Research Studies
Study Tracker
Cytotoxicity and Genotoxicity of Fluoride Toothpastes in Buccal Cells.
| Acta Stomatologica Croatica | Vladislavic J, Tadin A, Zeljezic D, Sutej I, Basic K, Vladislavic N, Peros K. | 60(2):236–252.
Croatia Cytotoxicity Genotoxicity Toothpaste Human Study
View PDF

Abstract

Original full-test study online at
https://europepmc.org/article/PMC/PMC13292665#free-full-text

Objectives

This study investigated whether toothpastes containing different fluoride compounds influence cytotoxic and genotoxic alterations in buccal mucosal cells, with particular attention to the type of fluoride, the presence of fluoride itself, and the duration of exposure.

Materials and Methods

Eighty-eight participants were randomly assigned to four parallel groups: a control group using fluoride-free toothpaste and three intervention groups using formulations containing sodium fluoride, sodium monofluorophosphate, or amine fluoride. Buccal cell samples were obtained at baseline (T0), after 30 days (T1), and after 45 days (T2), and evaluated using the buccal micronucleus cytome assay to quantify nuclear abnormalities and cytotoxic markers.

Results

All fluoride-containing toothpastes led to higher frequencies of micronuclei, nuclear buds, and “broken egg” cells at T1 and T2 compared with the control group (P < 0.001). Amine fluoride and sodium monofluorophosphate produced sustained increases in cytogenetic markers, including nuclear buds (AmF: P = 0.013; NaMFP: P < 0.001) and “broken egg“ cells (AmF: P < 0.001; NaMFP: P = 0.004), while sodium fluoride demonstrated a slower, progressive increase in “broken egg“ cells (P = 0.036).

Conclusion

The findings suggest that fluoride-based toothpastes may modulate cytogenetic responses in buccal epithelial cells, and that these effects differ according to the fluoride compound and exposure duration. The findings should be interpreted with caution due to study limitations, and further research is needed to clarify the long-term biological consequences of repeated fluoride exposure in oral hygiene.

Keywords: MeSH Terms: Mouth Mucosa, Fluoride Treatment, Cytotoxins, Micronucleus Tests

Author Keywords: buccal mucosal cells, fluoride toothpastes, sodium fluoride, sodium monofluorophosphate, amine fluoride, cytotoxicity, genotoxicity

Introduction

Toothpaste formulations typically include various chemical elements, and while these elements are generally safe when used as directed, it is essential to be aware of potential risks associated with certain components. To date, fluoride stands as the sole nonprescription toothpaste additive with demonstrated efficacy in preventing tooth decay. The introduction of fluoride toothpastes into daily oral hygiene has led to a significant reduction in the incidence of dental caries in many populations around the world. Fluoride has a remineralizing effect on tooth enamel, and it reduces tooth demineralization, thereby preventing the development of caries. Toothpastes containing 1000 to 1500 ppm fluoride have been proven effective in the prevention of dental caries and are internationally supported (1). The main fluoride compounds currently found in toothpastes are sodium fluoride and sodium monofluorophosphate, although stannous fluoride and amine fluoride are also used (2). Fluoride toothpastes differ not only in their total fluoride content but also in the chemical form and bioavailability of the fluoride compound. Sodium fluoride (NaF) dissociates rapidly, releasing free fluoride ions capable of interacting directly with the oral mucosa. Amine fluoride (AmF), an organic surfactant-bound fluoride compound, exhibits strong adhesion to soft and hard tissues, leading to prolonged retention in the oral cavity and enhanced substantivity (3). Sodium monofluorophosphate (NaMFP) requires enzymatic hydrolysis to liberate fluoride, resulting in a more gradual ion release and potentially different epithelial exposure kinetics (4). These pharmacochemical distinctions may influence the magnitude and pattern of cellular responses in the oral epithelium, particularly over repeated daily exposure.

The acknowledgment of potential toxicity of fluoride was largely overlooked due to its commendable reputation for caries prevention. Although the doses present in toothpastes are usually low, there are concerns regarding long-term daily exposure, especially in sensitive populations, such as children or individuals with weakened oral barriers. In recent years, there has been a resurgence of interest in investigating its adverse effects. This renewed focus stems from an increasing awareness that fluoride can interact with cellular systems, even at low doses. Numerous studies have shown that fluoride has the potential to induce oxidative stress (5), influence intracellular redox balance (6), promote lipid peroxidation (7), and induce alterations in gene expression, ultimately leading to apoptosis (8, 9). Fluoride has been tested in various in vitro and in vivo systems assessing mutagenicity and clastogenicity, including several studies on exposed humans. The results of in vitro cytogenetic studies are mixed, however, the majority of evidence indicates that sodium fluoride can induce chromosomal aberrations and sister chromatid exchanges in cultured mammalian cells (10, 11). Studies have shown that fluoride varnish exhibits a concentration dependent cytotoxic effect on human gingival fibroblasts, with cell viability decreasing as fluoride concentration increases (12, 13). A study by Kleinsasser using the Comet assay demonstrated a minor genotoxic impact on human oropharyngeal epithelial cells and peripheral lymphocytes when aminofluoride was applied (14). The frequent observation of positive results in tests assessing chromosomal aberrations and sister chromatid exchanges support the hypothesis that fluoride may influence the genome, potentially leading to DNA rearrangements (15). However, the analysis of human epidemiological data from over 30 studies fails to show a clear relationship between fluoride exposure and disease, particularly in studies examining the correlation between water fluoridation and cancer (16). The oral mucosa serves as the first point of contact when using toothpaste, making it essential to understand the effects of fluoride on this sensitive tissue. As a crucial protective barrier, any damage to the mucosa can lead to various health issues, including heightened vulnerability to infections, ulcerations, and other pathological changes. The oral mucosa is renewed through a continuous process in which cells generated by mitosis within the basal layer gradually migrate toward the surface, replacing exfoliated cells. This basal layer contains progenitor and stem cells, and any chromosomal damage that occurs during their division such as chromosome breakage or loss can manifest as micronuclei (MNi) in the resulting daughter cells (17). After a brief genotoxic insult, micronuclei typically become detectable in exfoliated buccal cells only after the time required for these newly formed cells to reach the epithelial surface, a process estimated to take approximately 5–7 days. However, it has been observed that the peak expression of micronuclei may be delayed for up to 21 days (18). The micronucleus test was introduced at the end of the 19th century by Howell and Jolly, and since then, it has become the most widely and most reliable assay to evaluate cytogenetic damage (19). Chromosomal abnormalities detected in human lymphocytes are a well-known predictor of future cancer risk, and similar associations have been demonstrated for micronuclei (MNi) in these cells (20). As an alternative to peripheral lymphocytes, gingival and buccal epithelial cells can also be used to assess genotoxic and cytotoxic effects. These epithelial cells offer several advantages: they can be collected quickly and non-invasively, do not require cell culture, and can be analyzed without mitogenic stimulation or metaphase preparation. For these reasons, the micronucleus assay applied to epithelial tissues is regarded as a highly sensitive approach for monitoring genetic damage in individuals exposed to various genotoxic agents (21).

The relationship between fluoride exposure and buccal cell genotoxicity and cytotoxicity remains a subject of ongoing research. To date, several studies have explored the cytotoxic and genotoxic effects of fluoridated toothpastes on buccal mucosal cells. An in vivo study by Tadin et al. (22), evaluating the toxicity of fluoride in toothpaste on buccal epithelial cells, concluded that sodium fluoride (NaF) does not exert cytotoxic or genotoxic effects on these cells. Similarly, another study found that the simultaneous daily use of NaF toothpaste and mouthwash over a 12-week period did not result in statistically significant fluoride dependent cytotoxic or genotoxic effects on exfoliated buccal mucosa cells for most endpoints in the buccal micronucleus cytome assay (23). However, none of these studies have systematically examined the effects of three distinct fluoride active substances and compared their impacts. This study aimed to evaluate the cytotoxic and genotoxic effects of toothpastes containing fluoride, incorporating different active substances; sodium fluoride, sodium monofluorophosphate, and amine fluoride.

The primary objective of this study was to compare the cytotoxic and genotoxic responses of buccal mucosal cells following the use of three different fluoride-containing toothpastes and fluoride-free control toothpaste. The study also aimed to examine whether exposure duration influenced the observed cytogenetic effects. We hypothesized that fluoride-containing toothpastes may differ in their ability to induce cytogenetic alterations compared with a fluoride-free formulation, and that these effects may vary across fluoride compound types and exposure durations.

Materials and Methods

Study Design and Participants

The present study employed a prospective, randomized, triple-blind, parallel group clinical trial design to investigate the cytotoxic and genotoxic effects of toothpastes containing different fluoride-based active substances (sodium fluoride, sodium monofluorophosphate, amine fluoride), as well as to compare their effects. In this study, the investigators, participants, and the evaluator responsible for cytological scoring were all blinded to the type of toothpaste used. All statistical analyses were performed by an independent researcher who had access only to anonymized coded data and remained blinded to group allocation until completion of all analyses. A total of 100 participants aged 18 to 60 years were recruited from the Department of Restorative Dental Medicine and Endodontics at the School of Dental Medicine in Split, from students of the University of Split School of Medicine, and from adult volunteers who provided informed consent, as outlined in the recruitment flow presented in Figure 1. Inclusion criteria required participants to be in good general and oral health (ASA I) with a minimum of 20 teeth present in both jaws. Exclusion criteria included individuals with infectious diseases, chronic inflammatory conditions, recent use (within the past six months) of antibiotics, corticosteroids, or anti-inflammatory drugs, as well as those with oral mucosal lesions, periodontal disease, fixed prosthetic restorations, or orthodontic appliances. Pregnant individuals, patients who had undergone radiation therapy in the head and neck region, and those with known allergies to ingredients in oral hygiene products were also excluded. Participants were also excluded if they had a history of chronic alcohol consumption and/or long-term smoking (more than 10 pack-years). Written informed consent was obtained from all participants after a thorough explanation of the study objectives. Basic demographic information (age and gender) was recorded for all participants. The sample size for the present trial was determined using the effect size reported in an earlier in vivo investigation by Tadin et al. (2019) (22), which examined fluoride and sodium lauryl sulfate related cytotoxic and genotoxic changes in buccal epithelial cells. In that study, the mean micronucleus frequency differed between users of fluoride-free toothpaste (0.55 ± 0.51) and those using a fluoride-containing formulation (1.15 ± 0.88), corresponding to a Cohen’s d of 0.835. Using this effect size, together with a significance threshold of a = 0.05 and a statistical power of 80%, the minimum number of participants needed for each study arm was calculated to be 19 (22). To account for potential attrition during follow-up, the minimal required sample size of 19 participants per group was exceeded by intentionally recruiting 25 participants per group at the time of randomization. The study was carried out at two collaborating institutions: the Department of Restorative Dental Medicine and Endodontics, School of Dental Medicine, University of Split, Split, Croatia, and the Institute for Medical Research and Occupational Health, Zagreb, Croatia. The study was conducted in accordance with the Declaration of Helsinki (1975, revised 2002) and approved by the Ethics Committee of the School of Dental Medicine, University of Zagreb (No: 05-PA-30-20-9/2023), and the Ethics Committee of the School of Medicine, University of Split (No: 2181-198-03-04-22-0003). Written informed consent was obtained from all participants prior to enrollment. All aspects of the trial adhered to CONSORT recommendations, and the study was prospectively registered on ClinicalTrials.gov under the identifier NCT05596149.

Figure 1

Open in a separate windowFigure 1– Flowchart of participant recruitment and follow-up.

Materials, Clinical Procedure and Sample Collection

Before the study interventions were introduced, participants completed a 30-day washout period in which they exclusively used toothpaste without fluoride, a duration chosen to allow full turnover of the buccal epithelium and minimize any residual effects of previously used products. After baseline sampling, participants who met all eligibility criteria were randomly allocated into one of four study arms. The allocation sequence was generated using computer-based block randomization to ensure balanced group sizes throughout enrollment (24). An independent researcher, who was not involved in participant contact, sample collection, or microscopic evaluation, managed the allocation process and prepared the coded toothpaste tubes. All toothpastes were transferred into identical, opaque, precoded containers to maintain blinding. Coding was concealed from participants, the examiner responsible for buccal cell collection, and the examiner who performed microscopic scoring, thus ensuring full triple blinding. Baseline comparability among the four groups was confirmed using the Chi-square test for categorical variables and the Kruskal–Wallis test for age, with no significant differences detected (Table 2). Three of the groups used toothpastes containing fluoride, while the fourth control group continued to use fluoride-free toothpaste. The fluoride toothpastes tested contained either sodium fluoride (NaF), sodium monofluorophosphate (NaMFP), or amine fluoride (AmF). The composition of the toothpastes was the same, except for the active substances mentioned (Table 1). To maintain blinding, the toothpastes were labeled with letters (A, B, C, D) by pharmacy staffs that were not familiar with the randomization process. All experimental dentifrices were prepared at a specialized pharmaceutical formulation laboratory in Zagreb. The toothpastes were prepared using standardized manufacturing procedures (controlled mixing, homogenization under reduced pressure) to ensure a uniform dispersion of ingredients and to maintain identical physicochemical characteristics across formulations. The base excipient composition was kept constant, with the fluoride active substance being the only variable. All participants received standardized written and verbal instructions regarding brushing frequency and technique. Participants were instructed to brush their teeth with the allocated toothpaste twice daily, once in the morning and once in the evening, using a pea-sized amount (approximately 0.5 cm) and following the modified Bass technique for a total of three minutes. Adherence was monitored through periodic follow-up contacts and short self-report checklists. They were explicitly advised not to use any additional toothpastes or oral hygiene products, including mouthrinses or topical fluoride preparations. Throughout the study period, all individuals received the same model of toothbrush (Swissdent Profi Soft, Swissdent Care AG, Zurich, Switzerland) to standardize brushing conditions. For cytological sampling, buccal epithelial cells were collected from both cheeks using a cytobrush (Cytobrush Plus, GmbH Dietramszell-Linden, Germany). The baseline sample (T0) was obtained immediately before participants began using their assigned toothpaste, following a 30-day washout phase with fluoride-free toothpaste. Follow-up samples were collected at two additional time points: T1, after 30 days of exposure to the assigned toothpaste, and T2, after 45 days. Participants in the control group used the fluoride-free toothpaste for the full 45-day duration, and the collection of buccal mucosa cell samples was carried out identically to the fluoride toothpaste groups at T0, T1, and T2. Before each sampling session, participants refrained from eating, drinking, and smoking for at least one hour, and were asked to rinse their mouths with water to remove superficial exfoliated cells. All buccal mucosa samples were collected by a single trained examiner using a standardized buccal exfoliation protocol to ensure methodological consistency. The same examiner performed all sampling at T0, T1, and T2 to minimize interexaminer variability. Using a cytological brush, the buccal mucosa was gently brushed on both sides, and the buccal cell swab was placed in Falcon tubes containing chilled saline solution (+4°C). The samples were centrifuged within 60 minutes of collection, and the cell suspension was applied to glass slides and fixed with a methanol and acetic acid solution (3:1) (School of Medicine, Split). Afterward, the preparations were sent to the Institute for Medical Research and Occupational Health in Zagreb, where they were stained with 1% Schiff reagent and 0.2% Light-Green dye (Merck, Darmstadt, Germany). The coded slides were independently evaluated by trained expert using a blind approach. A scoring scheme proposed by Thomas and Fenech in their protocol published in Nature was carefully applied (25). The analysis of the stained preparations was performed using a light microscope (Olympus CX 40, Tokyo, Japan) set to 600× magnification, which allowed precise visualization and reliable evaluation of the cellular features. In addition to anomalies in the organization of genetic material (micronuclei, nuclear buds, “broken eggs“, nucleoplasmic bridges), the analysis of the slides included the identification and classification of various anomalies resulting from chromosomal instability and DNA damage, which pointed to specific forms of cell death. These anomalies were categorized according to established HUMNxl criteria, and 2,000 buccal exfoliated cells per participant were analyzed to ensure the reliability of the results (26). In the context of cytotoxicity parameters, cellular changes such as condensed chromatin (apoptosis), karyorrhexis (cell death characterized by nuclear fragmentation – apoptosis and necrosis), pyknosis (nuclear condensation – apoptosis), and karyolysis (nuclear dissolution – apoptosis and necrosis) were identified and categorized as indicators of early and late stages of apoptosis, pointing to cytotoxic effects. In addition to markers of cell death, the slides were analyzed for the presence of micronuclei, nucleoplasmic bridges, nuclear buds, and the “broken egg“ phenomenon (specific membrane damage) as indicators of chromosomal and DNA damage (25). To ensure objective and unbiased scoring of cytogenetic and cytotoxic parameters, all slides were anonymized before evaluation. Each participant and sample timepoint (T0, T1, T2) was assigned a unique alphanumeric code generated by an independent researcher who was not involved in slide analysis. All cytological evaluations were performed by an experienced evaluator with postgraduate training in cytomorphology and prior calibration in the HUMNxl buccal micronucleus cytome assay scoring protocol. The evaluator was fully blinded to group allocation and to all sampling time points throughout the study. To ensure scoring consistency, repeated intraobserver calibration sessions were carried out on a randomly selected set of slides before a full analysis was initiated.

Table 2

Characteristics Total
(n=88)		 Control
(n = 22)		 AmF
(n = 19)		 NaMFP
(n = 25)		 NaF
(n = 22)		 P – value
Age (X, SD) 36.0 (15.9) 43.5 (16.1) 32.6 (14.1) 32.5 (15.7) 35.36
(16.1)		 0.092
Gender (n, %) Male 31
(35.2)		 9
(40.9)		 5
(26.3)		 9
(36.0)		 8
(36.4)		 0.912
Female 57 (64.8) 13
(59.1)		 14 (73.7) 16 (64.0) 14
(63.6)
X – mean value; SD – standard deviation; n – number of participants; % – percentage of participants relative to the total number. P-value for age refers to between-group comparison of mean age. P-value for gender refers to the overall Chi-square test of sex distribution (male vs female). Statistical significance was set at P < 0.05
Open in a separate window

Table 1

Toothpaste Ingredients
Toothpaste A – 1000 ppm Amine fluoride Zeodent, Sodium lauryl sulfate, Sorbitol 70% solution, Sodium carboxymethylcellulose, Carbomer (Carbopol 980),
Sodium hydroxide 18% solution, Sodium saccharin, Essential oil of eucalyptus, Purified water, Amine fluoride (1000 ppm)
Toothpaste B – 1000 ppm Sodium monofluorophosphate Zeodent, Sodium lauryl sulfate, Sorbitol 70% solution, Sodium carboxymethylcellulose, Carbomer (Carbopol 980), Sodium hydroxide 18% solution, Sodium saccharin, Essential oil of eucalyptus, Purified water, Sodium monofluorophosphate (1000 ppm)
Toothpaste C – 1000 ppm
Sodium fluoride		 Zeodent, Sodium lauryl sulfate, Sorbitol 70% solution, Sodium carboxymethylcellulose, Carbomer (Carbopol 980), Sodium hydroxide 18% solution, Sodium saccharin, Essential oil of eucalyptus, Purified water, Sodium fluoride (1000 ppm)
Control – 0 ppm F Zeodent, Sodium lauryl sulfate, Sorbitol 70% solution, Sodium carboxymethylcellulose, Carbomer (Carbopol 980), Sodium hydroxide 18% solution, Sodium saccharin, Essential oil of eucalyptus, Purified water
Open in a separate window

Statistical Analysis

Statistical analyses were performed using SPSS version 25.0 (SPSS Inc., Chicago, IL, USA). Descriptive statistics (mean and standard deviation) were calculated for all variables. The Shapiro–Wilk test demonstrated a non-normal distribution of cytogenetic and cytotoxic parameters; therefore, non-parametric tests were used throughout. Between-group comparisons of the four toothpaste groups at each time point (T0, T1, T2) were conducted using the Kruskal–Wallis test. Within-group comparisons across time points were performed using the Friedman test. Whenever the Friedman test indicated statistical significance, pairwise post-hoc analyses were carried out using the Wilcoxon signed-rank test with Bonferroni correction to control for Type I error. Superscript letters in Tables 3 and 4 denote statistically significant pairwise differences identified in these Bonferroni-adjusted post-hoc tests (adjusted P < 0.05). Comparisons of age among groups were assessed using the Kruskal–Wallis test, while gender distribution was analyzed using the Chi-square test. Statistical significance was set at P < 0.05.

Table 3

Genotoxic parameters Toothpastes used
Control
(n = 22)		 AmF
(n = 19)		 NaMFP
(n = 25)		 NaF
(n = 22)		 P – value
Micronucleus
T0 4.02 (1.45) 4.84 (3.13) 5.64 (2.38)A 5.73 (2.55) 0.065
T1 4.06 (2.54)a,b,c 6.53 (2.48)a 7.44(2.43)A,b 7.14 (3.11)c < 0.001
T2 3.95 (2.36)a,b,c 6.79 (2.3)a 6.28 (2.09)b 6.14 (2.21)c < 0.001
P – value 0.923 0.104 0.007 0.074
Nuclear buds
T0 5.32 (2.85) 5.37 (3.4)A,B 5.47(2.52)A,B 6.23 (2.74)A 0.599
T1 4.27 (2.35)a,b,c 8.05(4.03)A,a 9.28(4.41)A,b 10.05(4.18)A,c < 0.001
T2 4.32 (2.34)a,b,c 7.79(4.49)B,a 7.2 (4.15)B,b 7.77 (3.41)c 0.004
P – value 0.776 0.013 < 0.001 < 0.001
Binucleated cells
T0 7.91 (2.6)A,B 7.16 (2.95) 8.76 (3.76) 9.27 (3.34) 0.323
T1 5.27 (2.45) A,a,b 6.79 (3.36)c 9.2 6(3.74)a 10.05 (3.8)b,c < 0.001
T2 5.59(4.44)B,a,b,c 8.68 (3.6)a 10.2 (4.46)b 10.32 (4.9)c < 0.001
P – value < 0.001 0.097 0.091 0.714
Nucleoplasmic bridges
T0 0.27 (0.55) 0.84 (2.29) 0.16 (0.37) 0.23 (0.43) 0.603
T1 0.27 (0.46) 0.26 (0.56) 0.24 (0.52) 0.32 (0.57) 0.930
T2 0.36 (0.58) 0.26 (0.45) 0.28 (0.46) 0.32 (0.48) 0.960
P – value 0.819 0.558 0.535 0.850
“Broken egg“
T0 4.23 (2.18) 4.11(2.21)A,B 5.44(2.62)A,B 5.77 (2.14) 0.051
T1 4.86 (2.23)a 6.05 (2.44)A 8.2(3.89)A,a,b 5.18 (4.01)A,b 0.007
T2 4.41 (2.02)a,b,c 7.89 (2.83)B,a 9.04(5.17)B,b 6.73 (3.99)A,c < 0.001
P – value 0.166 < 0.001 0.004 0.036
The data are presented as mean and standard deviation values. Statistical analysis was performed using the Kruskal–Wallis test to compare differences between toothpaste groups at each time point. Within each toothpaste group, differences across T0, T1, and T2 were evaluated using the Friedman test. When the Friedman test was significant, post-hoc pairwise comparisons were performed using the Wilcoxon signed-rank test with Bonferroni correction. Uppercase superscript letters indicate statistically significant differences between time points within the same toothpaste group, while lowercase superscript letters indicate statistically significant differences between toothpaste types after Bonferroni correction (adjusted P < 0.05). Abbreviations: T0 – after 30 days of using fluoride-free toothpaste; T1 – 30 days after using the fluoride-containing toothpaste; T2 – 45 days after starting the use of the fluoride-containing toothpaste.
Open in a separate window

Results

A total of 88 individuals were enrolled in the study, with a mean age of 36.03 ± 15.93 years. The key demographic variables for all four study groups are summarized in Table 2. No statistically significant differences were observed among the groups with respect to either age or sex distribution. The genotoxicity assessment across the four toothpaste groups—amine fluoride (AmF), sodium monofluorophosphate (NaMFP), sodium fluoride (NaF), and a fluoride-free control showed distinct temporal patterns in several cytogenetic markers, including micronuclei, nuclear buds, binucleated cells, nucleoplasmic bridges, and the “broken egg” phenomenon over the three sampling points (T0, T1, T2). As presented in Table 3, no significant differences in micronucleus frequency were observed at baseline (T0). However, by T1 and T2, all fluoride-containing toothpastes demonstrated significantly higher micronucleus counts than the control group (P < 0.001). Notably, NaMFP toothpaste showed a clear increase between T0 and T1 (P = 0.007). A similar pattern was observed for nuclear buds: all fluoride formulations displayed higher values than the control at T1 (P < 0.001) and T2 (P = 0.004). Significant within-group increases over time were observed for AmF toothpaste at both T1 and T2 compared to T0 (P = 0.013), and for NaMFP toothpaste, which displayed consistent rises from baseline (P < 0.001). NaF toothpaste demonstrated an increase from T0 to T1, with no further rise at T2 (P < 0.001). For binucleated cells, NaMFP and NaF toothpastes exhibited significantly higher levels than the control at T1, with an additional difference observed between AmF and NaF (P < 0.001). By T2, all fluoride toothpastes maintained higher binucleated cell counts compared with the control (P < 0.001). The control group showed a progressive decline in binucleated cells over time, with significantly lower values at both T1 and T2 compared to T0 (P < 0.001). No statistically significant differences were detected in nucleoplasmic bridges across groups or time points, suggesting minimal sensitivity of this parameter to the tested formulations.

For the “broken egg” phenomenon, notable variations were observed. At T1, NaMFP toothpaste showed significantly higher counts compared with the control and NaF toothpaste (P = 0.007). By T2, all fluoride-containing toothpastes displayed markedly elevated counts relative to the control group (P < 0.001). Within-group analyses showed increases for AmF and NaMFP toothpastes at both T1 and T2 relative to baseline (P < 0.001 and P = 0.004, respectively), while NaF toothpaste demonstrated a rise between T1 and T2 (P = 0.036). Table 4 provides an overview of the cytotoxic parameters assessed across the study groups. For karyolysis, no statistically significant differences were observed between groups at any time point. All groups maintained comparable cell counts, with within-group analyses showing a trend of increased karyolysis from T0 to T1, although this did not reach statistical significance. In karyorrhexis, significant differences emerged between groups at T1. The fluoride toothpastes exhibited increased cell counts compared to the control (P = 0.002). Within-group comparisons showed that cell counts peaked at T1 across all fluoride groups. Notably, in the control group, T2 counts were significantly higher than T0 (P = 0.016), while AmF toothpaste showed a statistically lower count at T2 compared to both T0 and T1 (P < 0.001). NaMFP and NaF toothpastes similarly showed more karyorrhexis cells at T1 than at T0 and T2 (P < 0.001). The pyknotic nuclei parameter showed significant increases at T1 in NaMFP toothpaste compared to both the control and AmF toothpaste (P < 0.001). NaMFP also had a significantly higher count of pyknotic cells at T2 compared to the control (P = 0.006), although the increase within its own time points was not statistically significant. Additionally, NaF toothpaste displayed significantly more pyknotic cells than the control at T1 (P < 0.001). In the control group, T2 counts of pyknotic cells were significantly lower than T0 (P = 0.032). Lastly, condensed chromatin exhibited significant inter-group differences at T1, with all fluoride-containing toothpastes showing higher counts than the control (P < 0.001). Significant within-group reductions in condensed chromatin were noted over time for both NaMFP and NaF toothpastes, with notable decreases from T1 to T2 and from T0 to T2 (P < 0.001 and P = 0.004, respectively).

Table 4

Cytotoxic parameters Toothpastes used
Control
(n = 22)		 AmF
(n = 19)		 NaMFP
(n = 25)		 NaF
(n = 22)		 P – value
Karyolysis
T0 110.73 (27.26) 117.47 (22.77) 116.64 (24.12) 118.27
(22.93)		 0.806
T1 130.68 (30.32) 126.21 (34.71) 130.56 (34.57) 128.5
(30.0)		 0.939
T2 125.32 (31.34) 119.53 (33.81) 131.64 (34.17) 120.41
(23.73)		 0.466
P – value 0.066 0.611 0.077 0.232
Karyorrhexis
T0 15.23 (5.54)A 19.74(5.19)A 19.48 (7.53)A 17.59 (7.75)A 0.112
T1 16.73 (6.75)a,b,c 22.95 (6.53)B,a 24.48 (9.33)A,B,b 24.27
(7.15)A,B,c		 0.002
T2 18.0 (4.31)A 15.4(5.46)A,B 18.4 (7.31)B 17.41 (5.15)B 0.470
P – value 0.016 ? 0.001 ? 0.001 ? 0.001
Pyknosis
T0 11.95 (4.41)A 10.68 (5.57) 14.4 (6.62) 14.09 (5.03) 0.144
T1 11.36(3.57) a,b 13.79(5.57)c 18.44 (8.25)a,c 16.95 (7.44)b 0.001
T2 10.09 (4.36)A,a 15.89 (10.44) 20.08
(10.26)a		 15.14
(8.95)		 0.006
P – value 0.032 0.343 0.066 0.053
Condensed chromatin
T0 20.82 (3.83) 25.16 (7.89) 24.12 (5.68)A 24.59 (8.65)A 0.171
T1 18.09
(5.18) a,b,c		 23.95
(8.46)a		 28.48 (10.06)B,b 25.64
(6.94)B,c		 ?0.001
T2 19.5 (6.71) 21.89 (5.38) 19.16 (7.3)A,B 18.73 (7.99)A,B 0.447
P – value 0.644 0.670 ? 0.001 0.004
The data are presented as mean and standard deviation values. Statistical analysis was performed using the Kruskal–Wallis test to compare differences between toothpaste groups at each time point. Within each toothpaste group (comparisons across T0, T1, and T2), differences were evaluated using the Friedman test. Post-hoc pairwise comparisons following a significant Friedman test were performed using the Wilcoxon signed-rank test with Bonferroni correction. Uppercase superscript letters indicate statistically significant differences between evaluation periods within the same toothpaste group, while lowercase superscript letters indicate significant differences between toothpaste types based on Bonferroni-adjusted P values (adjusted P < 0.05). Abbreviations: T0 – after 30 days of using fluoride-free toothpaste; T1 – 30 days after using the fluoride-containing toothpaste; T2 – 45 days after starting the use of the fluoride- containing toothpaste.
Open in a separate window

Discussion

This study investigated the cytotoxic and genotoxic effects of several fluoride-containing toothpastes on buccal mucosal cells, with the aim of determining whether cytogenetic alterations differ according to fluoride compound, the presence of fluoride, and exposure duration. Fluoride-containing formulations showed distinct patterns compared with the non-fluoride control, and differences were also observed among individual fluoride compounds. Some markers decreased while others increased over time, indicating that cytotoxic and genotoxic responses fluctuate with continued exposure. Overall, these findings suggest that fluoride-containing toothpastes may induce modest but compound-specific variations in cytogenetic markers, thus supporting the possibility of a more complex cellular response to prolonged fluoride exposure (27). All fluoride-containing toothpastes showed higher frequencies of micronuclei and nuclear buds compared with the non-fluoride control, indicating a modest genotoxic response that varied among compounds. These findings are broadly consistent with previous research demonstrating that fluoride exposure can induce DNA damage and alter cell-cycle dynamics in oral epithelial cells (28, 29). The differing patterns observed among AmF, NaMFP, and NaF likely reflect their distinct fluoride-release mechanisms. NaMFP undergoes enzymatic hydrolysis, resulting in gradual ion release and prolonged low-level exposure (30), whereas AmF binds to salivary proteins and mucins, creating a sustained fluoride reservoir on oral surfaces (31). In contrast, NaF dissociates rapidly, producing higher immediate ion availability and potentially more acute cellular stress, consistent with reports of NaF-associated mitochondrial and metabolic disruption (32). Similar time-dependent cytotoxic effects have been observed in studies evaluating commercial toothpaste formulations (Ghapanchi et al.) (33), suggesting that prolonged or repeated fluoride exposure, together with formulation specific factors, may influence cytogenetic outcomes (34, 35). Fluctuations in binucleated cell counts and other genotoxic markers showed broadly similar trends across the fluoride-containing toothpastes, although the timing and magnitude of these responses differed among compounds. These compound-specific patterns likely reflect differences in fluoride release kinetics, as slower release formulations may produce more sustained but less acute cellular effects. In contrast, the non-fluoride control exhibited relatively stable genotoxic parameters, supporting the notion that the observed variations are attributable to fluoride exposure rather than natural temporal changes.

Nucleoplasmic bridges showed minimal variation across groups, suggesting that this endpoint may be less sensitive to the types or concentrations of fluoride tested. More distinctive patterns were observed in the “broken egg” phenotype, where fluoride-containing toothpastes consistently produced higher counts than the control. This aligns with evidence that fluoride retention in the oral cavity can prolong epithelial exposure and contribute cumulatively to nuclear abnormalities (36, 37). The stronger early response observed with NaMFP is consistent with its enzymatic hydrolysis and gradual ion release, whereas similar findings in NaF varnish studies (Escobar-García et al.) (38) support the broader concept that prolonged fluoride contact, regardless of formulation, may influence chromatin and nuclear morphology.

However, the literature on this endpoint remains mixed. Ribeiro et al. reported no genotoxic effects at low NaF concentrations in rat oral mucosa, suggesting that dose, assay type, and exposure duration strongly influence outcomes (39). Their study used the comet assay, which detects strand breaks rather than chromosomal level anomalies; thus, the micronucleus cytome assay applied in our study captures complementary forms of genomic instability not detectable by comet analysis. The distinct temporal pattern observed with NaF, characterized by a more progressive increase, may reflect its rapid ion availability, which can impose acute oxidative or metabolic stress before repair mechanisms fully compensate. This interpretation is supported by evidence that NaF can impair mitochondrial function and cellular energy balance (40), raising the possibility of similar effects in buccal epithelial cells under prolonged or repeated exposure. Cytotoxicity-related markers provided additional insight into the cellular response to fluoride exposure. Karyolysis showed minimal variation across groups, suggesting that gross membrane disruption and advanced nuclear degeneration were not strongly affected by the fluoride formulations tested. This finding is consistent with the results of Tadin et al., who likewise reported no fluoride-dependent changes in karyolysis in buccal cells (22). In contrast, karyorrhexis exhibited more pronounced, though transient, changes. Fluoride-containing toothpastes generally showed an early increase in nuclear fragmentation, followed by a decline, a pattern compatible with an initial oxidative or metabolic stress response followed by partial cellular adaptation or repair. AmF tended to show a milder profile, which may relate to its protein binding properties and sustained but lower intensity fluoride release (31). NaMFP and NaF demonstrated patterns consistent with studies showing that fluoride induced oxidative stress can elevate cytotoxic markers before cellular homeostasis is restored (41). The potential contribution of formulation excipients should also be considered: Tadin et al. reported that sodium lauryl sulfate (SLS) can amplify fluoride related nuclear abnormalities by increasing membrane permeability (22). The findings of Vladislavic-Zorica et al. further support the notion of time-dependent fluctuations, as increases in karyorrhexis and condensed chromatin at 30–60 days were followed by a reduction at 90 days when using NaMFP and NaF-based toothpastes (42). Differences between their outcomes and ours likely reflect formulation variability, particularly the presence of whitening agents, which may impose additional oxidative burden and alter the cytotoxic trajectory compared with standard fluoride formulations. Pyknosis and condensed chromatin provided additional insight into the cytotoxic trajectory associated with prolonged exposure to different fluoride formulations. Fluoride-containing toothpastes generally produced early increases in pyknotic nuclei, which is consistent with an acute cellular stress response. NaMFP showed the most pronounced early effect, likely reflecting its enzymatic hydrolysis and gradual ion release pattern, while NaF’s rapid ion availability aligns with reports of short-term oxidative stress induction. These observations are supported by Tadin et al., who found that formulations containing sodium lauryl sulfate (SLS) amplified pyknosis, indicating that surfactants may potentiate fluoride-related membrane stress (22). Additional evidence from Tabatabaei et al. underscores the importance of concentration and exposure time, as NaF demonstrated a clear dose and time-dependent cytotoxic effect in gingival fibroblasts (41), although such responses may differ between gingival and buccal epithelial cells. Condensed chromatin followed a similar pattern, with fluoride formulations inducing a transient early increase suggestive of chromatin condensation during cellular stress, followed by a decline that may reflect partial adaptation or activation of repair mechanisms. The findings of Tadin et al. support the role of SLS in enhancing chromatin condensation (22), whereas Puizina et al. reported decreasing condensed chromatin levels over time (23), partially aligning with the later decline observed in our study. The differences between studies likely reflect variations in toothpaste composition, particularly the presence of whitening agents or surfactants, as well as differences in population characteristics and exposure conditions. This study provides valuable insights into the cellular effects of fluoride-containing toothpaste, but it also has some limitations. The observation period of the study is short (30 and 45 days), therefore it may not capture the full extent of cytotoxic and genotoxic effects associated with prolonged toothpaste use. It is challenging to control all variables within the oral cavity environment, where complex interactions between fluoride, saliva, and the oral microbiome can significantly influence cytotoxic and genotoxic effects. Additionally, the study relied solely on the micronucleus test for genotoxicity evaluation, which, while effective, does not detect all types of DNA damage, such as oxidative DNA lesions. Further research is required. It should extend observation periods to cover longer durations, potentially several months or years, to capture cumulative cellular effects of fluoride exposure. Incorporating additional genotoxicity assays, such as the comet assay, would offer a more thorough analysis by identifying single-strand and double-strand DNA breaks (43). A crossover study design could allow participants to act as their own controls, helping to control for individual oral hygiene habits. Although participants were instructed not to use additional fluoride sources (e.g., mouthrinses), residual variability in dietary fluoride intake or drinking water fluoride content cannot be completely excluded. Moreover, assessing saliva pH and oral microbial composition would provide valuable context for understanding how these en-vironmental factors influence fluoride’s cytotoxic and genotoxic effects, ultimately giving a more nuanced view of fluoride’s impact in the oral cavity.

Conclusion

This study examined potential cytotoxic and genotoxic effects of fluoride-containing toothpastes on buccal mucosal cells, assessing the influence of fluoride compound type, fluoride presence, and exposure duration. Statistically significant differences were observed between fluoridated and non-fluoridated toothpastes, as well as among different fluoride compounds, suggesting that fluoride exposure may be associated with cytogenetic changes in oral mucosal cells. Observed variations among different fluoride compounds may reflect differences in their release profiles, potentially influencing cellular responses over time. While some markers suggested possible cellular adaptation or stabilization, others raised questions regarding cumulative effects that warrant further investigation. Given the methodological limitations of the buccal micronucleus cytome assay, these findings should be interpreted with caution. The study underscores the complex interactions between fluoride and oral epithelial tissues and highlights the need for further research using complementary genotoxicity assays to better characterize early and progressive cellular effects. Complementing these investigations, longitudinal studies are needed to better understand the long-term biological implications of fluoride use in oral hygiene products, particularly in the context of prolonged exposure and varying fluoride release mechanisms.

Acknowledgements

This study was supported by the University of Zagreb Support Project awarded to the School of Dental Medicine, University of Zagreb. It is also part of the project conducted by the Institute for Medical Research and Occupational Health titled “Evaluation of Efficacy and Toxicity of Biologically Active Substances WP5: Ecogenetic Research in Biomonitoring of Populations in vivo and in vitro” (Class: 643-02/23-01/00016, Reg. no. 533-03-23-0006).

The authors wish to thank all participants for their invaluable contribution to this study.

Footnotes

Conflict of interest

The authors declare no conflicts of interest.

References