Dentinogenesis involves the secretion of an extracellular matrix, which is subsequently mineralized. The formative cells, the odontoblasts, are derived from the ecto-mesenchymal cells of the dental papilla and lie on the pulpal interface of dentin, maintaining communication with the matrix through the odontoblast process. The extracellular matrix of dentin is rich in collagenous and noncollagenous components. Initial secretion of the unmineralized dentin matrix is followed by proteolytic matrix remodeling, together with secretion of further matrix components, which permit and regulate the mineralization process in the formation of dentin [1]. As well as a fundamental role in the regulation of matrix synthesis and secretion, the odontoblast is also responsible for the channeling of calcium and phosphate to the mineralization front at the predentin/dentin interface [2].

It is now clear that fluoride is capable of influencing the mineralization process of calcified tissues [3]. Lower concentrations of fluoride can reduce acid dissolution of mineral during dental caries by stabilization of the hydroxyapatite crystal lattice through fluoride substitution [4, 5]. However, higher concentrations can give rise to fluorosis characterized by abnormalities in mineralization [6, 7]. Though ionic substitution of the crystal lattice represents a predominantly physicochemical process, the tissue changes observed during fluorosis probably have a greater cellular basis. The characteristic mottled appearance of enamel during fluorosis does not appear to be paralleled by such obvious tissue changes in dentin and bone, although small hypomineralized interglobular spaces have been reported in the dentin matrix of fluorotic teeth [8].

At the molecular level, several studies have suggested that fluoride leads to extracellular matrix alterations in dentin [9, 10, 11, 12, 13, 14, 15]. Both collagenous and noncollagenous components appear to undergo structural alterations during fluorosis, although the precise mechanisms are unclear. Focus has been directed to the noncollagenous components, particularly the small leucine-rich proteoglycans, decorin and biglycan [16] and dentin phosphoprotein [17], in view of the putative roles of these molecules in mineralization. In vivo studies have indicated that fluoride exerts an increase in the levels of dermatan sulfate-substituted forms of decorin and biglycan with an overall increase in the charge density of the proteoglycans [13]. This later study confirmed previous in vitro studies investigating the influence of fluoride on proteoglycan synthesis within a rat incisor organ culture system, suggesting a reduction in sulfation [9] and a reduction in the length of the glycosaminoglycan chains as a result of posttranslational influences, hence leading to an overall reduction in the molecular size of the proteoglycans synthesized [12]. In vivo studies in rats have also shown that prolonged ingestion of high levels of fluoride (>20 ppm) results in a reduction in the phosphate content of the dentin phosphoproteins [14].

In view of the putative role of these noncollagenous components of dentin in mineralization, the aim of the present study was to investigate the influence of excess fluoride upon the mineralization process in dentin. An in vitro organ culture model [18] was adopted for the study enabling the normal tissue architecture of the dentin-pulp complex to be maintained together with direct application of known concentrations of fluoride to the cells. This model also allows us to study differences in mineralization over periods of 14 days, which were not possible in previous in vitro studies [12]. Both analytical and morphological approaches were undertaken with a particular focus on the transition of predentin to dentin where noncollagenous matrix modifications are important in the mineralization of dentin [1].

Materials and Methods

Rat Incisor Organ Culture

The organ culture of rat incisor tooth sections was performed according to the method of Sloan et al. [18]. This established culture model preserves the normal tissue architecture of the dentin- pulp complex with the odontoblasts lining the formative surface of the dentin matrix and maintenance of the underlying fibroblasts and other pulpal cells. Briefly, upper and lower incisor teeth were dissected from 28-day-old male Wistar rats sacrificed by cervical dislocation. Animals were maintained and humanely treated within the University’s animal facility in accordance with good practice and Government regulatory procedures. The incisors were placed in sterile washing medium, consisting of DMEM (Sigma, Dorset, UK), supplemented with 4 mM L-glutamine and penicillin/streptomycin/amphotericin (100 U/ml, 100 µg/ml and 25 µg/ml, respectively). Then, 2 mm of the growing root were removed and discarded and 2–3 transverse sections of rat incisors, approximately l–2 mm in thickness were cut with a segmented, diamond-edged rotary saw (TAAB Laboratories Equipment Ltd., Berkshire, U.K.), cooled with washing medium. The tooth sections were immediately placed in several changes of washing medium at 37°C prior to culture.

The tooth sections were transferred to individual wells of a plastic, 96-well plate containing 100 µl of embedding medium, consisting of DMEM, 4 mM L-glutamine, penicillin/streptomycin/amphotericin (100 U/ml, 100 µg/ml and 25 µg/ml, respectively), 0.15 mg/ml L-ascorbic acid, 10% heat inactivated fetal calf serum and 1% low melting point agarose. Culture medium (same composition as embedding medium with the omission of the agarose) was also supplemented with either 0 mM, 1 mM, 3 mM or 6 mM sodium fluoride. Such concentrations were used as these have previously been demonstrated not to be lethal to odontoblasts and to allow ECM synthesis to continue [12]. The embedding medium was allowed to cool until semisolid and each embedded tooth section was transferred to a sterile Millipore filter on a sterile plastic support, in a 35 mm Petri dish. The filters were suspended on the surface of 4 ml of fluoride-free or fluoride-supplemented (1 mM, 3 mM or 6 mM) culture media in Trowel-type cultures. Tooth sections were cultured at 37°C in 5% CO2 in air for 14 days, with the culture medium being changed every 2 days.

Quantification of Fluoride Incorporation into Dentin

The levels of fluoride incorporation into the dentin of tooth sections following culture in the absence and presence of 1 mM, 3 mM and 6 mM sodium fluoride were determined by established techniques for the quantification of fluoride in mineralized tissues [19]. Tooth sections (6 sections per fluoride concentration) were cultured at 37°C in 5% CO2 in air for 14 days, as described above. Following removal from culture, washing and blotting on a tissue, whole tooth sections were ashed at 600°C for 18 h and weighed. The ashed tooth sections were dissolved in 400 µl 1 M perchloric acid prior to the addition of 1200 µl 1 M sodium acetate, to give a final pH of 5.2. An equal volume of Total Ionic Strength Adjustment Buffer II (TISAB II) (Orion Research Incorporated, Massachusetts, USA) was added to each solubilized tooth section and the fluoride concentration was determined by a combined fluoride/reference electrode (Orion Research Incorporated), against known fluoride standards (0.1–200 ppm). Tooth sections (n = 6) cultured at 37°C in 5% CO2 in air for 14 days in the absence of sodium fluoride served as controls. The standard electrode potentials (mV) obtained were plotted against log10 of the fluoride concentrations of the standard solutions and the concentration of fluoride within the ashed tooth samples were expressed as parts per million (ppm). Levels of fluoride within the tooth sections following incubation in the different fluoride concentrations were compared and statistically analyzed by one-way ANOVA.

Tetracycline Hydrochloride Incorporation as an Indication of Dentin Mineralization

Analysis of dentin mineralization during culture was performed examining tetracycline hydrochloride incorporation into dentin in the absence and presence of sodium fluoride, through pulse labeling of the marker and microscopic visualization. The tooth sections were cultured in semisolid culture media and liquid culture media for 24 h, supplemented with 0 mM, 1 mM, 3 mM and 6 mM sodium fluoride, as described above. The tooth sections were removed from the semisolid embedding medium after 24 h and re-embedded in fluoride-free or fluoride-supplemented semisolid medium containing l µg/ml tetracycline hydrochloride (Sigma) (6 tooth slices per fluoride concentration). The re-embedded tooth sections were cultured for a further 24 h at 37°C in 5% CO2 in air with 4 ml fluoride-free or fluoride-supplemented liquid culture media also containing 1 µg/ml tetracycline hydrochloride. Tooth sections were then removed from the tetracycline hydrochloride-supplemented embedding and culture media after 24 h and re-embedded in fresh, tetracycline-free embedding and culture (fluoride-free or supplemented) medium for a further 12 days at 37°C in 5% CO2 in air.

Following 14 days in culture, tooth sections were removed, washed in Sorensens phosphate buffer (4 × 30 min), fixed and dehydrated through graded ethanols (3 × 15 min). The tooth sections were washed in LR White resin (4 × 30 min) and subsequently embedded in LR White resin for 24 h. Thin (approx 500 µm) slices of the embedded tooth sections were cut with a Microslice (Cambridge Scientific Instruments, Cambridgeshire, UK). Sections were then successively polished with graded aluminium oxide abrasive strips (Agar Scientific, Essex, UK) in distilled water to a thickness where predentin, dentin and enamel were identifiable when the ground section was examined under a light microscope. Sections were then examined with an Olympus Provis AX70A light/fluorescent microscope (Olympus Microscopes, Middlesex, UK) at ×400 magnification.

EDAX Analysis of Calcium and Phosphorus Ratios within Fluoride-Exposed Dentin

Following removal from organ culture, the tooth sections exposed to 0 , 1 , 3 and 6 mM sodium fluoride (n = 6/fluoride concentration) were fixed in 2.5% gluteraldehyde for 24 h and dehydrated through graded ethanol washes (4 × 15 min). Tooth sections were critical-point dried on a Polaron E3000 Critical Point Drying Apparatus (Thermo Electron Company, West Sussex, UK) and ion etched for 2 h using a B304/404 Microlap ion beam in an argon atmosphere (Ion Tech. Ltd., Middlesex, UK). The tooth sections were then carbon-coated on an EMScope SC 5000 Sputter Coater (Thermo Electron Company).

Calcium and phosphorus (Ca/P) ratios were determined at 30 random points in each section within the dentin along a region representing newly deposited dentin matrix, immediately above the identifiable predentin-dentin transition area, located approximately 10–20 µm from the pulpal border. Measurements were made with a Jeol Scanning Electron Microscope (SEM) with X-ray Microanalyser (JEOL Ltd., Hertfordshire, U.K.), at 15 KeV for 100 sec (2000 counts per second), counting the whole frame at a magnification of ×16,000 and quantified by comparison with an apatite standard (Agar Scientific). The changes in the Ca/P ratios obtained for each fluoride concentration were statistically analyzed by one-way ANOVA.

Results

Quantification of Fluoride Incorporation into Dentin

The levels of fluoride incorporation into the dentin of tooth sections following culture for 14 days in the absence and presence of 1, 3 and 6 mM sodium fluoride are shown in Table 1. The mean levels of fluoride within tooth sections culture at differing fluoride concentrations are given. Statistical analysis of the data using one-way ANOVA yielded a P value of <0.001, indicating that the difference among group means is extremely significant P values were then calculated for statistical differences between the fluoride conditions. However, this test does not take into account that same sets of data are used in multiple comparisons where there is a 5% probability that random chance would cause each P value derived to be less than 0.05. Therefore, P values were corrected using the Bonferroni method. Significant differences in the level of fluoride incorporation were evident on comparing those tooth sections cultured in fluoride (1, 3 and 6 mM) with those that received no fluoride supplementation (P < 0.001). However, no significant differences in levels of fluoride incorporation were observed when comparing tooth sections cultured at the various fluoride concentrations of 1 mM, 3 mM and 6 mM.

Table 1 Incorporation of fluoride into tooth sections following incubation within fluoride supplemented media
Full size table

Calcium/Phosphorous (Ca/P) Ratios within Fluoride-Exposed Dentin Mineral

The Ca/P ratios of the dentin mineral in tooth sections following culture for 14 days in the absence and presence of 1, 3 and 6 mM sodium fluoride are shown in Table 2. Percentage increases in the Ca/P ratios compared with the no fluoride control are shown in Figure 1. Statistical analysis using one way ANOVA indicated an extremely significant difference between groups (P < 0.0001). Table 2 shows P values calculated when comparing individual fluoride conditions, which were corrected to allow for multiple analyses using the Bonferroni method. All Ca/P ratios were demonstrated to significantly increase as a function of fluoride concentration, compared to tooth sections unexposed to sodium fluoride (P < 0.0001). No significant increase in Ca/P ratio was evident on comparing tooth sections exposed to 1 and 3 mM sodium fluoride (P > 0.8), 1 and 6 mM sodium fluoride (P > 0.05) or 3 and 6 mM fluoride (P > 0.05).

Table 2 Calcium phosphate ratio within newly formed dentin following incubation in fluoride supplemented media
Full size table
Figure 1
figure 1

Percentage increase in calcium/phosphorus (Ca/P) ratios of dentin of tooth sections following culture in the presence of 1, 3 and 6 mM sodium fluoride for 14 days compared with no fluoridecontrols (determined at 30 random points).

Full size image
Figure 2
figure 2

Tetracycline hydrochloride incorporation into rat incisor tooth sections following culture in the absence or presence of 1, 3 or 6 mM sodium fluoride for 14 days (×400). The fluorescent zone of incorporation at the predentin/dentin interface is indicated by arrows. Areas marked indicate the predentin matrix (PD), dentin matrix (D) and autofluorescence within the enamel layer (E).

Full size image

Tetracycline Hydrochloride Incorporation During Dentin Mineralization

Tetracycline hydrochloride incorporation into the dentin of tooth sections was examined following culture in the absence and presence of sodium fluoride for 14 days, and typical images obtained under the fluorescent microscope are shown in Figure 2. For tooth sections cultured in the absence of fluoride, a narrow and discrete fluorescent band was evident, as indicated by the arrows, and corresponds to the incorporation of tetracycline hydrochloride into newly mineralized dentin formed over the 14-day culture period. For all tooth sections incubated in the presence of 1, 3 or 6 mM fluoride, the fluorescence from the tetracycline incorporation in the predentin-dentin transition area appeared much more diffuse in nature (indicated by an arrow) and a narrow, discrete fluorescent band was not identifiable.

Discussion

The present study has demonstrated altered patterns of mineralization in the transition of predentin to dentin during the in vitro culture of tooth slices in the presence of fluoride, which shows correlations with the extracellular matrix changes previously reported in this region of the tissue [12, 13, 14]. Specifically, it has been possible to demonstrate incorporation of fluoride into the dentin matrix during culture, changes in the Ca/P ratios of the mineral and a more diffuse pattern of tetracycline incorporation, implying alterations in the regulation of the mineralization process.

The concentrations of fluoride adopted in the study were in the millimolar range to allow direct comparison with previous studies investigating extracellular matrix changes [12]. At these concentrations of fluoride, the morphological appearance of mature odontoblasts in culture is unaffected, but total collagen synthesis is reduced [15], confirming the previously reported effects of fluoride on the extracellular matrix [12, 13, 14]. These observations are at variance with data from exposure of cultured tooth germs to fluoride where de novo dentin formation and mineralization was reported to be unaffected [20, 21]. Such differences may reflect the stages of development of these tissues since matrix-vesicle-mediated mineralization [22] would be predominant in the tooth germ cultures; in contrast, dentinogenesis would be proceeding through matrix-mediated mineralization [23] in the cultured mature tooth slices.

Though there appeared to be a dose dependency in Ca/P ratios at the highest fluoride concentration studied (6 mM), this was not reflected in the fluoride uptake into the tissue or the altered patterns of tetracycline incorporation. In vivo, fluoride concentrations have been reported to be highest at or near to the pulpal surface [24], probably reflecting the continued apposition of matrix at this site. Previous studies have identified that fluoride exposure leads to reductions in both Ca2+ and PO 3−4 (including incorporation into phosphoproteins) uptake and transport in dentin and enamel tissues, following chronic fluoride exposure [20, 25]. The increase in the Ca/P ratio in the newly formed dentin, reported herein, is in agreement with previous studies which have identified similar trends within enamel and whole tooth germ explants [25, 26, 27]. These studies have suggested that fluoride influences the cellular activities of both odontoblasts and ameloblasts, and the alterations in Ca/P ratios with fluoride exposure may reflect alterations in the substitution of hydroxyl ions in the hydroxyapatite crystal lattice by fluoride ions [24]. Phosphate ions may also be substituted by fluoride ions, in conjunction with carbonate ions [24]. Such substitutions would provide a sufficient localized fluoride concentration to disrupt normal apatite nucleation and crystal growth, forming hypomineralized regions and also increasing Ca/P ratios [27]. Loss and substitution of hydroxyl and phosphate ions at fluoride concentrations above 100 ppm can lead to nonspecific deposition of calcium fluoride, giving rise to increased Ca/P ratios [24, 25]. This may partly explain the dose dependency of the Ca/P ratios in the cultured tooth slices at the highest fluoride concentration studied.

The rather diffuse incorporation of tetracycline within the dentin matrix of tooth sections exposed to fluoride implies alterations in the regulation of the mineralization process and would be in accord with the observations of small hypomineralized interglobular spaces in fluorotic dentin [28]. Changes in the pattern of dentin mineralization after a single acute dose of fluoride have also been reported [27] and decreases in tetracycline incorporation and dentin apposition have been observed during exposure to high fluoride concentrations [29, 30, 31]. The more diffuse appearance of tetracycline incorporation into the dentin matrix of tooth sections cultured in the presence of fluoride in the present study suggest that epitactic nucleation sites involved in matrix-mediated mineralization may have become disrupted. This would be in accord with the extracellular matrix changes previously reported in this region of the tissue [12, 13, 14, 15] and reinforces the importance of these extracellular matrix changes in the regulation of mineralization [1].

Thus, the present study has demonstrated that exposure of the dentin-pulp complex to higher concentrations of fluoride influences the mineralization process in the transition from predentin to dentin, which may well have a mechanistic basis in the fluoride-induced extracellular matrix changes arising in this region of the tissue. In view of the continued apposition of dentin throughout life, these observations indicate that exposure to high levels of fluoride may exert effects at the cellular level well beyond tooth development during primary, physiological secondary and tertiary dentinogenesis. There are also implications for the regeneration and repair of dentin after injury. The critical importance of growth factors sequestered within the dentin matrix [32] in dentin repair and their association with extracellular matrix components [33] imply that biological repair processes may also be susceptible to the effects of excessive fluoride exposure.