Synthesis and Characterization of Novel Calcium-Silicate Nanobioceramics with Magnesium: Effect of Heat Treatment on Biological, Physical and Chemical Properties

Abstract

Glass-ceramic nanopowder with a composition of 55SiO₂-35CaO-10MgO (mol %) was synthesized by the sol–gel method and was heat treated at three temperatures (T1 = 835 °C, T2 = 1000 °C, T3 = 1100 °C) in order to obtain different materials (C1, C2, C3, respectively) varying in crystal structure. Bioactivity and oxidative stress were evaluated in simulated body fluid (SBF) for various time periods (up to 10 days). The structure of the synthesized materials and their apatite-forming ability were investigated by X-ray diffractometry (XRD), Fourier transform infrared spectroscopy (FTIR), Scanning Electron Microscopy and Energy Dispersive Spectroscopy (SEM/EDS). The antibacterial properties of the synthesized materials were evaluated against three Gram-positive and four Gram-negative bacterial strains and their biocompatibility was verified on a primary cell line of human gingival fibroblasts (HGFs) by the MTT (3-[4, 5-dimethylthiazol-2-yl]-2, 5 diphenyl tetrazolium bromide) assay. The crystallization of the materials was increased by sintering temperature. Heat treatment did not inhibit the bioactive behavior of the materials as apatite formation started after 3 days in SBF. C2, C3 showed some indications of apatite forming even from the first day. Regarding cell viability, a variety of biological behaviors, concerning both dose and time points, was observed between the positive control and the tested materials by both the MTT assay and oxidative stress analysis. In conclusion, the nanobioceramic materials of this study possess a multitude of attractive physicochemical and biological properties that make them suitable candidates for bone regeneration applications, fillers in nanocomposite scaffolds, or as grafts in bone cavities and periodontal lesions.

1. Introduction

Bioactive ceramics and glass-ceramics are a unique group of synthetic materials that react with biological fluids, presenting enhanced capability of biointegration [1,2,3,4,5,6]. Numerous studies, both in vitro and in vivo, have demonstrated the ability of these materials to bond with living tissues, especially bone, through the development of a surface apatite layer [7,8,9,10,11,12,13]. Their physical, chemical, biological, and mechanical properties are of paramount importance and determine their suitability for various biomedical applications [14,15,16,17,18,19], for example, in bone regeneration and dentine bonding of root filling materials.

Bioceramic nanoparticles have attracted the interest of the scientific community because of their advantages and attractive properties, including high surface area, green catalytic capability, green adsorbents capability, targeted drug delivery potential, and non-hemolytic or even antibacterial properties [20,21,22,23,24,25,26]. There is a growing body of literature aiming to evaluate the toxicity of nanomaterials and correlate it with specific nanoparticle properties. The biocompatibility and cytotoxicity of biomaterials are key issues that must be addressed at a pre-clinical stage, before biomedical applications [27,28,29,30]. Research has shown that the smaller the size of the bioactive glass particles (nanoparticles), the higher their bioactivity and the more suitable they are for biomedical applications. However, different types of silica-based nanomaterials, such as fumed silica, non-porous silica, and porous silica materials, have different cytotoxicity responses, with the last ones exhibiting the lowest values in a broad range of cell types [31,32,33,34,35,36,37,38].

The results of recent in vivo studies with bioactive glass nanoparticles indicated that these materials do not produce toxic shock or systemic toxicity and inflammation, and exhibit antimicrobial properties [21,39,40,41,42]. One of the most common tests for the evaluation of biological responses, according to ISO 10993, is the in vitro test of cellular toxicity. Many research groups characterize cell cytotoxicity as the initial step in determining whether or not a material’s biocompatibility is achieved [33,43,44].

Bioactive glass-ceramics can be synthesized with either melt-derived or sol–gel methods. The sol–gel method is commonly used for the synthesis of ceramic nanoparticles. It is an inorganic polymerization reaction based on a dual hydrolysis/polycondensation mechanism that can be accordingly modulated for the desired result. The process involves the use of alkoxide precursors diluted in a solvent (i.e., alcohol) in the presence of water. By monitoring the basic parameters affecting the formation of nanoparticles, such as the H₂O/alkoxide ratio, sintering temperature and time, aging temperature, and drying conditions, it is possible to produce either a sol having a colloidal nanoparticle dispersion or a gel where the macromolecule that spreads throughout its volume is formed [17,45,46,47]. This method results in materials with high surface area, which is correlated with an increase in the growth rate of apatite layer and better bioactivity.

There are two dominant categories in ceramic biomaterials, calcium-phosphate ceramics, and calcium-silicate ceramics, which have comparable biological properties. Calcium-silicate ceramics have shown higher bioactivity than calcium phosphates and fast apatite-forming ability [48,49]. They have also been associated with enhanced proliferation and differentiation of various cell types [50,51]. Although calcium silicates have similar properties to calcium-phosphate ceramics, they are more soluble, which causes cell death through significant increase in pH and alkaline-induced toxicity [52,53].

Magnesium—an important trace element of bone and extracellular matrix (ECM)—is closely linked to cellular differentiation and calcification of hard tissues and it also has an indirect effect on metabolism. The addition of magnesium to calcium-silicate systems affects their mechanical properties, reduces pH, accelerates bone formation, directly stimulates osteoblast proliferation, and enhances cellular differentiation [54,55,56,57]. Consequently, calcium-silicate nanomaterials containing Mg possess superior mechanical properties, enhance bone formation, and lead to higher expression of osteogenic differentiation gene markers [58,59,60,61,62,63,64,65].

The objective of the present study was to synthesize bioceramic nanomaterials in the ternary SiO₂-CaO-MgO system through the sol–gel method and to evaluate their in vitro bioactivity, biocompatibility, and antibacterial properties after thermal treatment at various temperatures.

2. Materials and Methods

2.1. Glass-Ceramic Synthesis of Bioceramic Nanopowder

In the present study, the Stöber-based sol–gel method that was developed for the synthesis of silica nanoparticles was used to synthesize a bioceramic nanopowder of the composition 55SiO₂-35CaO-10MgO (in mol %) by an acid-based catalyzed hydrolysis of TEOS [66,67,68,69,70]. In brief, silicon oxide in the form of tetraethyl orthosilicate (TEOS) solution was added in a mixture of d.d.H₂O, ethanol and 2M nitric acid (HNO₃). Then, while maintaining continuous magnetic stirring at 400 rpm and after the completion of the hydrolysis of TEOS, the remaining reagent-soluble calcium and magnesium nitrates (Ca(NO₃)₂·4H₂O, Mg (NO₃)₂·6H₂O) were added in the mixture. Mechanical stirring was applied for 1 h at room temperature, and then the beaker containing the mixture was placed in a conventional ultrasonic bath, whereupon the 2M ammonia solution was added dropwise, at a rate of 1 mL/min. Gelation was achieved at around 14 h after the addition of ammonia by the conversion of the solvent to a soft gel. The mixture was dried at 75 °C for 2 days to complete the aging stage and receive a dry gel. In order to decompose the formed agglomerates, the synthesized material was subjected to wet mechanical ball milling after the sintering process [71,72,73,74,75,76,77]. The milled material was then dried in a low temperature oven at 30–40 °C and stored in a desiccator.

2.2. Morphological and Structural Characterization of Synthesized Bioceramic Nanopowder

2.2.1. Differential Thermal and Thermogravimetric Analysis (TG-DTA)

The thermal behavior and properties of the synthesized powder were analyzed using differential thermal analysis and thermogravimetric analysis (TG-DTA). A quantity of 36.5 mg of the initial nanopowder was heated from 26 °C to 1380 °C with a heating rate of 10 °C/min, using a Setaram SETSYS 16/18 (1750 °C rod) device. All measurements were performed in alumina crucibles and under air flow (50 mL/min).

Based on the results of the thermal analysis performed after the aging stage, and, in particular, on the characteristic crystallization temperature (T_c), glass transition temperature (T_g), and peak temperature (T_c, p_k), the initial powder was divided into three isomeric quantities, and each one was heated to a different temperature in the region of 800–1100 °C [78]. The three temperatures selected for further study were: 835 °C (C1), which corresponds to the beginning of the crystallization process, 1000 °C (C2), which corresponds to the end of the crystallization process, and 1100 °C (C3), which corresponds to the temperature where the ternary system exhibits no further crystallization. All the samples were heated from room temperature with a heating rate of 10 °C/min under air and then allowed to cool to room temperature.

2.2.2. Structural Characterization

The synthesized materials were in the nanoscale level as evidenced by Dynamic Light Scattering (DLS) measurements [66]. Structural characterization of the synthesized bioceramic nanopowder was carried out by X-ray diffractometry (XRD) and Fourier transform infrared spectroscopy (FTIR), while their morphology was evaluated with Scanning Electron Microscopy and Energy Dispersive Spectroscopy (SEM-EDS). All the FTIR spectra were obtained with the KBr pellet technique using a Perkin-Elmer FTIR spectrometer, model Spectrum 1000, in the MIR spectral region (4000–400 cm⁻¹), with a resolution of 4 cm⁻¹ and 32 scans. The XRD measurements were performed using a two-cycle Rigaku Ultima⁺ X-ray diffractometer (operating at 40 kV/30 mA, with CuK_α radiation, in Bragg–Brentano geometry) and the operating conditions were: step 0.02°, count 2.5 s/step, in the angle range of 5–75°. The XRD patterns were identified with JCPDS-ICDD. Open-source FullProf Suite program (version January 2021) was used for the Rietveld analysis [79]. The morphological characterization and evaluation, and the stoichiometric composition of the samples were evaluated with SEM-EDS using a JEOL electron microscope, model JMS-6390LV, equipped with an Oxford INCA Energy Dispersive Spectroscope. For the microanalysis, the samples were coated with a carbon layer of about 200 Ǻ to achieve conductivity of the materials under study.

2.3. In Vitro Apatite-Forming Ability in Simulated Body Fluid

The in vitro apatite-forming ability of the heat-treated bioceramic powders was evaluated by immersion in a conventional simulated body fluid (c-SBF) under renewal conditions [69,80]. The bioceramic powder mass to the liquid volume ratio of the solution was constantly kept at 1.5 mg/mL [80]. The replacement of the SBF solution was performed 6 h after the first immersion, then at 24 h (1 day) and then every other 2 days (0, 1, 3, 5, 7, and 10 days). All samples remained at a constant temperature of 37 °C in an incubator. After each time point, the samples were centrifuged and dried in a low temperature incubator at 30–40 °C. All experiments were performed in triplicate.

2.4. Preparation of Cell Lines of Human Fibroblasts (Human Gingival Fibrοblasts, HGFs)

Primary cultures of HGFs were established from gingiva biopsy of a healthy donor during the extraction of premolars or third molars, after approval by the Ethical Committee of the Institutional Review Board (#11/23 November 2017 Absence of inflammation and root caries were the necessary conditions in order to perform the biopsy. More specifically, small pieces (1–2 mm) of gingival tissue were placed in 25 cm² culture flasks with Dulbecco’s modified essential medium (DMEM, Biochrom, Cambridge, UK) supplemented with 10% fetal bovine serum (FBS, GibcoBRL) and antibiotics/antimycotics (penicillin, streptomycin, amphotericin B (GibcoBRL, Grand Island, NY, USA). Cell culture was performed in a sterilized incubator at 37 ± 1 °C and 5% CO₂, 95% atmospheric pressure, and 100% humidity. When the flask became 80–90% confluent, cells were transported by trypsinization (solution trypsin 0.25% 1 mM EDTA solution (GIBCO/Invitrogen)) to a larger flask of 75 cm². HGFs from 3rd to 6th passages were used.

Mitochondrial Activity—MTT Assay

In order to measure the cell viability, the indirect biocompatibility test of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) was performed. Eluates were collected after 24 h incubation time with the nanoparticles at a concentration of 1 mg/mL in Dulbecco’s minimal essential medium, DMEM. After centrifuging, the supernatant was collected and passed through 0.22 µm filters. More specifically, cells of HGFs (3 × 10⁴ cells per well) were seeded in 96-well plates to allow cell attachment. After 24 h of cell culture, DMEM was removed and replaced with eluates of the three tested materials at two concentrations (125 and 250 μg/mL). HGFs cultured in DMEM supplemented with FBS and antibiotics served as positive control, while HGFs cultured in DMEM without FBS served as negative control. Analysis of mitochondrial activity and thus cell proliferation was succeeded by measuring the mitochondrial dehydrogenase activity of living cells verified by the capability of transforming the yellow tetrazolium salt into blue formazan crystals. After 1, 3 and 5 days, the culture medium was removed from each well, followed by washes with PBS, and then DMEM was placed in each well. Following that, MTT solution was added in each well (10% of the total volume per well) and 3 h incubation at 37 °C and 5% CO₂ followed. After this period, the medium containing the MTT solution was discarded, and the insoluble formazan was dissolved with dimethyl sulfoxide (DMSO). Cell viability was evaluated by measurement of optical density with an ELISA-spectrophotometer (Epock, Biotek) at double wavelength (at 540–630 nm). The experiments were performed in triplicate. Statistical analyses were performed using Microsoft Excel 2007. Independent sample t test was used to compare between means. In all statistical analysis, level of significance (p-value) was set at α = 0.05. The level of statistical significance was 0.05 (p < 0.5).

2.5. Antibacterial Assay

Seven bacteria were used in the experiments: Bacillus cereus, Staphylococcus aureus ATCC 6538, Listeria monocytogenes Scott A, Escherichia coli O157:H7 (non-toxigenic) NCTC 12079, Salmonella enterica serovar Typhimurium, S. enterica serovar Enteritidis PT4, and Pseudomonas aeruginosa. Cultures were resuscitated from frozen stocks by streaking onto plate count agar (PCA) plates. To prepare cultures for the experiments, 10 mL of nutrient broth (NB) was inoculated with bacterial culture from a PCA plate and incubated for 24 ± 2 h at 37 °C. A 100 μL aliquot of this culture was then used to inoculate fresh NB (10 mL), which was then incubated for 24 ± 1 h at 37 °C. A 1 mL aliquot of this broth culture was centrifuged at 13,000× g for 5 min and the supernatant discarded. The cell pellet was resuspended in 1 mL Ringers diluent solution and the resulting suspension diluted 1 in 10 with Ringers solution. The viable bacteria in the inocula were enumerated by colony counting on PCA (37 °C, 24 h).

2.5.1. Growth Experiments

Growth experiments were carried out using a Bioscreen C instrument, which measures optical density changes in liquid media in microtiter plate wells. Each well contained 120 μL NB of 2.5× normal concentration (NB2.5), 150 μL 1 g/L test compound (control: sterile distilled water), and 30 μL bacterial suspension (sterility control: sterile distilled water). The final concentration of the antimicrobial compounds in the test wells was therefore 0.5 g/L. Plates were loaded into the Bioscreen C, which was set at 37 °C with optical density measurements at 600 nm every 30 min for 24 h.

2.5.2. Statistical Data Analysis of the Bacterial Cultures

All experiments were performed in triplicate and repeated once, giving six replicates in total. At the end of the experiment, the collected data were exported to Microsoft Excel for processing. For each triplicate measurement, the data were adjusted for initial absorbance by subtracting the mean of the time zero values from all measurements. Semi-log graphs were plotted of optical density (OD) against time, the period of maximum exponential growth was selected and the growth rate over the selected period was calculated using the formula

$$Growth Rate=\frac{ln\left(OD{t}_2/OD{t}_1\right)}{t_2-t_1}$$

(1)

where t₁ is the first time point of the selected exponential growth period and t₂ is the last. The growth rates of the controls and the test wells in each individual Bioscreen run were normalized by dividing each individual value by the mean of the three control values. Data from the different runs were then combined, and the normalized growth rates in the test wells were compared with those in the control wells. Lag phases were approximated by determining the time before an increase in absorbance (of at least 0.01 absorbance units) was observed at the start of growth.

2.6. Fluorescence Analysis for the Detection of ROS Levels

For the detection of intracellular reactive oxygen species (ROS) levels generated during cell culture with the eluates of nanobioceramics, we employed the cell-permeable ROS-sensitive probe Dihydroethidium (DHE). Measurement of fluorescence was performed on the supernatant received from cell culture wells before the addition of DMSO at each time point (1, 3, and 5 days) at two concentrations (125 and 250 μg/mL) for the three tested materials. The received supernatant from cell culture wells without tested materials was used as control. For fluorescent measurements, an excitation wavelength of 480 nm and emission wavelength of 576 nm were used. Oxidation of DHE 5 µM (incubated for 1 h) in supernatants was monitored by measurement of the fluorescence in 96-well black-walled microplates (Corning^®, Sigma Aldrich, Merck KGaA, Darmstadt, Germany) using a TECAN Trading ASC Fluorescence Multimode Microplate Reader. The relative fluorescence is expressed as “% maximal emission” as determined with the software “Magellan”.

3. Results

3.1. Morphological and Structural Characterization of Synthesized Bioceramic Nanopowders

3.1.1. Thermogravimetric Results

The thermal behavior of the synthesized material is depicted in Figure 1.

The TG-DTA analysis performed on the material under study (Figure 1) showed a total mass loss of 70% up to 835 °C, which occurred in four steps:

(a)

25–190 °C: Mass loss of about 31.6%, which was associated with an endothermic peak at 142 °C. It was attributed to the loss of adsorbed water and the degradation of HNO₃. As reported in the literature, there are two forms of water “trapped” in the crystalline cell structure of the sol–gel-based nanopowder when it is in the gel state: the unbound water inside the gel structure and water [81,82,83].

(b)

190–333 °C: Mass loss of about 13%, which was associated with an endothermic peak at 276 °C. It was attributed to incineration of organic residues of ethanol and acetone [84]. In addition, in this temperature range, the decomposition of the chemically adsorbed water was attributed to the decomposition of the polycondensation reaction.

(c)

333–640 °C: Mass loss of about 23.1%, associated with an endothermic peak at 560 °C. According to the literature, it is attributed to the decomposition of calcium nitrate tetrahydrate, which is first converted to anhydrous calcium nitrate Ca(NO₃)₂ (eliminating H₂O at about 350 °C) and then it decomposes into CaO and NO₂ [85,86].

(d)

640–835 °C/1000 °C: Mass loss of 1.4% due to dehydroxylation [81,83], achieved above 850 °C. When the viscous flow began and the hydroxyl groups reacted, the particles came closer together on the surfaces of the gels, thereby eliminating the inner pores and gaps. This was consistent with the Heat Flow curve, since there was no indication of a decomposition process, while at the same time there was no significant loss of mass at that particular temperature range and above.

The results of the above study showed a large mass loss (69%) up to 835 °C. This was due to the fact that the thermal behavior of the material was measured after the aging stage of 75 °C but before the drying stage (usually up to 750–800 °C). For this reason, the material had “retained” large quantities of bound and unbound H₂O, water vapor, gases, nitrates, and other byproducts created during its preparation process, resulting in a greater evaporation rate at 75 °C.

The Heat Flow curve showed a double exothermic peak, which started at 835 °C and ended at about 970 °C, with the most intense one located at 945 °C. This double peak was attributed to multiple crystallization mechanisms, such as these of enstatite, bredigite, and wollastonite [87,88]. This was supported by the FTIR and XRD analysis of samples C2 and C3, as discussed in the respective paragraph below. In addition, a low intensity exothermic peak observed at 1053 °C was probably attributed to poor crystallization. Finally, at 1336 °C an intense endothermic peak was observed with zero mass loss, which was attributed to melting of the material. It is important to mention that no further mass loss was observed from 835 °C (initiation of crystallization) until the end of the thermal measurements, suggesting a material with stable structure, without significant mass fluctuations.

3.1.2. FTIR Analysis of Heat-Treated Bioceramic Materials

The obtained FTIR spectra before immersion in SBF are presented in Figure 2.

All three materials, regardless of their heat treatment temperature, exhibited the characteristic broad bands of silicate glass [89]. Specifically, the FTIR spectra of the studied materials revealed two strong and broad bands at 1200–900 cm⁻¹ with a shoulder at about 1220 cm⁻¹—which were attributed to the asymmetric stretching vibrations of Si–O–Si [90]—and at 580–400 cm⁻¹—which were attributed to the rocking vibration of the O–Si–O bond [91,92,93,94]. Additional common features in the FTIR spectra of the three materials were: (a) a weak band at 1640 cm⁻¹ and a wide one at 3600–3200 cm⁻¹, which were mainly associated with the bending and stretching vibrations of the O–H bond, respectively, owing to the presence of moisture in the sample; (b) a weak band at 3780 cm⁻¹, which was attributed to a hydroxyl group of non-bridged silanoles (single silanols); and (c) a very weak and wide band in the spectral area of 1500–1400 cm⁻¹, which was attributed to the asymmetric C–O stretching vibration of the CO₃⁻² group.

The FTIR spectrum of C1 was attributed mainly to a silicate glass while the spectra of C2 and C3 revealed new sharp bands that were indicative, as expected, of the creation of crystalline phases. More specifically, for C1, except for the characteristic wide silica bands, the weak bands at 930 and 470 cm⁻¹ were indications of calcium and/or calcium-magnesium silicate phases [58,95,96,97]. For C2, the main crystalline phase identified by the FTIR spectrum was that of diopside (CaMg(Si₂O₆)), because of the bands at 958, 670 and 632 cm⁻¹. Also, the bands at 690 and 745 cm⁻¹ indicated the contribution of magnesium silicate phases such as enstatite (MgSiO₃). Finally, for C3, the main crystalline phase identified was that of wollastonite (CaSiO₃) owing to the bands at 896, 681, 642 and 565(shoulder) cm⁻¹.

3.1.3. XRD Results

The XRD patterns are presented in Figure 3.

From the XRD patterns presented in Figure 3 we observed that there were differences between the samples as they progressively turned from almost amorphous (C1) to more crystalline (C2, C3). Concerning C1, except for the calcium siliceous phases, the Mg incorporation into the silicate lattice was confirmed by the XRD peaks in those areas where broad but characteristic peaks appeared. These peaks were attributed to diopside (#71-1497), bredigite (#36-0399), calcium silicate (#33-0303, #45-0156), and calcium-magnesium silicate (#34-1350, #89-1484). The diffraction pattern of C2 showed evidence of more crystalline phases, such as diopside (#86-0932, #78-13, #76-0524), enstatite (#76-0524), merwinite (#35-0591), hillebrandite (#42-0538), calcium-magnesium silicate (#34-1350, #34-1216, #72-1498, #72-1499), and calcium silicate (#49-1672, #33-0303). Concerning C3, the two main crystalline phases were those of wollastonite (#42-0547, #72-2297, #72-2284) and bredigite (#36-0399). Moreover, as observed in the previous samples, there were more clear indications of diopside (#83-1818, #71-1494), calcite (#83-1762), and magnesium silicate (MgSiO₃) (#34-1350, #87-2037), which again verified the incorporation of Mg in the calcium-silicate network. The percentage of amorphous phases was calculated to be 78.5% for C1, 38.4% for C2 and 8.8% for C3. Some of the identified phases were not verified from the FTIR analysis, as the synthesized materials consisted mainly of calcium-silicate compounds, which presented a high overlapping in the MIR area.

3.2. In Vitro Apatite-Forming Ability in Simulated Body Fluid (SBF)

The results of the FTIR analysis of the samples after immersion in SBF are presented in Figure 4. After the first day of immersion in SBF, the spectrum of C1 remained unchanged, whereas the spectra of C2 and C3 exhibited a weak band at ~580 cm⁻¹, indicating the formation of a precursor calcium-phosphate phase (Ca-P). After the third day of immersion, the characteristic hydroxyapatite (HCAp) bands of bending vibration of phosphate group at 604 and 566 cm⁻¹ were revealed, while in all samples a new weak band at 800 cm⁻¹ was presented, attributed to the symmetrical stretching vibration of the Si–O–Si bond formed by the condensation of silanol groups. The above characteristic bands, attributed to the formation of HCAp, increased in intensity and became more distinct after increasing the immersion time in SBF, indicating that the formation of HAp was time-dependent for all samples. After 10 days of immersion, the sharpening and the shifting of the broad band at 1100–900 cm⁻¹, especially in the spectrum of C1, indicated the formation of a thick HAp layer.

This was further confirmed by the following XRD patterns of the samples after their immersion in SBF for 0, 3, 5, and 10 days (Figure 5), and SEM images collected after the immersion in SBF for 0 day (Figure 6) and 3, 5, and 10 days (Figure 7). The gradual intensity of the hydroxyapatite peaks—which increased at around 32° 2θ—was indicative of its progressive formation on the surface of the samples, while its onset started at 72 h for all nanobioceramics. After 72 h of immersion, the high amount of amorphous phase was decreased in all the samples, while the HAp and calcium-phosphate percentage calculated by Rietveld analysis increased. More specifically, concerning C3, all the identified crystalline phases of calcite, magnesium silicate, diopside and wollastonite were gradually decreased as hydroxyapatite crystallized. Hydroxyapatite percentages were 16.0% at the 3rd day of immersion, 71.1% at the 5th, and 76.1% after 10 days of immersion. Similarly, for C2 at the 3rd day of immersion hydroxyapatite was 29.6%, at the 5th day 32.6%, and at the 10th day 80.1%. Concerning C1, the percentages of hydroxyapatite varied from 64.3% at the 3rd day to 78.8% at the 5th day and 83.4% for the 10th day.

3.3. Mitochondrial Activity-MTT Assay

MTT assay results confirmed nontoxic biological behavior in most cases (Figure 8). Although at day 1 a decrease in cell proliferation was observed, indicating a mild cytotoxicity even at the lower concentration, at days 3 and 5 this result was compensated. At day 3 a statistically significant increase in mitochondrial activity (p < 0.01) and consequent cell proliferation was presented for C1, while a minor increase was reported for C2 and C3. The MTT assay performed with higher concentration eluates (250 μg/mL) revealed a nontoxic biological behavior for all tested materials as verified by the comparable values between tested and control groups. At day 5, an increase in cell proliferation was reported for the C1 group for both concentrations, while C2 presented comparable values to control. A statistically significant decrease in cell proliferation for the C3 higher concentration group compared to the C3 lower concentration group was recorded (p < 0.01) at the two examined time points (day 3, day 5), indicating a dose-dependent cytotoxicity in this group. Overall, we support that the MTT assay results confirmed a nontoxic biological behavior for all tested bioceramic materials (C1, C2, and C3) at the lower concentration (125 μg/mL).

3.4. Antibacterial Assay

The results of the growth experiments for C1, C2, and C3 are presented in

Table 1. Effect of compounds C1, C2, and C3 (0.5 g/L) on the maximum growth rate of seven food-related bacteria as determined by optical density measurements in the Bioscreen C instrument.

	Compound C1		Compound C2		Compound C3
Bacterium	Ratio Test: Control	t Test 1	Ratio Test: Control	t Test 1	Ratio Test: Control	t Test 1
Bacillus cereus	0.86	0.001	0.87	0.001	0.87	0.001
Staphylococcus aureus	0.83	0.049	0.88	0.096	0.86	0.071
Listeria monocytogenes	1.01	0.932	1.00	0.987	0.90	0.165
Escherichia coli	0.97	0.238	1.03	0.226	0.94	0.071
Salmonella enterica ser. Typhimurium	0.79	0.095	0.91	0.011	0.97	0.235
Salmonella enterica ser. Enteritidis	0.90	0.013	0.81	0.000	0.87	0.000
Pseudomonas aeruginosa	0.93	0.000	0.89	0.002	0.97	0.234

¹ t test of normalized control data versus normalized test data.

. Maximum growth rates are given in the form of the ratio between the test compound and the control, rather than as absolute values.

A large degree of inhibition was not observed in any of the bacteria when challenged with any of the three antimicrobial compounds. Statistically significant (p < 0.05) decreases in growth rate were observed in B. cereus and S. enterica ser. Enteritidis with all three compounds, and in S. aureus, S. enterica ser. Typhimurium and P. aeruginosa with one or two of the compounds. However, the largest reduction in growth rate was to 83% of the control value, so the biological significance of the small degrees of inhibition observed is very limited. Large significant increases in lag phase or decreases in maximum optical density were not observed with any combination of bacteria and antimicrobials (data not shown).

3.5. Oxidative Activity of Bioactive Glass Nanoceramics

To gain further knowledge about the role of ROS production in gingival human fibroblasts after contact with the synthesized nanomaterials, their oxidative activity was determined by the fluorescent method. In Figure 9, a significant increase in ROS production after treatment with all the tested nanomaterials was observed when compared with the untreated cells (control) (p < 0.01).

In details, Figure 9A shows that C1 induced the higher increase in ROS in the lowest tested concentration (125 μg/mL) (p < 0.01) after 3 days of incubation while C2 and C3 did not present any significant difference between the tested time points. Furthermore, the presence of ROS was observed at all time points; however, fluctuations among different materials and incubation times, following the distinct trend on MTT analysis, were observed.

4. Discussion

In general, silica-based bioactive materials present high biocompatibility, bioactivity, and positive biological effects of their reaction products after implantation in bone and tissue applications [10,98,99,100]. They are used in various applications in bone regeneration and drug delivery. They can stimulate biological and molecular mechanisms of the body and regenerate damaged bone tissue. Most bioceramic systems are based on calcium phosphates such as Bioglass 45S5 and 1393, hydroxyapatite (HA), and β-tricalcium phosphate (β-TCP), while calcium-silicate ceramics such as akermanite Ca2Mg (Si₂O₇) have also shown similar biological behavior and high bioactivity [11,101,102,103]. Selecting the appropriate chemical composition is important for the design of a bioactive material for bone tissue engineering. Silicon (Si) is an essential trace element involved in metabolic activity as it plays an important role in processes related to bone growth. Calcium (Ca) promotes the proliferation of osteoblasts, their differentiation, and extracellular matrix maturation (ECM) calcification. Finally, magnesium (Mg) is a very important trace element in bone metabolism and plays a critical role in cellular differentiation and accelerating bone remodeling (osteoconductivity) [8,20,55,56,57,58,59,104,105,106,107].

In the present study the ternary composition 55SiO₂-35CaO-10MgO (in mol %) was selected for its high content of calcium and magnesium in relation to similar systems that have shown promising results [58,108]. This nanobioceramic was synthesized with the sol–gel method for the purpose of being incorporated into biopolymeric implants, possibly enhancing their mechanical properties, but also enabling them to induce osteogenesis and biointegration. Thermal treatment was performed to induce different crystalline phases that could have a positive effect on bioactivity. According to the thermal analysis, C2 and C3 samples were crystalline glass-ceramic materials, while C1 was primarily amorphous. All materials, regardless of the thermal treatment, consisted of nanoparticles, as was revealed by DLS measurements and TEM analysis presented in a previous study [66]. The thermally treated specimens presented similar bioactive behavior by developing the double peak of apatite even after 3 days of immersion, which agreed with the literature [105,109,110,111,112,113]. The most amorphous material after immersion in SBF revealed a high amount of apatite even from the 3rd day of immersion, while the C2 and C3 nanobioceramics, which were more crystalline, presented a gradual increase in hydroxyapatite percentages. An interesting observation was that C3, from even the 5th day of immersion, had a considerably high percentage of hydroxyapatite (71.1%), suggesting that the onset and crystallization of hydroxyapatite may be affected by the amount and type of crystalline phases. Despite that, all materials after 10 days of immersion presented similar percentages of hydroxyapatite formation, a finding that supported their potential use as regenerative materials in bone tissue engineering. Goudouri et al. synthesized an Mg-containing silicate glass–ceramic to be used as a scaffold for cementum/alveolar bone regeneration in the system 60SiO₂-30CaO-10MgO (in mol %) and reported no apatite or calcium-phosphate layer formation even after 10 days in SBF when the glass was heated to 1050 °C. The findings of the present study suggested that an increased amount of calcium along with a decrease in silica content are necessary for increased bioactivity, especially after high temperature thermal treatment.

Regarding cell viability, variations were observed in mitochondrial activity between the positive control and the tested materials. A significant increase in cell viability was observed for C1 and C2 at day 5 of cell culture. Dose and time dependence were observed on cell viability only for C1, whereas it was not observed for C2 and C3. For only C3, there was a time dependence (day 3) on cell viability. Despite minor differences, it can be safely concluded that all concentrations were at nontoxic levels.

Previous studies have shown that degradation of bioceramics may regulate the growth and metabolism of various cells by releasing ions such as Si, Ca, P, and Na into the culture medium. Cell viability has been demonstrated by previous studies revealing that, despite the fact that ions such as Mg, Ca, Na, etc., can play an important role in numerous processes related to bone formation, at high doses in the environment they are questionable, because of toxic effects they may induce [112,114]. So, it must be ensured that metal ion release dosages are within a range of nontoxic yields in order to ensure cell viability [59,115,116,117]. The results of the present study confirmed the bioactivity of the produced materials. They supported the finding that bioactive materials present high surface reactivity and allow surface remodeling when they come in contact with fluids such as cell culture medium. In the literature a cascade reaction after immersion of such materials in DMEM has been reported: first, a rapid exchange of alkali ions with hydrogen (H+ or H₃O+) occurs, leading consecutively to the formation of a silica-rich surface layer [118]. However, the described burst release of alkali ions leads to a strong increase in local pH, potentially resulting in pH-dependent cytotoxicity [119,120,121,122]. This is the reason that preconditioning was proposed as a procedure necessary before cell seeding. It supported that in case of bioglass 45S5-BG-based scaffolds 2 weeks of preconditioning were necessary before cell seeding [123] and 72 h in 45S5 granules. In the present study only 24 h preconditioning of tested materials in DMEM was performed. Previous studies have shown that cell viability and proliferation in the presence of bioactive ceramics is cell type- and composition-dependent [21,124,125,126,127,128,129,130,131]. The amount of Mg in this study was the smallest in the ternary composition (55SiO₂-35CaO-10MgO in mol %) compared to silicon and calcium. The investigation of ion release from these nanobioceramics, calculated by inductively coupled plasma spectrometry (ICP), showed that both Si²⁺ and Ca²⁺ ion release increased gradually in the first 24 h, whereas Mg²⁺ ions remained at steady and much lower values. This implied that Mg remained in the calcium silica network or in the developed crystalline phases [66], and its poor ion release was not sufficient to induce a significant enhancement of cell proliferation [66]. The MTT assay results could be correlated with the ion release during preconditioning. Increased alkali ion release combined with inadequate Mg ion release could explain the cytotoxicity observed at day 1 in all tested groups [66]. As already mentioned, this cytotoxicity could be characterized as pH-dependent. The range of the time periods for the passivation—proposed in other studies concerning bioactive materials—was broad, varying from 1 h to 24 days [132] and depended on the fluid used and the composition of the material. As cell viability was continuously higher with longer periods of passivation, extended preconditioning may be necessary in order to enhance the biocompatibility of the tested materials. Boccaccini et al. [132] supported the notion that dynamic passivation is an effective way to increase the compatibility of BGs for cell culture settings. In our study a passive preconditioning model was used. A dynamic one could be more effective concerning cell viability even at day 1. All these different parameters (material, time, fluid, passive or dynamic preconditioning) should be fields for future study, in order to establish the most beneficial preconditioning protocol for the tested materials before cell seeding [133]. However, our results at days 3 and 5 confirmed that the synthesized materials presented a biocompatible behavior, as besides the pH-dependent cytotoxicity at day 1, an increase in cell viability and proliferation was reported in all tested groups at the time points that followed.

However, various studies have shown that extracts of bioactive glasses and glass-ceramics can have a limited effect on fibroblast proliferation and that the type of released ions as well as their concentration are critical in monitoring cell response [134,135,136,137,138]. Contrary to bulk materials, the biological response of nanomaterials is highly dependent on their specific surface area [139], environment pH, particle size (nano, micro), pore size, composition, synthesis, and cell type cultures, so it was difficult to compare the present results with previous findings in the literature. In the study of Tavakoli et al. [140], increased genotoxicity was recorded in periodontal fibroblasts when the concentration of bioactive glass nanoparticles in the eluates was higher than 4 mg/mL. In the present study a low concentration of nanobioceramic nanoparticles was used in the eluates (1 mg/mL), and thus, the ion content of the extracts may not have been adequate to trigger a strong biological effect on cells. Furthermore, the majority of studies with modified calcium-silicate ternary systems have used human mesenchymal stem cells (hMSCs), osteoblasts, or cementoblasts as the main body of research has been devoted to applications of these materials in orthopedics and dentistry for bone regeneration, etc. Future research is needed to evaluate the biological behavior of these materials at higher concentrations and in direct contact with the cells. Despite differences in the effect of ions released from various bioactive materials on cells, a profound effect on the expression of some critical growth factors and important proteins, including the vascular endothelial growth factor, basic fibroblast growth factor, epidermal growth factor, collagen I, and fibronectin has been recorded in most of these studies [119,120,121,122,138,141]. These findings elucidated the enormous potential of bioactive glasses and glass-ceramics in soft tissue regeneration and wound healing [141,142]. The nanobioceramics in the present study led to a favorable behavior on gingival fibroblasts, suggesting that they can also be used with promising results in soft tissue regeneration.

The role of reactive oxygen species (ROS) in cellular processes is a critical parameter for the cells’ fate. Intracellular ROS generation can be used as an index of mitochondrial alterations and oxidative stress induced by the interaction of cells with biomaterials [143]. An enhanced production of ROS and oxidative damage can disturb mitochondrial function, affecting cell viability [144,145]. Also, many studies have shown that active radicals (such as O²⁻, H₂O₂, etc.) are strongly related to the antibacterial properties of materials [146]. By being strong oxidants, some ROS (such as OH, O²⁻ and H₂O₂) can cause damage to microorganisms, ensuring their destruction, while oxygen radicals can interact with the components of the bacterial cell as well as other cell components (such as mitochondria), causing irreversible changes in their cell-wall structure, which can lead to bacterial death [146,147,148,149]. It is well known that the effects of ROS on cellular functions are dose-dependent, because at low levels ROS play a major role in cell survival signaling processes such as proliferation, differentiation, and wound healing through the activation of a number of growth factors. However, high levels of ROS may damage cells because they are injurious to cell survival by causing inflammation, inducing apoptotic pathways, and even leading to cell death [150,151,152,153]. For example, Hunang et al. [154] confirmed that mesoporous silica nanoparticles (MSNs) significantly promoted A375 cell proliferation owing to the decrease in endogenous reactive oxygen species in cells. Also, He et al. [155,156] compared mesoporous silica nanoparticles with a covalently grafted drug delivery system exhibiting a remarkable inhibition effect on ROS levels and on the proliferation activity of LX-2 cells in the long term. Furthermore, Asharani et al. [157] showed that Ag-nanoparticles increased the production of ROS in a dose-dependent manner and reduced the ATP content of the cells, which in turn resulted in damage to cell mitochondria and apoptosis at a later stage. Calcium-silicate nanomaterials are capable of provoking an oxidizing effect as they have hydroxyl groups on their surface [101,102,115,158,159,160,161]. In the present study, the antibacterial experiments for these materials, showed a weak antibacterial effect, which could not be correlated with the significant increase in ROS levels generated in the presence of nanobioceramic eluates. However, this increase in ROS levels was inadequate to significantly affect mitochondrial activity and induce cytotoxicity in gingival fibroblasts cells. Different antioxidant mechanisms may be implicated in bacteria and cells that need further clarification. Further research work is needed to investigate their oxidative behavior at the level of interaction with cells for prolonged culture times and their capacity to induce osteogenic differentiation of mesenchymal stem cells.

5. Conclusions

Crystalline and relatively amorphous nanobioceramics were successfully synthesized by the sol–gel method and subsequently subjected to different thermal treatments. All materials presented apatite formation in SBF and biocompatibility with human periodontal fibroblasts. The similar expressions of ROS in the presence and absence of nanobioceramic eluates was correlated with the limited antibacterial activity of the materials and verified the absence of oxidative stress on cells from the presence of released ions. Nanobioceramics synthesized in the context of this study possess a multitude of attractive physicochemical and biological properties that make them suitable candidates for bone regeneration applications, as fillers in nanocomposite scaffolds, or as grafts in bone cavities and periodontal lesions.