Morphological and physicochemical impact of an experimental epigallocatechin-3-gallate based gel on erosive dentin wear

Abstract

Purpose

The aim of this study is to evaluate in vitro the morphological and physicochemical effects of an experimental epigallocatechin-3-gallate (EGCG) gel and its combination with sodium fluoride (NaF) on the progression of dentin erosion.

Materials and Methods

Initial lesions were induced in bovine dentin samples (n=10/group) and divided into: Control (gel without active ingredients), NaF (1.23% NaF), EGCG (10% EGCG), and EGCG+NaF (10% EGCG + 1.23% NaF). Half of the sample was protected and the other half treated and subjected to erosive/abrasive challenges (0.3% citric acid, pH=2.6 and simulated brushing). Volume loss (VL) and surface roughness (Ra) were evaluated by 3D laser confocal microscopy. Dentin composition was analyzed by Raman spectroscopy, considering mineral/matrix ratio (MMR), collagen organization, and crystallinity.

Results

The EGCG+NaF group presented the lowest VL values (p<0.05). The Control group had the highest MMR values, while the EGCG+NaF group presented the lowest (p<0.05). For the Amide I/III ratio, the Control group had the lowest values and the EGCG+NaF group the highest (p<0.05). No statistically significant differences among the groups regarding the crystallographic order (p= 0.668).

Conclusion

The combination of EGCG and NaF demonstrates the ability to mitigate volume loss under erosive-abrasive challenges, preserve structural integrity, and enhance collagen cross-linking within the demineralized organic matrix of dentin.

Introduction

Dental erosion (DE) is a multifactorial condition characterized by the progressive and irreversible loss of dental hard tissues through chemical dissolution induced by acids of non-bacterial origin (1). The acids responsible for erosion can have intrinsic origins, such as hydrochloric acid, or extrinsic origins, such as citric acid (2). Erosive demineralization initially leads to softening of the enamel, with initial lesions presenting smooth and shiny surfaces above the cementoenamel junction. If left untreated, dental erosion progresses to irreversible enamel loss, altering the morphology of the tooth, including flattened surfaces, concave cusps, and increased dentin hypersensitivity (3). According to Schlueter & Luka (4), the prevalence of erosion in the deciduous dentition ranges from 30% to 50%, and in the permanent dentition, from 20% to 45%. Furthermore, the management of this condition depends on the severity, origin, and symptomatology of the patient (5). When combined with mechanical forces, such as toothbrushing abrasion, this softened surface becomes more susceptible to progressive enamel loss (6) and, subsequently, of dentin. As the causative factor remains active for a prolonged period, the extent of the lesion may increase, consequently forming cavities as the dentin deteriorates (7).

Erosive wear progresses more rapidly and extensively in dentin than in enamel. Therefore, early dentin erosion lesions warrant particular clinical attention, as they are often subclinical and remain undetected during routine dental examinations (8). The dentin erosion process begins in the peritubular dentin, which has a higher degree of mineralization and surrounds the dentinal tubules. Subsequent dissolution of hydroxyapatite crystals within the intertubular dentin exposes the underlying collagen-rich organic matrix (7). These collagen fibers that make up this organic matrix can be biochemically degraded by proteases, such as host-derived matrix metalloproteinases (MMPs), present in dentin and saliva, thus contributing to the progression of dentin wear (9).

Although there is no universally accepted gold standard for the treatment of dental erosion, topical fluoride application remains a fundamental approach (10). Among fluoride compounds, titanium tetrafluoride (TiF₄) and sodium fluoride (NaF) have both demonstrated anti-erosive properties. Nevertheless, NaF is the most extensively investigated and clinically established fluoride agent, making it the preferred choice for the present experimental formulation (10, 11). In parallel, antioxidant agents, including proanthocyanidins, vitamin E, epigallocatechin-3-gallate (EGCG), and quercetin, have emerged as promising adjunctive therapies for the management of eroded dental substrates, demonstrating encouraging results (12).

The use of polyphenolic agents found in various plants and seeds (13), such as epigallocatechin-3-gallate (EGCG), present in green tea, exhibits inhibitory activity against MMPs (14, 15). In the exposed collagen matrix of dentinal tubules, EGCG acts on MMP collagenases, altering their structure, favoring the preservation of this matrix and thus contributing to the control of dentin wear (15). Previous studies have demonstrated that the incorporation of this polyphenol into restorative materials promotes antiproteolytic activity and absence of cytotoxicity (16). An in situ study with EGCG gels demonstrated a reduction in erosive dentin wear and indicated a physical occlusion effect of the dentinal tubules on the dentin surface (15). Therefore, EGCG appears to be a promising compound for the control of dentin erosion.

However, literature lacks studies on the combination of EGCG with NaF. Therefore, this in vitro study evaluated the morphological and physicochemical effects of an experimental EGCG gel on controlling the progression of erosion. The null hypotheses tested were as follows: H01: There will be no statistically significant difference in the morphological structure of eroded dentin treated with a gel containing 10% EGCG, either alone or in combination with 1.23% NaF; H02: There will be no statistically significant difference in the physicochemical properties of eroded dentin treated with a gel containing 10% EGCG, either alone or in combination with 1.23% NaF.

Materials and Methods

Ethical aspects

This study was conducted using healthy bovine teeth with approval from the local university's Ethics Committee on Animal Use under opinion number 4498260521.

Sample calculation

The definition of the sample size was calculated using as reference the study by Faraoni et al. (17) and considering a statistical power of 80%, α error of 5% and prediction of sample loss at the end of the study of 20%.

Specimen preparation

Specimens were prepared from the roots of recently extracted bovine teeth with healthy crowns and fully formed roots, obtained from young animals with an average age of 3 years. Initially, the teeth were cleaned and stored in a 0.1% thymol solution (Sigma-Aldrich Co.) at 4°C until use. Sixty dentin fragments measuring 4 mm in width by 4 mm in length were sectioned from the buccal and lingual surfaces of the roots using a diamond disc attached to a cutting machine (Isomet Low Speed, Buehler Ltd., Lake Bluff, Illinois, USA) under refrigeration. The specimens were manually polished with #600 and #1200 silicon carbide sandpaper (3M, Sumaré, São Paulo, Brazil) using circular motions under abundant water cooling (18), and placed in an ultrasonic cleaner (Odontobras, Brazil) for 3 minutes in deionized water between each change of sandpaper and after the final polishing. Subsequently, they were evaluated under a stereomicroscope (SZ2-ILST, Olympus SZ61, Tokyo, Japan) to check for cracks or fractures, and only those without such defects were selected (N=45).

Initial erosion, randomization of specimens and delimitation of the test area

Initial erosive lesions were created by immersion in 0.3% citric acid (pH ≈ 2.6) for 10 min (19). The specimens were then rinsed with distilled water, dried with soft absorbent paper (Kleenex – MMC Brasil Indústria e Comércio Ltda., Suzano, SP, Brazil), and re-evaluated under a stereomicroscope (SZ2-ILST, Olympus SZ61, Tokyo, Japan). Three specimens were excluded, resulting in 40 specimens, which were randomly allocated into four experimental groups (n = 10) using the Research Randomizer program (https://www.randomizer.org/) and stored at 4°C under relative humidity. Half of each specimen surface was protected with adhesive tape (UPVC tape – Graphic Tape, Chartpak, Leeds, USA) and coated with varnish (Colorama), leaving an exposed test area of approximately 4 × 2 mm. The protected area served as the baseline reference for volume loss and dentinal tubule analyses.

Experimental materials

All gels were produced by the same manufacturer, and their compositions are described in Table 1.

Table 1.

Description of experimental gels

Material	Manufacturer	Composition
Epigallocatechin-3gallate 10%		Hydroxyethylcellulose, propylene glycol, methylparaben, Imidazolidinyl urea, deionized water and 10 μM EGCG.
Sodium fluoride 1.23%	Company F&A Pharmaceutical Laboratory LTDA-São Paulo, Brazil	Hydroxyethylcellulose, propylene glycol, methylparaben, Imidazolidinyl urea, deionized water and 1.23% NaF.
Epigallocatechin-3 gallate 10% and sodium fluoride 1.23%		Hydroxyethylcellulose, propylene glycol, methylparaben, Imidazolidinyl urea, deionized water and 10 μM EGCG and 1.23% NaF.

Application of experimental gels

The experimental gels were applied by a single operator using a microbrush (KG Sorensen, Cotia, SP, Brazil) over the entire surface of the test area. After 5 minutes, the excess was carefully removed with a cotton swab (Cotonete). The gels were applied according to each experimental group: Control – gel without active agents; NaF – gel containing 1.23% NaF; EGCG – gel containing 10% epigallocatechin-3-gallate; EGCG+NaF – gel containing 10% epigallocatechin-3-gallate combined with 1.23% NaF. Subsequently, to form the acquired salivary pellicle, all specimens were immersed in artificial saliva (0.96g/1000 mL – KCl; NaCl; MgCl₂; K2HPO4; CaCl₂; carboxymethylcellulose; 70% sorbitol; nipagin; nipazole; and deionized water) for two hours (19).

Erosive-abrasive challenge

The erosive–abrasive cycling was performed over 5 days, with four daily erosive challenges consisting of immersion in 0.3% citric acid (pH = 2.6) for 5 min (20), followed by rinsing with deionized water and immersion in artificial saliva for 1 h. The erosive solution was renewed after each exposure, while deionized water and artificial saliva were replaced daily. Immediately after the first and last erosive challenges, the specimens underwent an abrasive challenge using a brushing machine (Odeme Dental Research, Luzerna, SC, Brazil) with soft Oral B Indicator toothbrushes (Procter & Gamble do Brasil S.A., Manaus, AM, Brazil). Each specimen was brushed for 45 cycles (15 s), using a 1:3 suspension of fluoride-free dentifrice, MalvatriKids Baby (Megalabs, Rio de Janeiro, RJ, Brazil), and artificial saliva (21). The suspension remained in contact with the specimens for a total of 2 min, after which they were rinsed with distilled water. At the end of each day, the specimens were stored at 4°C under relative humidity.

3D Confocal Microscopy Analysis

For physical analysis using 3D confocal microscopy, the tapes were removed from the specimens, which were then positioned parallel to each other with the aid of a parallelometer on the stage of the 3D laser scanning confocal microscope (LEXT OLS4000, Olympus, Tokyo, Japan). Image capture was performed at 1,024x magnification of the treated areas of each group using the OLS4000 software (Olympus, Tokyo, Japan). These images allowed for the evaluation of dentinal tubule diameter in µm and tubular area in µm³. All images were acquired by a single operator. For volume loss analysis, images at 20x magnification were obtained from the central region, showing the reference area (untreated and protected dentin) and the treated area (dentin subjected to treatment, erosion, and abrasion). The difference between the volume of the reference area and the test area determined the volume loss, measured in µm³, using the OLS4000 software (Olympus, Tokyo, Japan). All images were, once again, acquired by a single operator. Surface roughness (Ra) was evaluated in µm for each treated area of each group using the same parallel positioning of the specimens on the previously mentioned 3D laser confocal microscope. Measurements were performed using the mentioned software (17).

Raman Spectroscopy Analysis

The FT-Raman spectra were recorded using a Bruker Vertex 70v with a 1064 nm laser line. The beam diameter was 100 µm, with a power of 900 mW, a spectral resolution of 4 cm⁻¹, and 100 scans. The specimens were positioned perpendicularly to the light beam emitted by the equipment. Measurement control and data acquisition were performed using the OPUS 7.5 software provided by the spectrometer vendor. The data were processed and analyzed for greater precision using the OriginPro 8.5 software, where the peak positions and Raman bandwidths were obtained by fitting Lorentzian curves. The evaluation of the mineral and organic matrix volume, representing the volumetric fraction of the mineral relative to collagen, was performed using the mineral-to-matrix ratio (M/M) defined between the intensity of the 960 cm⁻¹ peak, corresponding to the phosphate group (PO₄³⁻) vibration, and the intensity of the 1650 cm⁻¹ peak, corresponding to the amide I group vibration, for each experimental group (22). Collagen cross-linking was assessed by the ratio of the amide II to amide I peak areas (23) for each experimental group, and the integrity of the collagen triple helix was evaluated by the height of the amide III peak for each experimental group (24).

Statistical analysis

The Jamovi 2.3.21.0 program (www.jamovi.org) was used for statistical analysis. The Shapiro-Wilk test was applied to verify normal distribution. The data obtained from the 3D Confocal Microscopy analysis showed normal distribution, and the One-Way ANOVA test followed by the Tukey post hoc test was used to compare the means of the evaluated groups. For the Raman Spectroscopy data, the mineral-to-matrix ratio and collagen organization were analyzed using the One-Way ANOVA test to compare the means of the evaluated groups, while the degree of dentin crystallinity was assessed using the Kruskal-Wallis test for group comparisons. In all analyses, a significant level of 5% was considered.

Results

3D Confocal Microscopy Analysis

The analysis of volume loss (VL) in µm² revealed a presence of statistically significant differences between the Control group and the other groups (p < 0.05). However, no significant difference was observed between the NaF group and the EGCG group (p = 0.078). The EGCG+NaF group exhibited the lowest VL values when compared to the other groups (p < 0.05). Regarding surface roughness (Ra), the EGCG+NaF group demonstrated lower Ra values compared to the other groups (p < 0.05). Similarly, no significant difference was found between the NaF group and the EGCG group (p = 0.081) (Table 2).

Table 2.

Mean and standard deviation M(±SD) of TSL volume (µm3) and roughness Ra (µm) between the evaluated groups.

Group	TSL Volume (µm3) Mean (±SD)	Roughness (Ra) (µm) Mean (±SD)
Control	-15.44 (±0.27)A	-0.982(±0.11)A
NaF	-7.45 (±0.17)B	-0.335(±0.08)B
EGCG	-6.56 (±0.18)B	-0.456(±0.11)B
EGCG+NaF	-1.99 (±0.59)C	-0.032(±0.06)C

*ANOVA One Way post-hoc Tukey

The photomicrographs of dentin from the treated area revealed diverse surface topography among the evaluated groups (Figure 1). The presence of inorganic precipitates or partial/total obliteration of dentinal tubules was observed in the EGCG+NaF group. The images illustrated the presence of tubules with larger diameter and greater quantity in the Control group.Raman Spectroscopy Analysis

Figure 1.

Surface of dentin blocks after treatment obtained by 3D confocal microscopy.

Mineral/Matrix Ratio (MMR)

The results of the mineral-to-matrix ratio (MMR), calculated as the intensity ratio of the PO₄³^⁻ peak (960 cm^⁻¹) to the Amide I peak (1600 - 1700 cm^⁻¹), are described in Table 3. The Control group exhibited the highest MMR values compared to the other groups (p < 0.05), while the EGCG+NaF group showed the lowest MMR values (p < 0.05). No statistically significant difference was observed between the NaF and EGCG groups.Collagen organization The data pertaining to collagen organization, obtained from the ratio of the intensity of the Amide I peak (1600 - 1700 cm⁻¹) relative to the Amide III peak (1200 - 1300 cm⁻¹), are presented in Table 4. The Control group exhibited the lowest values for the Amide I/Amide III ratio compared to the other groups (p < 0.05). Conversely, the EGCG+NaF group displayed the highest values (p < 0.05). No statistically significant difference was observed between the NaF and EGCG groups (p < 0.05). These results suggest a higher degree of collagen cross-linking in the EGCG+NaF group and greater collagen disorganization in the Control group.Degree of dentin crystallinity The data regarding the degree of dentin crystallinity were obtained from the full width at half maximum (FWHM) of the PO₄³⁻ peak (960 cm⁻¹) and are described in Table 5. No statistically significant difference was found (p= 0.668; χ²= 1.56) among the groups, indicating no difference in the crystallographic or atomic order between the studied groups.

Table 3.

Results of mean and standard deviation M(±SD) mineral/matrix ratio (MR) and amide III height of the evaluated groups.

Group	PO43- /Amide I M(±SD)
Control	40,18 (± 7,12)A
NaF	12,48 (± 2,53)B
EGCG	14,97 (± 2,67)B
EGCG+NaF	4,01 (± 0,92)C

*ANOVA One Way

Table 4.

Mean and standard deviation results M(±SD) of the amide I/amide III ratio data of the groups evaluated.

Group	Amide I /Amide III M (±SD)
Control	2.02 (± 0.71)A
NaF	4.27 (± 0.93)B
EGCG	5.01 (± 0.99)B
EGCG+NaF	8.42 (± 1.58)C

*ANOVA One Way

Table 5.

Median (Med), maximum and minimum of the full width at half intensity of the maximum peak of PO43- (960 cm-1) (FWHM).

Group	Med	Max	Mín
Control	19,5A	22	18,4
NaF	18,7A	20,6	17,8
EGCG	18,5A	21,8	17,7
EGCG+NaF	20,6A	22,1	18,3

*Kruskal-Wallis

The graphs relating to Raman spectroscopy of the vibrational peaks of PO₄₃-, amides I, II and III are presented in Figure 2.

Figure 2.

Graphs obtained by Raman spectroscopy of the Control, NaF, EGCG and EGCG+NaF groups. Letters: A – PO43- peak; B – vibrational peak of amide I; C – vibrational peak of amide II; D – vibrational peak of amide III.

Discussion

Erosive tooth wear has become a frequent condition, and its progression can lead to dentin hypersensitivity, directly affecting the individual's quality of life. Erosion may also be exacerbated by oxidative stress, which promotes the generation of free radicals and contributes to tissue damage (25). Studies are being carried out to better understand its mechanisms and etiologies (26), techniques for evaluating dental erosion (27), as well as control (28) and treatment (29) protocols.

3D laser confocal microscopy enables high-resolution analysis of tissue morphology and structural loss by providing a 3D view of the altered substrate (30). Thus, 3D confocal microscopy was used to evaluate the experimental gels proposed in controlling the progression of dentin erosion through inorganic dentin analysis. The group treated with the combination of EGCG and NaF showed less wear of the dentin structure compared to the other groups. Therefore, H0 was rejected.

The photomicrographs of the evaluated groups showed differences, indicating that the treatments provided greater potential for partial or complete obliteration of the dentinal tubules. In addition, it was observed that the combination of EGCG with NaF resulted in greater surface obliteration due to the presence of precipitates, even under erosive-abrasive challenges. Previous studies have shown that the protective effect of NaF against dentin erosion depends on the presence of an organic matrix, which increases surface area and diffusion, increasing the amount of fluoride adhered to the surface (31).

Although the organic matrix composed of collagen can be affected by abrasion (32), the present study demonstrated a potential synergistic effect between EGCG and NaF. This argument may be further reinforced by the similar performance observed in the groups individually treated with NaF and EGCG in this study. A possible explanation for this discovery is the presence of a more preserved demineralized organic matrix in the combined group (EGCG+NaF). This is because monovalent fluorides, such as NaF and AmF, work by forming adsorbed fluoride or a rich CaF-like layer2 in the dental structures. However, this layer dissolves rapidly during erosive events (33), a phenomenon that may have been mitigated or reduced by the mechanism of action of the antioxidant polyphenol EGCG as a potent inhibitor of dentin proteases.

Recent studies published in 2025 confirm that the efficacy of these polyphenolic inhibitors lies in their ability to shield collagen ultrastructure even under severe acid cycling regimens (34). In addition, research from 2026 shows that the incorporation of EGCG into hydrophilic vehicles enhances fluoride retention on the dentin surface, reducing the calcium desorption rate by up to 35% compared to the use of fluorides alone (35).

It is important to clarify that when dentin is exposed to an erosive challenge, rapid tissue dissolution occurs, leading to the exposure of a demineralized organic matrix on the surface (36). Maintaining the integrity of this matrix is essential to preserve a barrier that prevents deeper degradation of the substrate, as it functions as a scaffold (34). Degradation of the demineralized organic matrix promotes the release of inactive matrix metalloproteinases (MMPs) derived from dentin, which are activated in an acidic environment, initiating a process of disintegration of exposed collagen fibrils (37).

Recently, this approach against enzymatic degradation has been expanded: it has been shown that, in addition to MMPs, EGCG acts in the bimodal inhibition of endogenous cysteinal cathepsins, which act synergistically in the destruction of type I collagen (38). In response to this question, trials from 2026 have proposed the use of cross-linking agents of natural origin to mechanically stabilize dentin against structural collapse (39).

The group treated with EGCG combined with NaF had a lower surface roughness (RA) compared to the other groups. It is possible that the use of polyphenols such as EGCG, absorbed by the acquired salivary film, leads to its stabilization due to structural modifications, and may interact with the organic matrix of the dentin and promote the cross-linking of collagen, reducing its susceptibility to degradation (40). Rehage et al. (41) demonstrated through a study in situ that the thickness of the film increased immediately after washing with an EGCG solution. Nanotribological mappings carried out in 2025 corroborate this finding, revealing that the salivary film modified by polyphenolic gallates alters the coefficient of friction of eroded dentin, cushioning the mechanical abrasive impact of brushing (42). Therefore, the acquired film acts as a protective factor for both enamel and dentin, especially when modified by polyphenols, which justifies the results of this study.

Collagen is a protein composed of three helical polypeptides associated in a triple helix (43). Raman spectroscopy allows the analysis of collagen structures through vibrational peaks related to amides I, II, and III (24). Significant differences in the mineral-to-matrix ratio were observed among the experimental groups, with the EGCG + NaF group exhibiting the lowest values. Therefore, H₀₂ was rejected. This pattern of structural preservation resembles results found in 2025, where molecular Raman monitoring demonstrated that the integrity of the mineral-matrix ratio functions as a direct predictive indicator of dentin resistance to hypersensitivity progression (44).

Amide III allows you to investigate the integrity of the collagen triple helix, while amide I provides insights into cross-linking. The highest values for the Amide I/Amide III ratio were observed in the group treated with the combination of EGCG and NaF, suggesting an increased capacity for collagen crosslinking. Recent advanced spectroscopic analyses confirm that the increase in this molecular ratio reflects a conformational reorganization of collagen that physically prevents bacterial and endogenous hydrolases from accessing the protein's active site (45).

New laboratory and clinical assays using alternative protocols and thickeners for products combining the polyphenol EGCG and NaF, aimed at controlling the progression of dental erosion, should be developed.

Conclusion

The experimental gel combining EGCG and NaF demonstrated a promising effect in controlling volume loss against erosive-abrasive challenges, structural integrity and collagen crosslinking of the demineralized organic matrix of dentin.