1 INTRODUCTION

Calcium phosphates (CaPs), mainly hydroxyapatite (HAp), due to their chemical and structural similarity with bone and teeth mineral components are widely used for replacement, regeneration, and augmentation of calcified tissue.1, 2 As resolved previously,3-5 biological HAp is not stoichiometric and may contain various amounts of substitutions (i.e., F−, HPO42−, CO32−, Na+, Mg2+, Sr2+). Many synthesis methods have been developed to produce substituted and non-stoichiometric HAp with different particle morphology.6-15 Such materials can be used for preventive dentistry as an active additive to dental care products.

In the last two decades CaPs nanomaterials have been studied as remineralizing agents for tooth enamel protection from early caries.16-18 In the scientific literature two processes are proposed to explain the CaPs protective effect on tooth enamel. It has been suggested that CaPs can boost the remineralization of lost minerals due to their ability to release Ca2+ and PO43− ions, thus supplying mouth cavity with extra Ca2+ and PO43− ions.19, 20 The other process is associated with CaPs particles that easily block dentine tubules or settle on the enamel surface defects.21, 22

Calcium deficient hydroxyapatite (CDHAp) is potentially a good alternative to stoichiometric HAp to grant additional functionality. It can be easily substituted with various ions. Accordingly, we chose fluoride ions (F−) for substitution of hydroxyl ions (OH−) due to verified anti-caries performance in clinical studies for many decades,23 and strontium ions (Sr2+) for substitution of calcium ions (Ca2+), as strontium salts are being introduced into dentifrice for desensitizing purposes.24 Moreover, some study confirmed uptake of F− and Sr2+ by plaque, thereby affecting metabolism of bacteria and inhibiting formation of bacterial acids25, 26 that are critical for inhibition of caries. Fluoride ions replace hydroxyl ions into CDHAp structure, thus improving mechanical properties and decreasing solubility, leading to resistance against acids secreted by bacteria in plaque.27, 28 J. Feartherstone cites three reasons why fluoride is effective against caries: (i) inhibits bacterial metabolism, (ii) inhibits demineralization at the crystal surface during an acid challenge, (iii) enhancing remineralization and thereby forming a low-solubility layer similar to the acid resistant mineral fluorapatite, on the remineralized crystals. Studies in caries research have shown that only if fluoride is introduced into a new crystal surface during remineralization, it beneficially altered enamel solubility.29 Strontium mostly is known for its ability to prevent bone resorption as well as confront osteoporosis,30 but in dentistry in the very beginning it has been used for teeth desensitizing effect as mentioned previously. Recently, Krishnan et al. observed, that incorporation of Sr2+ up to 25 mol% into HAp structure increases cell viability compared to pure HAp phase, that is important factor for well-being of tissues in the mouth cavity.31 Also, scientists from Italy noticed better mineralization of hard tissues for Sr2+ containing samples.32

Simultaneous addition of Sr2+ and F− to bioactive glasses has been studied.33-36 The findings showed, that incorporation of these two ions into bioactive glass had beneficial and synergistic effects on HAp crystallization and pH increase was observed in the research media, thus reducing loss of minerals and promoting remineralization of demineralized enamel and dentine. Moreover, beneficial antibacterial effect was observed for glass ionomers with Sr2+ and F-.30

Taking into account the studies described above, we have concluded, that there is luck of information about simultaneous incorporation of strontium and fluoride ions into CDHAp structure. CDHAp chemical and crystal structure is more similar to natural tooth enamel and dentine comparing to bioactive glasses. We were curious, is there any difference of remineralization potential among pure CDHAp and substituted CDHAp? Therefore, in current study we synthesized needle-like Sr and/or F substituted CDHAp and evaluated enamel recovery ability with the same materials. Also the possibility to protect enamel from caries and erosion caused by acidic environment was evaluated. In vitro tests, simulating application of dental care products containing substituted CDHAp nanoparticles on bovine teeth enamel were performed. Special attention was paid for in vitro remineralization equipment and experimental design. Our findings are complementary to the field of remineralizing agents used in dentifrice, toothpastes, mouthwashes, and so forth.

2 MATERIALS AND METHODS

Our research had two parts—synthesis and characterization of CaP and remineralization studies (see Figure 1). Remineralization study includes preparation of CaP pastes and in vitro studies, where bovine enamel samples were treated.

A schematic illustration of experimental methods

2.1 Preparation and characterization 2.1.1 Wet chemical precipitation

All CDHAp were synthesized trough wet precipitation method according to these parameters: Ca/P or Ca + Sr/P molar ratio between 1.50 and 1.67, theoretical Sr substitution level 10 wt% in respect to total weight of the product and theoretical F substitution level 3 wt% in respect to total weight of the product.

To obtain CDHAp, CaO was used as calcium source from which Ca(OH)2 suspension was obtained. Subsequently, neutralization reaction between 0.3 M Ca(OH)2 and 2 M H3PO4 at 45°C was realized. H3PO4 addition was done by incremental mode till end pH 7.2, then CaP suspension was stirred for 1 h and vacuum filtered.

Sr substituted CDHAp (SrCDHAp) production was done as described above and SrO was used as Sr source.

The synthesis of F substituted CDHAp (FCDHAp) had two stages: (i) production of CDHAp, (ii) pH-cycling of obtained CDHAp suspension to incorporate fluorine into the structure. Stage (ii) included addition of 0.09 M NH4F solution with following pH-cycling of obtained suspension. The pH of the suspension was lowered to 4 by addition of 2 M H3PO4 and subsequentialy raised to 7 by incremental addition of 1 M NaOH solution. Described pH cycling was repeated three times. The end stage of the FCDHAp synthesis was the washing of obtained precipitates with deionized water during vacuum filtration process.

To obtain SrFCDHAp precipitates we combined SrCDHAp and FCDHAp production techniques sequentially.

For determination of phase composition and evaluation of characteristic absorption bands part of the filtered precipitates was dried and calcined at 1100°C (1 h). Remaining wet precipitates were stored in a fridge at 4°C into the closed containers till further use for experiments.

2.1.2 Physico-chemical characterization The phase composition and characteristic functional groups of the powders were investigated using X-ray powder diffractometry (XRD, PANalytical X'Pert Pro, Cu Kα1, 40 kV, 30 mA) and Fourier transformation infrared spectroscopy (FTIR, Varian Scimitar 800, KBr pellet method at 4 cm−1 resolution co-adding 30 scans over a range of wavenumbers from 400 to 4000 cm−1). The morphology of synthesized products was observed with field emission scanning electron microscopy (FE-SEM, Tescan Mira/LMU), and high-resolution transmission electron microscopy (HR-TEM, Tecnai G20) with a field emission gun operating at 200 kV. Fluorine content was determined by elemental analysis using energy-dispersive x-ray analysis (EDX, Oxford Instruments X-MaxN detector size 150 mm2). Strontium content was analyzed by AAS (Varian SpetrAA880). Crystallite size (d) were calculated from XRD pattern using Scherrer equation37 at 2θ = 25.9°, corresponding to the 002 hkL crystal plane. Brunauer–Emmett–Teller (BET) method was used to determine specific surface area (SSA) of the precipitated powders by N2 absorption (Quadrasorb SI-KR/MP, Quantachrome Instruments). Equivalent particle diameter (dBET) was calculated using following equation: (1)where q is the true density of CDHAp powder (g/cm3) and Sw is the SSA. Density measurements of synthesized CDHAp powders were done with helium pycnometer Micro UltraPyc 1200e (Quantachrome Instruments, Boynton Beach, Florida, USA).

The evaluation of bovine enamel samples was carried out with following instrumental methods. XRD patterns were recorded using Ni filtered monochromatized Cu Kα1 radiation (λ = 0.1540 nm) generated at 40 kV and 30 mA and in the 2θ range from 5° to 70°. FTIR spectra were recorded in the attenuated total reflectance (ATR, GladiATR™, Pike technologies, USA) mode and obtained at 4 cm−1 resolution co-adding 50 scans over a range of wavenumbers from 400 to 4000 cm−1. Before every measurement, a background spectrum was taken and deducted from the sample spectrum. XRD and FTIR were taken for etched and treated enamel samples. The surfaces of demineralized and remineralized samples were observed with SEM at an accelerating voltage 15 kV and images were taken under 30,000× and 100,000× original magnification. The thickness of newly formed CaP layer on the enamel slabs were analyzed by contact type 3D profilometer (Taylor Hobson Form Talysurf Intra 50). The scanning speed was set to 1 mm/s and standard stylus 112–2009 with probe tip radius 2 μm was used.

2.1.3 Ability to increase pH of acidic media We tested ability of the nanoparticles to increase pH of acidic media, thus mimicking acidic environment in the oral cavity after a meal. The experimental set-up is shown in Figure 2. For this experiment, 5 mg of the nanoparticles were dispersed into 40 ml H2O and homogenized with sonotrode Hielscher for 10 s (amplitude 59%, 82 W). Titration station Mettler Toledo with software LabX and pH sensor was adapted to perform the experiment. Following process steps were done: Heating of the nanoparticles suspension up to 36°C (in a water bath); Addition of 10 μl of 0.05 M citric acid every 10 min (maximum 90 μl). The selected time period was chosen according to the results from preliminary experiments, in which a plateau occurred on the pH change curve after 10 min. After the addition of the last 10 μl of citric acid, pH changes of the system were observed for 30 min.

A schematic illustration of experimental equipment to test the ability of the nanoparticles to increase pH of acidic media

2.1.4 Cytotoxicity

Samples for cytotoxicity were prepared from wet precipitates. For the preparation of the test samples, the moisture content of each paste has been taken into account to obtain suspensions with concentrations 300 and 500 μg/ml.

For the assessment of cytotoxicity of the nanoparticles, cell line HGF-1 (ATCC® CRL-2014™) was used. Dulbecos modified eagle medium (DMEM, Millipore) and 10% fetal bovine serum (FBS, Millipore) was used as a cultivation media in 24 well plate. Approximately 180,000 cells were seeded in each well. Cells were cultivated for 24 h prior to expose of test material. After 48 h incubation at 37°C and 5% CO2 in the atmosphere, test material was removed and cells detached from reaction plate surface. Hemocytometer was used for cell counting. ANOVA test was applied for data statistical analysis, where p = .05.

2.2 Preparation of pastes

To evaluate ability of the nanoparticles to protect enamel hard tissue, a simple composition of pastes was created. Water, glycerol, 2-hydroxycellulose (HOECel) and relevant synthesis product were used (See Table 1). To obtain homogeneous mas, IKA T18 basic ULTRA-TURRAX homogenizer and sonotrode UP200t Hielscher (82 W, amplitude 49%, 2 min) was used to frustrate larger agglomerates and to reduce the number of air bubbles into the paste. The amounts of used components are summarized in Table 1.

TABLE 1. The composition of paste Nr. Paste designation CaP (wt%) HOECel (wt%) Glycerin (wt%) Water (wt%) 1. Rem_CDHAp 20 3 5 72 2. Rem_FCDHAp 20 3 5 72 3. Rem_SrCDHAp 20 3 5 72 4. Rem_SrFCDHAp 20 3 5 72 Abbreviations: CaP, calcium phosphates; CDHAp, calcium deficient hydroxyapatite; FCDHAp, F substituted CDHAp.

The viscosity of prepared pastes was measured with rotational viscometer RheolabQC (Anton Paar) at 5 rpm for 10 min and it was comparable with commercial toothpaste BioRepair®. According to ISO 11609:2010 the pH of all pastes was under 10.5.

2.3 Mineralization study in vitro 2.3.1 Preparation of bovine enamel blocks

Enamel blocks were prepared from bovine teeth (obtained as by-products from slaughterhouse). Bovine teeth were cleaned from soft tissue debris, then each tooth was sectioned horizontally below the cementoenamel junction using a water-cooled diamond saw (IsoMet Low Speed Saw). Soft tissue debris from pulp cavity were removed. All teeth were carefully rinsed under deionized water and stored in to 0.01% thymol solution at 4°C prior further use.

To obtain enamel blocks with surface dimensions 3 × 4 mm2, tooth crown (lingual surface) was attached to the PE plate and labial surface of the crown was cut horizontally and vertically (Figure 3).

Preparation of bovine enamel blocks

The surface of enamel blocks was prepared according to ISO 11609:2010. As the surface roughness of samples should not exceed 0.3 μm, bovine enamel blocks were grinded with sandpapers (600, 800, 1000, 1500 grit) for 5 min. Grinding procedure also removed plaque and dental calculus from the surface. Afterwards the samples were washed into the ultrasound bath for 5 min to remove dusts from the enamel surface. The roughness of surface was measured by 3D profilometry to be in conformity with ISO standard.

Before in vitro studies, enamel samples were subjected to demineralization to obtain exposed enamel tissue structure. The surfaces of specimens were immersed in the 34% H3PO4 for 15 s, then rinsed with deionized water.38

Prepared specimens were embedded into special sample holder (see paragraph b.), where a half of each enamel block surface was covered with nail varnish to serve as an untreated control.

2.3.2 Equipment for in vitro study

Custom made sample holders were prepared to ensure relatively equal experimental conditions (Figure 4). The sample holder was designed with closed cavity inside to have the ability to float. Such position of sample holder ensured that embedded enamel blocks were faced downwards, thus deposition of minerals from artificial saliva due to gravity was excluded during in vitro study. Moreover, during and after the experiment, it was not necessary to take out enamel specimens from the sample holder for surface treatment with the pastes and for physico-chemical analysis. To perform enamel surface brushing, we adapted, BUEHLER MINIMET 1000″ grinding machine by designing special frame for sample holder to fix it during treatment with the pastes (Figure 5).

Sample holder for in vitro studies: (A) holder with tooth samples, (B and C) holder immersed in the artificial saliva

The equipment for treatment of enamel blocks with the pastes

Custom made sample holder was used also for XRD analysis (Figure 6), thus enamel block was positioned in the center of the holder. It should be noted that central sample was not coated with partly masking nail varnish.

Custom made sample holder for XRD analysis. XRD, X-ray powder diffractometry

2.3.3 Treatment of enamel blocks with the pastes

The surface treatment of enamel samples was performed according to scheme shown in Figure 7.

Schematic image of enamel sample treatment

Enamel blocks were placed in the artificial saliva (0.33 g KH2PO4, 1.27 g KCl, 0.58 g NaCl, 0.17 g CaCl2, 0.2 g [NH2]2CO, 1 L deionized H2O) at 37°C. Incubator (Biosan Orbital Shaker – Incubator ES – 20) ensured dynamic condition (shelf rotation speed 100 rpm) between treatment episodes with pastes. To simulate everyday tooth brushing, following steps were performed (repeated two times a day for 7 days): Removal of the sample from artificial saliva; Sample rinsing with deionized water; Sample demineralization with 50 mM citric acid for 3 min; Sample rinsing with deionized water; Sample brushing with the pastes for 3 min with applied toothbrush force of 0.2 N per 1 cm2; Sample rinsing with deionized water; Incubation of the sample in to the container with artificial saliva.

All experimental groups are summarized in Table 2.

TABLE 2. Experimental groups Sample group Comment Rem_Dynamic Treatment with Rem_CDHAp Twice a day for 2 min 5 days Profilometry every day Rem_Control Treatment without paste Twice a day 2 min 7 days Rem_CDHAp Treatment with Rem_CDHAp Twice a day 2 min 7 days Rem_SrCDHAp Treatment with Rem_SrCDHAp Twice a day 2 min 7 days Rem_FCDHAp Treatment with Rem_FCDHAp Twice a day 2 min 7 days Rem_SrFCDHAp Treatment with Rem_SrFCDHAp Twice a day 2 min 7 days Abbreviations: CDHAp, calcium deficient hydroxyapatite; FCDHAp, F substituted CDHAp.

After a set of in vitro experiments, nail varnish was removed and both sides of enamel surface investigated.

3 RESULTS

Table 3 presents data obtained from XRD, BET, and EDX analyses. All synthesis products after thermal treatment are biphasic mixtures composed of HAp and β-tricalcium phosphate (β-TCP). According to EDX results, only 50% of added fluoride and 74%–78% of strontium were found into the FCDHAp, SrFCDHAp and SrCDHAp samples. The highest SSA values were measured for SrCDHAp following by CDHAp, SrFCDHAp, and FCDHAp. We observed that fluoride incorporation into CDHAp structure results in lower values of SSA compared to CDHAp and SrCDHAp powders. Calculated particle size of all synthesis products ranges from 22 to 29 nm and are close to calculated crystallite size (d002).

TABLE 3. Characteristics of synthesized nanopraticles FCDHAp SrCDHAp SrFCDHAp CDHAp Ca/P ratio 1.64 1.65 Ca + Sr/P ratio 1.64 1.64 HAp/β-TCP, wt %a 86/14 82/12 84/16 89/11 F, wt % 1.6 ± 0.1 – 1.5 ± 0.2 – Sr, wt % – 7.4 ± 0.4 7.8 ± 0.3 – SSA, m2/g 71.1 ± 0.2 94.5 ± 1.3 78.1 ± 0.4 82.3 ± 0.8 Particle size dBET, nm 29.7 ± 0.1 22.6 ± 0.3 26.5 ± 0.1 25.3 ± 0.2 Crystallite size d002, nm 30 27 27 27 Abbreviations: CDHAp, calcium deficient hydroxyapatite; FCDHAp, F substituted CDHAp; HAp, hydroxyapatite; SSA, specific surface area. 3.1 Physico-chemical characterization

FTIR showed characteristic absorbance bands of (PO4) and (OH) groups for all synthesized powders. Additionally, the spectra showed absorbance bands characteristic to (HPO4) and (CO3) functional groups. The presence of (CO3) absorbance bands is attributed to atmospheric CO2 in the synthesis environment and its inclusion in the product structure. In turn, the presence of (HPO4) confirms that the products are CDHAp.39 In the FTIR spectra of the thermally treated powders the respective absorbance bands were not detected, while extra maximums at 3543–3547 cm−1 and 735 cm−1 were detected for FCDHAp, and SrFCDHAp spectra.

XRD patterns of the synthesized powders showed crystalline structure with low crystallinity, as evidenced by relatively broad and low-intensity XRD peaks. Diffraction peaks of the powders became sharper and with higher intensity after thermal treatment at 1100°C, indicating increase of crystallinity. A slight shifting of maximums toward smaller 2θ angle were observed for Sr containing powders due to Sr incorporation into CDHAp structure.

SEM and TEM evaluation (Figure 8) shows needle-like morphology for all synthesized particles. In addition, TEM investigation reveals that observed particles actually were agglomerates consisting of smaller units and nanoparticles morphology was close to natural tooth crystals 50–70 nm in width and 20–25 nm in thickness.40

Morphology of the synthesized powders; left—SEM image, right—TEM image. SEM, scanning electron microscopy; TEM, transmission electron microscopy

3.2 Ability to increase pH of acidic media

Figure 9 represents the ability of nanoparticles to balance the pH of aqueous systems after acid addition. In all experiments, a steep decrease of pH after addition of 50 mM citric acid was observed. In case of enamel sample, pH decreased by 2.5 units; this means, the critical pH level was passed (pH ≤5.541) and enamel started to dissolve irreversibly, but for synthesized nanoparticle suspensions, pH decreased only by 1.0–1.2 units. After 10 min we observed uplift of the curves, thus indicating a slight compensation effect of pH drop in the aqueous systems. Comparing all four suspensions, relatively higher pH values was maintained for SrCDHAp, but the lowest pH was for SrFCDHAp suspension.

The function of pH in time for suspensions after incremental addition of acid

3.3 Cytotoxicity

The results of cytotoxicity studies showed that among various concentrations of the nanoparticles suspensions, a significant reduction of gingival fibroblast cell viability occurred in the case of SrCDHAp and FCDHAp samples. CDHAp and SrFCDHAp suspensions in various concentrations do not significantly affect the cell viability. Cytotoxic samples are marked with the “#” symbol in Figure 10.

Results of cytotoxicity studies of the nanoparticles in different concentrations: (A) 500 μg/ml, (B) 300 μg/ml; *p < .05

3.4 Remineralization in vitro 3.4.1 Fourier transformation infrared spectroscopy

Figure 11 summarizes FTIR spectra from all enamel samples and spectrum of 2-hydroxyethyl cellulose, as it was used as component for the pastes. The peaks and broad bands in the HOCel graph are attributed to (OH) (3000–3600 cm−1), (CH) (2900–2999 cm−1 and 1249–1454 cm−1), (CO) (1030 cm−1), and adsorbed H2O from atmosphere (1650 cm−1)42-44 FTIR data of enamel samples showed major chemical groups of CaP. The absorbance bands of (PO4) group were detected from 472 cm−1 to 599 cm−1 for etched enamel, control and all treated samples. The overlap of (PO4) and (HPO4) vibrations occurred from 720 cm−1 to 1160 cm−1, while characteristic vibration of (OH) group was recorded in the range 630–636 cm−1 and at ≈3565 cm−1 (Figure 12B; indicated with arrow). The absorbance bands of (CO3) at 1410–1450 cm−1 were detected for all enamel samples. Additionally, absorbance bands of amide groups were found from 1560 cm−1 to 1530 cm−1 for remineralized samples. Bending of (NH) group and stretching of (CN) group came from tooth enamel organic component.45 A small peak at 912 cm−1 (Figure 12A) for Rem_control. According to previous studies,37, 46 this place on the graph is attributed to (P-OH) absorption mode originating from octa-calcium phosphate (OCP).

FTIR spectra of remineralized samples, control sample, etched enamel, and hydroxyethyl cellulose. FTIR, Fourier transformation infrared spectroscopy

The zoomed in FTIR spectra: (A) OCP (800–990 cm−1) and (B) (OH) (3200–3800 cm−1) detection. FTIR, Fourier transformation infrared spectroscopy; OCP, octa-calcium phosphate

3.4.2 X-ray powder diffractometry

XRD patterns were similar for all enamel samples. The only difference was observed for peak intensities. However, one diffraction peak at 32.86° 2θ (HAp, ICDD 01–072-1243) for treated enamel samples was detected, but not for etched enamel. It should be noted that peaks have slight deviation compared to HAp record in the database. For Rem_CDHAp sample it was at 33.31° 2θ, for Rem_Control—32.89° 2θ, for Rem_SrFCDHAp—32.85° 2θ, for Rem_SrCDHAp—32.91° 2θ and for Rem_FCDHAp—32.93° 2θ. One additional peak was found for Rem_control at 4.8° 2θ, which is characteristic for OCP phase. (Figure 13)

XRD patterns of the treated enamel samples. XRD, X-ray powder diffractometry

3.4.3 Scanning electron microscopy

SEM images (Figure 14) clearly showed morphological differences between treated and untreated enamel surfaces. A layer of plate-like crystals on the Rem_control surface was observed. According to XRD, FTIR results and literature,47, 48 the observed crystals are typical for OCP phase. Similar crystals were found in the hollows of Rem_SrCDHAp sample surface, although neither XRD nor FTIR detected OCP phase, because of proportionally small phase amount present on the sample.

(A) SEM micrographs of Rem_control, (B) Rem_CDHAp, (C) Rem_FCDHAp, Rem_SrFCDHAp enamel samples. CDHAp, calcium deficient hydroxyapatite; FCDHAp, F substituted CDHAp; SEM, scanning electron microscopy

New CaP layer formed on Rem_CDHAp, Rem_FCDHAp, Rem_SrCDHAp, and Rem_SrFCDHAp samples consisted of chaotically oriented rod-like CaP crystals. In the case of Rem_SrFCDHAp, there was evidence for crystal orientation (Figure 14E). Nonetheless, this observation has to be approved by further experiments.

3.4.4 Energy-dispersive x-ray analysis

Along with SEM observation, EDX analysis was performed. The presence of F was detected for samples brushed with FCDHAp and SrFCDHAp paste, but Sr was found for samples brushed with SrCDHAp and SrFCDHAp paste (Table 4).

TABLE 4. EDX results for remineralized enamel samples F, wt % Sr, wt % Rem_Control EE –