The puncture resistance test measured maximum stress and elongation at break of the film using a texture analyzer (TA.XT, Stable Micro System Co., Ltd., England). The experimental conditions included a load cell of 30 kg, spherical probe P/5S, contact force of 5 g, and a 0.30 mm/sec test speed. Testing was conducted on films of size 2.25 cm². Maximum stress, Young’s modulus (MPa), and elongation at break (%) were recorded. All experiments were performed in triplicate (Landová et al. 2014).

• Ex vivo mucoadhesive retention time

A modified USP disintegration apparatus lined with porcine buccal mucosa was used as the experimental model. The mucosa was rinsed with deionized water and then with simulated saliva (pH 6.2) at room temperature. The porcine buccal mucosa was affixed to a glass slide, and the wet film, previously moistened with one drop of simulated saliva, was applied to the mucosal surface using gentle fingertip pressure for 10 seconds. The glass slide was positioned vertically in the apparatus and allowed to move up and down, with its lowest point extending beyond the surface at the highest position. The beaker was filled with 800 mL of simulated saliva. The time required for the film to detach from the mucosa was recorded (Vasantha et al. 2011).

• Scanning electron microscope (SEM)

The mucoadhesive film was sectioned into 1 cm² samples and mounted onto aluminum stubs. To enhance conductivity, samples were sputter-coated with gold for 10 seconds. Cross-sectional morphology was examined using a scanning electron microscope (model JEOL, JSM 840, Tokyo, Japan) (Vry et al. 2020).

This study evaluated the ex vivo retention of active compounds through porcine buccal mucosa using a Franz diffusion cell (Garcia-Tarazona et al. 2023). Fresh buccal porcine mucosa was obtained from a slaughterhouse. The tissue was placed between the donor and receptor compartments and clamped together. The receptor compartment was filled with isotonic phosphate buffer (pH 7.4). The diffusion cell was maintained at 37 ± 2 °C, and the receptor compartment was continuously stirred throughout the experiment. After 6 h, the film samples were gently removed from the buccal mucosa, which was then collected. The active compounds retained in the buccal mucosa were determined by homogenizing the tissue with methanol to extract the penetrated compounds. The sample was filtered, and the concentration of phytochemicals was analyzed using HPLC.

This HPLC method quantified key active compounds in the optimized Thai herbal extract. Analysis was performed on a ZORBAX Eclipse Plus-C18 column (4.6 × 100 mm, 5 μm) at 30 °C with a 0.8 mL/min flow rate, 20 μL injection volume, and a 30-minute run time. Detection of caffeic acid, ferulic acid, and α-mangostin was carried out at 320 nm. The mobile phase consisted of a gradient of potassium dihydrogen phosphate buffer (pH 2.5) and acetonitrile. Calibration curves were prepared individually for caffeic acid (1.25–20 µg/mL), ferulic acid (10–70 µg/mL), and α-mangostin (25–125 µg/mL), each demonstrating high linearity (R² > 0.99), thereby ensuring accurate quantification of target compounds. Results are expressed as µg/mL.

The cytotoxicity of the mucoadhesive film was evaluated using a fibroblast cell line, based on the protocol adapted from Ninan et al. (2016) with modifications. Mucoadhesive film precursor solutions (100 µL) were dispensed into 96-well plates and allowed to form films. After cross-linking, the films were sterilized with 70% ethanol and rinsed with sterile 1× phosphate-buffered saline (PBS). Each film sample was incubated in 1 mL of Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 µg/mL streptomycin at 37 °C in a humidified 5% CO₂ atmosphere for 24 h to obtain the mucoadhesive film extract. NIH 3T3 fibroblast cells (American Type Culture Collection, Rockville, MD, USA; ATCC® CRL-1658™) (1 × 10⁴ cells/well) were seeded into 96-well plates and cultured under identical conditions for 24 h. The culture medium was then replaced with 100 µL of the mucoadhesive film extract, and the cells were incubated for an additional 24 h. Cell viability was assessed using the MTT assay. A 0.5 mg/mL MTT solution was added to each well and incubated in the dark for 4 h. The resulting formazan crystals were dissolved in a solubilization buffer, and the plate was incubated for another 2 h. Absorbance was measured at 570 nm using a microplate reader. Cell viability was expressed as a percentage relative to the untreated control (Ninan et al. 2016).

The selected mucoadhesive films were studied for accelerated stability by wrapping them in aluminum foil. The films were kept in an incubator maintained at 45 ± 0.5 °C for 24 h, then switched to 8 ± 0.5 °C for 24 h for five cycles. The mucoadhesive films were examined for changes in color, shape, and odor (Ways et al. 2018).

Statistical analysis was conducted to confirm the significance of the results. Data from each combination were analyzed in Minitab to assess the main effects, interaction effects, and the optimal combination of extracts for maximum antimicrobial activity. ANOVA was performed with a significance level of 0.05 to identify statistically significant factors and interactions (Minitab, version 18, USA) (Montgomery 1984).

The preparation of C. nutans, C. asiatica, G. mangostana, and A. vera extracts resulted in yields of approximately 5% for each extract, ensuring that all extracts were standardized to the same concentration for further study. The optimal combination of herbal extracts was found to target S. mutans and S. oralis. A combination of G. mangostana and C. nutans exhibited potent antibacterial activity. In the disc diffusion assay, this optimized combination produced inhibition zones of 8.720 ± 0.651 mm for S. mutans and 9.145 ± 0.091 mm for S. oralis. Although these values were slightly lower than those achieved with tetracycline, the positive control (9.365 ± 0.856 mm and 9.355 ± 0.431 mm, respectively), the herbal combination still yielded statistically significant inhibition (p < 0.05), as detailed in Table 1.

Further optimization using Minitab’s desirability function analysis (Fig. 2) confirmed this binary combination as the most effective among the combinations tested. The optimal combination of G. mangostana and C. nutans achieved a high composite desirability score of 0.9496 – approaching the ideal value of 1.0. Specifically, the desirability values for inhibition against S. mutans and S. oralis were 0.93932 (y = 9.3650 mm) and 0.93785 (y = 9.3550 mm), respectively. This indicates that the optimized formula successfully meets the desired antibacterial performance criteria for both target organisms.

Figure 2. 

The optimal combination of herbal extracts for anti-oral pathogen efficiency.

These results suggest a strong synergistic interaction between the phytochemical constituents of G. mangostana and C. nutans compared with single-extract combinations. Such synergy is essential in developing multicomponent herbal interventions, particularly for controlling oral pathogens, where the complexity of biofilms and microbial interactions requires multi-targeted mechanisms of action.

The optimized 1:1 combination of G. mangostana pericarp and C. nutans leaf extracts demonstrated potent anti-oral pathogen effects against S. mutans and S. oralis, with an MIC of 15.62 µg/mL and an MBC of 62.5 µg/mL. Specifically, α-mangostin – an abundant xanthone in G. mangostana – along with caffeic acid and ferulic acid – two key polyphenols in C. nutans – have been shown to exert antimicrobial effects at relatively low concentrations. The current findings highlight the potential of combining these phytochemicals to enhance antibacterial efficacy through synergistic interactions (Kaomongkolgit et al. 2009; Zulkipli et al. 2017). Optimization strategies, particularly those employing a design of experiments (DoE) approach, as demonstrated by Saelee et al. (2023), have proven effective in identifying synergistic herbal combinations that maximize bioactivity while minimizing required dosages. In this context, our study highlights the value of targeted phytochemical pairing to enhance antimicrobial performance, supporting the development of natural, plant-based alternatives to conventional antibiotics in oral healthcare (Saelee et al. 2023).

Key compounds were quantified to elucidate the contribution of individual bioactive constituents within the optimized combination. In the most effective herbal extracts, which utilized a 1:1 ratio of G. mangostana and C. nutans, the concentrations of caffeic acid (273.76 ± 0.15 µg/mL), ferulic acid (23.20 ± 0.30 µg/mL), and α-mangostin (9.92 ± 0.23 ng/mL) were determined. These concentrations are consistent with levels previously associated with antimicrobial activity, further validating their role in the observed inhibition of oral pathogens. Caffeic and ferulic acids have been shown to disrupt bacterial cell membranes, inhibit nucleic acid synthesis, and prevent biofilm formation – key mechanisms underlying their antimicrobial actions (Sivaranjani et al. 2018). α-Mangostin, in particular, is known for its ability to permeabilize bacterial membranes and has demonstrated efficacy against various Streptococcus species due to its potent lipophilic structure and antioxidant properties (Pedraza-Chaverri et al. 2008).

These findings support the hypothesis that combining polyphenolic acids from C. nutans with xanthones from G. mangostana produces a synergistic antibacterial effect greater than that of individual components. Such synergy is crucial for developing multitargeted botanical therapies capable of overcoming microbial resistance and disrupting complex oral biofilms. Therefore, this study reinforces the potential of rationally optimized, phytochemical-rich herbal combinations as viable and effective alternatives to synthetic antimicrobial agents for managing oral infections (Saelee et al. 2023).

Thirty-six experimental runs were analyzed using a design of experiments (DoE) approach (Table 2). The resulting films exhibited thicknesses ranging from 0.7 to 1.7 mm. Surface pH values across all formulations were between 6.5 and 7.5, close to salivary pH in the oral cavity (6.5–7.5). This indicates that the prepared mucoadhesive film of C. nutans and G. mangostana can be applied without causing mucosal irritation. Each formulation demonstrated a mucoadhesive retention time exceeding 6 h, affirming their potential for prolonged contact with the mucosal surface. The optimal formulation was identified through a design of experiments analysis, as shown in Table 2 and Fig. 3. Several parameters were evaluated, including tensile strength, mucoadhesion time, radial swelling index (%), and surface pH. The formulation containing 2% SiO₂, 1% CMC, and 15% chitosan offered the best overall balance, showing strong yet flexible films, sustained mucoadhesion for more than 6 h, and a near-oral cavity pH.

Figure 3. 

The optimal plot of mucoadhesive film formulation.

The incorporation of SiO₂ served as structural reinforcement within the polymeric matrix. SiO₂ enhanced mechanical properties by increasing both tensile strength and Young’s modulus, owing to strong interfacial interactions between the nanoparticles and polymer chains (Oliveira et al. 2010; Giri et al. 2012). This reinforcement is essential for maintaining the integrity of the film under mechanical stress during application and continuous contact within the dynamic oral cavity (Pooja et al. 2024). However, excessive SiO₂ loading can lead to agglomeration, resulting in poor film uniformity and reduced mechanical strength. Thus, its concentration must be carefully optimized (Oliveira et al. 2010). CMC, a hydrophilic, anionic cellulose derivative, plays a dual role in the formulation, acting as both a plasticizer and a flexibility enhancer. Its ability to absorb water promotes swelling and facilitates close contact with the mucosal surface through hydrogen bonding, which is a key mechanism for effective mucoadhesion (Schattling et al. 2017). Within the chitosan matrix, CMC disrupts the rigid interactions among chitosan chains, thereby increasing elongation at break and improving film flexibility (Li et al. 2012; Valizadeh et al. 2019). However, excessive CMC content can compromise tensile strength, emphasizing the importance of fine-tuning its concentration to achieve the desired balance between elasticity and mechanical stability (Alves et al. 2020).

Chitosan, a biocompatible, cationic polysaccharide, provides the essential adhesive base for the mucoadhesive film. Its positive charge enables strong electrostatic interactions with negatively charged mucin, enhancing mucoadhesion (Morales and McConville 2011; Ways et al. 2018). Additionally, chitosan contributes to the film’s overall mechanical strength. However, at higher concentrations, it can lead to increased brittleness, as reflected in decreased elongation at break values (Barik et al. 2024). The optimized formulation benefits from the synergistic integration of these three components. This combination yields a mucoadhesive film with excellent mucoadhesive properties, structural integrity, and compatibility for sustained oral delivery of herbal actives.

The present findings are consistent with previous reports demonstrating the potential of herbal-based mucoadhesive films for oral applications. For example, Mythri et al. (2011) developed a chitosan-based mucoadhesive film incorporating herbal extracts and observed prolonged mucoadhesive time and effective localized delivery, though their study utilized single-plant extracts. Similarly, Saelee et al. (2023) employed a design of experiments (DoE) approach to optimize herbal combinations in oral care formulations, underscoring the value of systematic optimization for enhancing synergistic antibacterial effects. Compared to these studies, our formulation uniquely combines four complementary herbal extracts – G. mangostana, C. nutans, C. asiatica, and A. vera – which provide a broader spectrum of bioactivities, including antibacterial, anti-inflammatory, and wound-healing effects. This multiextract approach represents a novel strategy that leverages phytochemical synergy to enhance efficacy while maintaining biocompatibility. In addition, our film achieved a mucoadhesive retention time exceeding 6 h, comparable to or exceeding the performance reported in previous works, with favorable mechanical properties and ex vivo mucosal retention of active compounds (Barik et al. 2024).

Following a 4-h application in the Franz diffusion cell system, analysis of residual porcine buccal mucosa tissue revealed significant retention of the active compounds: caffeic acid (235.28 ± 0.15 µg/mL), ferulic acid (19.4 ± 0.3 ng/mL), and α-mangostin (8.54 ± 23.46 µg/mL). This finding is consistent with the intended objective of developing a localized, topical treatment that does not require systemic absorption. Maintaining the active compounds within the buccal mucosa enables prolonged interaction at the target site, thereby enhancing their potential antimicrobial efficacy while minimizing the risk of systemic exposure.

The presence of α-mangostin – a lipophilic xanthone – in the receptor phase indicates its ability to permeate the buccal mucosa, aligning with previous reports on enhanced mucosal lipophilic permeation (Mane et al. 2014). In contrast, caffeic acid and ferulic acid, both hydrophilic phenolic acids, were primarily retained within the mucosal tissue, supporting the role of the buccal mucosa in localized delivery and tissue retention of hydrophilic phytochemicals (Yong et al. 2005).

The dual disposition – systemic absorption of α-mangostin alongside localized retention of caffeic and ferulic acids – underscores the multifunctional capability of the optimized mucoadhesive film. It not only facilitates transmucosal drug delivery but also offers potential for localized therapeutic effects at the site of oral infection or inflammation. These outcomes highlight the critical role of polymeric matrix design in modulating both the release kinetics and mucosal interaction of bioactive compounds.

Moreover, cytotoxicity testing of the optimized formulation showed >90% cell viability at all time points (24, 48, and 72 h), confirming its excellent biocompatibility. This is consistent with prior evidence supporting the safety of SiO₂, chitosan, and CMC for mucosal drug delivery applications (Li et al. 2012). These materials are known for their non-toxic profiles, soft tissue compatibility, and low irritation potential when administered via the buccal mucosa route.

Stability testing under accelerated conditions revealed no significant changes in the films’ color, shape, or odor, indicating good physical stability. Furthermore, scanning electron microscopy (SEM) images (Fig. 4) revealed a uniform surface morphology and a well-integrated matrix in both top and cross-sectional views. These structural characteristics enhance the film’s physical robustness and ensure a consistent distribution of actives. Collectively, these findings strongly support that the selected polymeric composition – 20% chitosan, 1% CMC, 2% SiO₂ – effectively balances physical properties, mucoadhesion, buccal mucosa retention, and biocompatibility. The optimized mucoadhesive film (Fig. 5) thus represents a promising delivery platform for natural anti-oral pathogen agents, supporting both localized and systemic therapeutic strategies in oral healthcare.

Figure 4. 

SEM images of the top view (a) and cross-sectional view (b).

Figure 5. 

The optimized mucoadhesive film.