Multifunctional Zn and Ag co-doped bioactive glass nanoparticles for bone therapeutic and regeneration

Bone cancer has traditionally been treated using surgery, radiotherapy, and/or chemotherapy. The nonspecific distribution of chemotherapy and implantable infections are significant risk factors for the failure of the bone to heal. Multifunctional zinc and silver co-doped bioactive glass nanoparticles (yAg–xZn-BGNPs) with a diameter of 150 ± 30 nm were successfully synthesized using modified sol–gel and two-step post-functionalization processes, tailored to provide antibacterial and anticancer activity whilst maintaining osteogenesis ability. Co-doped BGNPs with Zn and Ag did not significantly alter physicochemical properties, including size, morphology, glass network, and amorphous nature. Apatite-like layer was observed on the surface of yAg–xZn-BGNPs and resorbed in the simulated body fluid solution, which could increase their bioactivity. Human fetal osteoblast cell line (hFOB 1.19) treated with particles showed calcified tissue formation and alkaline phosphatase activity in the absence of osteogenic supplements in vitro, especially with 0.5Ag–1Zn-BGNPs. Moreover, these particles preferentially disrupted the metabolic activity of bone cancer cells (MG-63) and had an antibacterial effect against B. subtilis, E. coli, and S. aureus via the disc diffusion method. This novel 0.5Ag–1Zn-BGNP and 1Ag–1Zn-BGNPs, with wide-ranging ability to stimulate bone regeneration, to inhibit bone cancer cell proliferation, and to prevent bacterial growth properties, may provide a feasible strategy for bone cancer treatment. The 0.5Ag–1Zn-BGNPs and 1Ag–1Zn-BGNPs can be applied for the preparation of scaffolds or filler composites using in bone tissue engineering.


Results and discussion
Dense monodispersed BGNPs with a diameter range of 150 ± 30 nm were successfully synthesized through the sol-gel and two-step post-functionalization processes as shown in Table 1. The zeta potential of 0Ag-0Zn-BGNPs was -33.4 ± 6.2 mV. The zeta potential shifted to the positive when the Zn and Ag were incorporated. This might be due to an increase in the overall amount of network modifiers in the glass network. Previous work reported that Zn incorporated into the silica network with high binding affinity compared to Ca and Sr (the alkaline earth metals); therefore, the two-step post-functionalization method was used to incorporate cations into the glass structure. In this study, 0Ag-0Zn-BGNPs with a nominal ratio of Si:Ca/Sr 1:2 were fabricated first, followed by the second post-functionalization with Zn and Ag ( Table 3). The particles were monodispersed, with a spherical structure homogeneous in size and shape, as shown in Fig. 1. However, the particle size of 1.5Ag-1.5Zn-BGNPs was a bit smaller than other compositions. This may indicate that the addition of Zn at the nominal ratio of 0, 1, and 1.5, and Ag at 0, 0.5, 1.0, and 1.5, did not significantly affect the morphology and homogeneity of the incorporated particles.
The elements in the composition of the prepared BGNPs were measured using X-ray fluorescence (XRF). The XRF results demonstrated the incorporation of Ca, Sr, Zn, and Ag through the two-step post-functionalization. www.nature.com/scientificreports/ The total amount of Ca and Sr in BGNPs decreased significantly-from approximately 53 to 18 wt%-when Zn and Ag were doped, as shown in Table 2. These results indicated that Zn and Ag had a high binding affinity to the silica network compared to Ca and Sr. When the nominal ratio of Ag increased, the amount of Ag in the composition increased, while the amount of Zn in the composition decreased. The total amount of Zn and Ag of 1.5Ag-1.5Zn-BGNPs was lower than that of 0, 1, and 1.5Ag-1Zn-BGNPs, implying not all doped Zn and Ag could incorporate into the BGNPs. ATR-FTIR spectra of BGNPs after calcination at 680 °C and 550 °C showed the unique, characteristic main bands corresponding to the vibration modes of the Si-O-Si bonds in the regions of 1300 and 400 cm −1 , as shown in Fig. 2a (Left panel). The broad band between 1300 and 1000 cm −1 was associated with the Si-O-Si asymmetric stretching vibration. The bands located at approximately 800 cm −1 related to the symmetric Si-O-Si stretching, and those at 450 cm −1 , to the rocking vibration of Si-O-Si bending 30 . The shoulder band at approximately 940 cm −1 was reduced when Zn and Ag were doped through the second post-functionalization. These results indicated the formation of bridging oxygen, resulting in an increase in network connectivity of the glass structure. The XRD pattern of typical amorphous BGNPs showed a broad halo at low degrees (around 23-26°), clearly indicating the structural disorder of the glass network prepared using the sol-gel method ( Fig. 2b (Right panel)). The XRD pattern confirmed the completely amorphous nature of all samples. There was no crystal pattern of metal oxide in the XRD results, implying that Zn and Ag were successfully incorporated into the glass network without a structural change 31,32 . Thus, it is possible to tailor BGNPs' properties by incorporating bivalent cationic ions, including Zn and Ag.
The in vitro bioactivity of the prepared dense monodispersed BGNPs was evaluated via apatite formation through immersion in PBS and SBF solution at pH 7.4 in an incubating shaker at 37 °C, shaking at 120 rpm for 4 h, 1 day, 7 days, 14 days, and 21 days of incubation. After the interval time, the pH of the sample solutions was recorded to monitor the hydrolytic stability. The pH of the SBF and PBS solutions for all samples did not significantly change, as shown in Fig. 3a,b. However, the pH solution of 0Ag-0Zn-BGNPs showed a greater increase than other groups. This was because the released ions played a role in stabilizing large changes in the    www.nature.com/scientificreports/ pH of the solutions. The cations in the particle compositions were exchanged with the arouse environment to stabilize silanol groups. The phosphate ions were precipitated on the particles' surface. These results indicated that the ions released from the particles did not alter the pH solutions over the period of study, implying that the degradation products could not trigger an adverse effect to the cellular environment 20 . The SEM images of particles after incubation in the SBF solution for 21 days revealed the formation of "cauliflower-like" structures on the particles' surfaces, as shown in Fig. 4, indicating that apatite deposits had appeared after immersion in the SBF for 21 days. The EDX-SEM profile showed the distribution of Si, Ca, Sr, Ag, Zn, and P in a cross section of BGNPs soaked in SBF. The XRD pattern of particles after incubation in the SBF solution for 21 days confirmed the formation of apatite on the particle surface (Fig. 5). The crystalline peaks were matched with the standard hydroxyapatite (JCPDS card No. 09-0432). The characteristics of major peaks were observed for hydroxyapatite at 2θ = 22.9°, 25.9°, 31.7°, 32.2°, 34.0°, 43.8° and 45.3°. Taken together, the ability of these BGNPs to form an apatite-like layer on the surface and to resorb could increase the bioactivity responsible for the bonding of bioactive glass to the host bone. Substituting Sr for Ca in the nanoparticle composition increased the rate of dissolution and ion release, resulting in the stimulation of apatite formation 33 .
The bright-field TEM images of particles after immersion in the PBS and SBF solutions for 21 days indicated that the particles had degraded, as shown in Fig. 6. In the PBS solution, there was a small change in the morphology of the particle, from a smooth to rough surface, indicating the formation of salt on the particles' surface (white arrows in TEM images). In the SBF solution, the reduction in the particle size and the transformation from spherical to irregular shape (red arrows in TEM images) indicated the biodegradation of the particles. These results were correlated with previous studies showing that BGs have the potential to stimulate osteogenesis through the formation of HCA layer, and the ions released from the degradation process 34,35 . Therefore, Zn-and Ag-doped BGNPs have great potential as nanocarriers for delivering therapeutic bivalent cations.
The effect of the particles on cell viability (direct method) against MC3T3-E1 (mouse pre-osteoblast cells), hFOB 1.19 (human fetal osteoblast cells), and MG-63 (human osteosarcoma derived osteoblastic cells) was investigated using MTT assay. Cells were exposed to the particle at a concentration range of 0-1000 µg/mL for www.nature.com/scientificreports/ 24 h. Untreated cells served as the control. A cell viability of greater than 70% relative cell viability was taken to represent no significant toxicity (ISO 10993-5). Zn-containing BGNPs had no toxicity to the MC3T3-E1 cells up to a concentration of 250 µg/mL, as shown in Fig. 7. At a concentration of 250 µg/mL, 0Ag-1.0Zn-BGNPs, 0.5Ag-1.0Zn-BGNPs, and 1.0Ag-1.0Zn-BGNPs were statistically significantly decreased compared to 0Ag-0Zn-BGNPs.
For the bone cancer model, to understand whether or not yAg-xZn-BGNPs could be used to kill cancer cells specifically, human bone cancer cells (MG-63) and non-tumor bone cells (hFOB 1.19) were compared. The cell viability of human osteoblast cells treated with particles is shown in Fig. 8. The cell viability of hFOB 1.19 treated with 1.5Ag-1.5Zn-BGNPs was statistically significant up to a concentration of 125 µg/mL (p < 0.05). At concentration above 125 µg/mL, the cytotoxicity of hFOB 1.19 treated with Zn-and Ag-containing BGNPs was statistically significantly decreased compared to hFOB 1.19 treated with particles without Zn and Ag (0Ag-0Zn-BGNPs), indicating that yAg-xZn-BGNPs were more toxic against hFOB 1.19 than MC3T3-E1 cells. Thus, the maximum concentration (cut-off level) at which the particles could be used with no toxicity to hFOB 1.19 was 125 µg/mL. MG-63 (human osteosarcoma derived osteoblastic cells) were exposed to x-Zn-BGNPs for 24 h as shown in Fig. 9. At concentrations up to 10 µg/mL, none of the particles significantly affected the cell viability of MG-63 following 24 h in culture. 0Ag-1ZnBGNPs statistically significantly decreased cell viability at concentrations of ≥ 100 µg/mL (p < 0.05). Moreover, 0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, and 1.5Ag-1Zn-BGNPs statistically significantly reduced bone cancer cells at concentrations of ≥ 125 µg/mL (p < 0.05). These results indicated that 0Ag-1Zn-BGNPs, 0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, and 1.5Ag-1Zn-BGNPs at a concentration of 125 µg/ mL killed the bone cancer cells (MG-63) preferentially whilst maintaining the cell viability of the normal bone cells (hFOB 1.19). This is because Zn is released from BGNPs in higher amounts under acidic conditions 22,36 . However, 1.5Ag-1.5Zn-BGNPs have less effectively killed the bone cancer cells resulting from a more compact glass network implying a decrease of ion release. A previous study reported that the intratumorally acidic microenvironment of bone cancer resulted in resorption activity 37 , indicating that Zn was highly released from the yAg-xZn-BGNPs in the bone cancer cells.yAg-xZn-BGNPs could induce osteogenic differentiation of normal bone cells (hFOB 1.19) after 21 days. To investigate the efficiency of osteogenesis, alizarin red S was used to indicate calcified tissue formation (Fig. 10a). The formation of mineralization was significantly increased when the bone cells were exposed to the yAg-xZn-BGNPs at a concentration of 125 µg/mL, especially in 0Ag-0Zn-BGNPs, 0Ag-1Zn-BGNPs, 0.5Ag-1Zn-BGNPs, and 1Ag-1Zn-BGNPs compared to the negative control (untreated cells in the basal condition). The ALP activity of treated cells was significantly observed compared to untreated cells in the basal condition, indicating that yAg-xZn-BGNPs have the capacity to activate osteogenic differentiation in the absence of osteogenic supplements (Fig. 10b). 0Ag-0Zn-BGNPs (46 Si:18 Ca:36 Sr) had a high capacity to induce calcified formation and ALP activity-similarly to the positive control (untreated cells in the osteogenic condition), because Ca and Sr play key functions in bone formation 38,39 . 0Ag-1Zn-BGNPs could induce bone regeneration caused by the function of Ca, Sr, and Zn 22 . An increase in the amount of Ag in the glass composition caused a slight reduction in mineralization precipitation and ALP activity because of the lower amounts of Ca, Sr, and Zn. In this study, Ag was incorporated in BGNPs owing to its remarkable antibacterial property. Therefore, the optimal amount of Ag was important to elucidate. 0.5Ag-1Zn-BGNPs and 1Ag-1Zn-BGNPs provided the effective desired results. The antibacterial effect of x-Zn-BGNPs was investigated against B. subtilis, S. aureus, and E. coli using a disc diffusion methodology, as shown in Fig. 11a,b. There was no inhibition zone formation for particles without Zn and Ag (0Ag-0Zn-BGNPs). The particles with Zn (0Ag-1Zn-BGNPs) exhibited an inhibition zone only against B. subtilis. The formation of inhibition zones around the paper discs in the agar plates after 16-18 h confirmed the antimicrobial activity of Zn-BGNPs containing Ag (0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, 1.5Ag-1Zn-BGNPs, and 1.5Ag-1.5Zn-BGNPs) against B. subtilis, S. aureus, and E. coli (Fig. 11a). For the same type of bacteria, the antibacterial effect of 0Ag-1Zn-BGNPs, 0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, 1.5Ag-1Zn-BGNPs, and 1.5Ag-1.5Zn-BGNPs against B. subtilis was significantly increased compared to the negative control (PBS). Ag-doped Zn-BGNPs (0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, 1.5Ag-1Zn-BGNPs, and 1.5Ag-1.5Zn-BGNPs) showed large inhibition zones against S. aureus and E. coli compared to the control. 0.5Ag-1Zn-BGNPs, 1Ag-1Zn-BGNPs, 1.5Ag-1Zn-BGNPs, and 1.5Ag-1.5Zn-BGNPs had an antibacterial effect against Gram-positive and Gram-negative bacteria that could have been facilitated by Ag's DNA replication-destroying function 40 . The preliminary results confirmed that Ag enhanced the antibacterial effects and assisted the Zn activity against bacterial growth. The possible mechanism of yAg-xZn-BGNPs to inhibit bacterial growth was that Zn and Ag ions were released through the degradation of yAg-xZn-BGNPs. Then they penetrated across cell membrane and generated reactive oxygen species (ROS) inside the cell 41 .

Conclusion
Multifunctional Zn and Ag co-doped BGNPs with a diameter range of 150 ± 30 nm were synthesized through the sol-gel and two-step post-functionalization processes. The incorporation of Zn and/or Ag, a network modifier, improved the bioactivity of BGNPs without altering the particle size and morphology. The amorphous nature of the synthesized particles could enhance bioactivity through apatite formation and by increasing resorption. Moreover, the modified particles stimulated calcium deposition and ALP activity, which were the key osteogenic differentiation markers. Particles co-doped with Zn and Ag had the ability to inhibit metabolic activity of the bone cancer cells (MG-63) preferentially whilst maintaining the cell viability of the normal bone cells

Methods
All reagents were from Sigma-Aldrich (Thailand) unless stated otherwise. Ethyl alcohol (99.5%), ammonium hydroxide, tetraethyl orthosilicate (TEOS), calcium nitrate tetrahydrate (99%), strontium nitrate (99%), zinc nitrate hexahydrate (≥98%), silver nitrate (99%), phosphate buffered saline (PBS), sodium chloride (NaCl), sodium hydrogen carbonate (Na-HCO 3     www.nature.com/scientificreports/ The SiO 2 -NPs were incorporated with Ca 2 (calcium nitrate tetrahydrate) and Sr 2+ (strontium nitrate) at a nominal ratio of 1.0 SiO 2 :0.5 CaO:1.5 SrO. Particles were dried at 60 °C in the oven overnight before calcination at 680 °C for 3 h at a heating rate of 3 °C /min to remove nitrate precursors and obtain Sr-containing BGNPs. After that, BGNPs were second-doped with Zn 2+ (zinc nitrate hexahydrate) and Ag + (Silver nitrate) at the nominal ratio of 1.0 SiO 2 :0.5 CaO:1.5 SrO:x Zn:y Ag, where x = 0, 1, and 1.5 and y = 0, 0.5, 1.0, and 1.5, through secondstep post functionalization as shown in Table 3. The doped particles were dried and heated at 550 °C for 3 h at a heating rate of 3 °C /min to obtain yAg-xZn-BGNPs (Fig. 12). Finally, these particles were washed with ethanol to remove excess Ca, Sr, Zn, and Ag that had not been incorporated into the particles. The cleaned particles were dried at 60 °C overnight. BGNP characterization. The particle size and surface charge of the BGNPs were measured using Nanosizer (Horiba; SZ-100V2). The particles were suspended in DI water. Three measurements were obtained for each suspension and average value was reported with standard deviation. The surface morphology the BGNPs was imaged using a Scanning Electron Microscope (SEM) operating at 20 kV. An X-ray diffractometer (XRD) was used to identify the crystallized pattern of the particles. The XRD pattern was collected with a Bruker AXS automated powder diffractometer using Cu Kα radiation (1.540600 A°) at 40KV/40mA. Data were collected in the 10-70° 2θ range with a step size of 0.02° and a dwell time of 1.0 s. Fourier transform infrared spectroscopy (FTIR; Thermo Scientific Nicolet iS5) was used in attenuated total reflection (ATR) mode at a wavenumber ranging from 4000 to 400 cm −1 , at a scan speed of 32 scan/min, and with a resolution of 4 cm −1 . To determine the elemental composition of BGNPs, X-ray fluorescence (XRF: Fischer/XUV773) operated in a vacuum with X-ray generators in the 8-20 kV range was used.
Bioactivity assessment. To compare the release profile of ions from the yAg-xZn-BGNPs, the release of Si, Ca, Sr, Zn, and Ag ions from yAg-xZn-BGNPs was monitored as a function of time for 21 days in two different solutions: simulated body fluid (SBF) at pH 7.4, and phosphate-buffered saline (PBS) at pH 7.4. The SBF solution was prepared following a previous study 42 . The SBF had to be pre-heated at 37 °C and pH-adjusted to 7.40 before starting the experiment. The in vitro bioactivity assessment was conducted by incubating 75 mg of yAg-xZn-BGNPs in the 50 mL of SBF and PBS solutions at pH 7.4 and 37 °C, with shaking at 120 rpm for 21 days. At the end of the incubation period, the yAg-xZn-BGNPs were collected by centrifugation and then immediately washed with ethanol and subsequently with acetone to terminate the reactions. The morphological structure of yAg-xZn-BGNPs was characterized using SEM and TEM.  After that, the cell culture media was replaced with media containing NPs at concentrations ranging from 0-1 mg/mL: namely 0, 10, 100, 125, 250, 500, 750, and 1000 µg/mL. Cells were exposed to particles for 1 day (direct contact). The control comprised cells without particle exposure. Cell viability was determined using the MTT colorimetric assay based on the conversion of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) into  Osteogenic differentiation. To evaluate the osteogenic differentiation of hFOB 1.19 cells, alkaline phosphatase (ALP) activity and alizarin red s were monitored. hFOB 1.19 cells in the basal medium were seeded in flat-bottom 48-well plates with a cell concentration of 2 × 10 3 cells/well. The cells were incubated at 37 °C overnight to allow the cells to attach in a monolayer. The osteogenic medium served as the positive control. The basal medium was supplemented with 10 nM dexamethasone (DEX, Sigma-Aldrich), 10 mM β-glycerophosphate (Sigma-Aldrich), and 100 µg/mL ascorbic acid (Sigma-Aldrich). After that, the cell culture media was replaced with media containing NPs at a concentration of 125 µg/mL. Cell culture media were routinely changed twice a week for 21 days. On the 21st day in culture, osteoblastic differentiation of hFOB 1.19 cells was measured via cellular ALP activity staining (Abcam) according to the manufacturer's instructions. Calcium phosphate deposits were detected. The cells were fixed with 4% paraformaldehyde and stained with 2% Alizarin Red S in PBS at pH 4.2 to detect calcified tissue formation.
Antibacterial activity. Antibacterial activity was determined using the disc diffusion method. Bacillus subtilis ATCC 6633, Staphylococcus aureus ATCC 25923, and Escherichia coli ATCC 25922 isolates were grown overnight on Mueller Hinton Agar (MHA). Direct colony suspensions were prepared in sterile saline to achieve a turbidity equivalent to 0.5 McFarland standards. The inoculum suspensions were streaked onto the dried surface of an MHA plate with a sterile cotton swab and left to absorb for no more than 15 minutes. The particles were re-suspended in the PBS solution at the concentration of 250 µg/mL in the ultrasonication bath for 30 minutes. Twenty microliters of BGNPs were pipetted onto 6-mm-diameter Whatman ® antibiotic assay discs in a sterile dish. PBS and antibiotic (ampicillin, AMP 10 µg, Thermo Scientific™) was served as the negative and positive Table 3. Compositions of bioactive glass nanoparticle (nominal ratio). www.nature.com/scientificreports/ control, respectively. Discs were placed aseptically onto the plates immediately. The plates were incubated at 37 °C for 16 to 18 h and the diameter of the inhibition zones or halo zones was measured using a sliding caliper.

Statistical analysis. Statistical analyses were performed by one-way analysis of variance (ANOVA) in
Minitab. A p value < 0.05 was considered significant. The graphs shown present the results as the mean values, with the standard deviations (SD) as the error bars. All the quantitative experiments were carried out at least in triplicate.

Data availability
The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.