Injectability, Processability, Drug Loading, and Antibacterial Activity of Gentamicin-Impregnated Mesoporous Bioactive Glass Composite Calcium Phosphate Bone Cement In Vitro

Calcium phosphate cement (CPC) is similar to bone in composition and has plasticity, while mesoporous bioactive glass (MBG) has the advantage of releasing Si, which can promote osteogenic properties and drug loading capacity. A sol–gel-prepared MBG micro-powder (mMBG) and further impregnated antibiotic gentamicin sulfate (Genta@mMBG: 2, 3, and 4 mg/mL) antibiotic were added to CPC at different weight ratios (5, 10, and 15 wt.%) to study CPC’s potential clinical applications. Different ratios of mMBG/CPC composite bone cement showed good injectability and disintegration resistance, but with increasing mMBG addition, the working/setting time and compressive strength decreased. The maximum additive amount was 10 wt.% mMBG due to the working time of ~5 min, the setting time of ~10 min, and the compressive strength of ~51 MPa, indicating that it was more suitable for clinical surgical applications than the other groups. The 2Genta@mMBG group loaded with 2 mg/mL gentamicin had good antibacterial activity, and the 10 wt.% 2Genta@mMBG/CPC composite bone cement still had good antibacterial activity but reduced the initial release of Genta. 2Genta@mMBG was found to have slight cytotoxicity, so 2Genta@mMBG was composited into CPC to improve the biocompatibility and to endow CPC with more advantages for clinical application.


Introduction
Complex bone defects caused by accident, infection, and aging are often accompanied by poor self-repair and must be restored with synthetic bone substitutes. An ideal synthetic bone substitute needs to be osteoconductive and osteoinductive at the repair site. Calcium phosphate bone cement, commonly known as CPC, is a mixture of tetracalcium phosphate (TTCP) and dicalcium phosphate anhydrous (DCPA) powder with a Ca/P atomic ratio of The pulverization process of TTCP and the surface modification of sm-DCPA have been shown in our previous study [4]. The process was as follows: 16.6 g of TTCP, 12.4 g of sm-DCPA, and 117 g of alumina spheres were added to a PE bottle and mixed for 24 h to prepare the CPC.

MBG Micro-Powder and Antibiotic-Impregnated mMBG (Genta@mMBG)
In this experiment, mMBG was prepared using an acid-catalyzed method based on the sol-gel method. The process was as follows: 1 g of surfactant F-127 and an acid catalyst of 2 M nitric acid (1 g) were mixed with the precursor material of SiO 2 -CaO-P 2 O 5 (the molar ratio of Si:Ca:P = 80:15:5), which was prepared by dissolving 6.7 g of TEOS, 1.43 g of calcium nitrate, and 0.73 g of TEP in 60 g of ethanol and stirring for 24 h. The above sol-gel was impregnated with a polyurethane sponge, dried at 100 • C, calcined to 600 • C at a rate of 10 • C/min, and held for 2 h. The sintered cake was pulverized and filtered through a 325 mesh sieve to obtain an irregularly shaped mMBG powder. Gentamicin at a stock concentration of 40 mg/mL was mixed into double distilled water (ddH 2 O) to adjust 50 mL of different concentrations of gentamicin (2, 3, and 4 mg/mL) in water; gentamicin was then mixed with mMBG to make a liquid ratio mixture of 1/50 (g/mL). After stirring for 24 h, the antibiotic-impregnated powder was filtered and dried at room temperature for 24 h to obtain drug loading Genta@mMBG (2Genta@mMBG, 3Genta@mMBG, and 4Genta@mMBG).

mMBG/CPC and 2Genta@mMBG/CPC Composite Bone Cement
CPC powder was mixed with different weight ratios (5, 10, and 15 wt.%) of mMBG and 10 wt.% of Genta@mMBG mixed with 0.67 M phosphate solution hardener. The pH of the hardener was adjusted to 6.02. An mMBG/CPC slurry was formed at a powder-to-liquid ratio of 0.8 g/380 µL. This was mixed well within 1 min, and the slurry was poured into a mold and left to set for 3 min. During sample preparation, the sample was kept under a pressure of 0.7 MPa (100 psi), and pressed into a cylindrical sample with a height of 12 mm and a diameter of 6 mm. It was then demolded, and the following physical and chemical properties were compared.

Characterization of MBG and Antibiotic Impregnated Genta@mMBG
The powder morphology of the mMBG particles was observed via transmission electron microscopy (JEM-2100F, JEOL, Tokyo, Japan) operated at 200 kV to characterize the internal structures and micrographs of mMBG and Genta@mMBG particles. The specific surface area and changes between mMBG and Genta@mMBG particles were determined using the Brunauer-Emmett-Teller (BET) method. The nitrogen adsorption and desorption isotherm data were obtained at −196 • C on a constant volume adsorption device (ASAP2020, Micromeritics, Norcross, GA, USA).

Characterization of mMBG/CPC and Genta@mMBG/CPC Composite Bone Cements
2.4.1. X-ray Diffraction (XRD) and Fourier Transform Infrared Spectroscopy (FTIR) Analysis XRD was used to analyze the crystallinity and phases of samples. XRD was used to analyze whether the crystallization of mMBG or Genta@mMBG on CPC was affected. The bone cement after the reaction was ground into powder and analyzed via XRD diffrac-tometer (XRD-6000, Shimadzu, Kyoto, Japan). Diffraction conditions using Ni-filtered Cu target Kα were typically operative at 30 kV, 20 mA, and scanning angles were in the range of 20-60 • with a scanning 2θ rate of 2 • /min. The relative XRD patterns were determined by comparing the different diffraction peaks with the JCPDS standard diffraction in the database file.
FTIR spectra were used to analyze the functional group changes of CPC after composite mMBG and Genta@mMBG. The sample was made by mixing the powder with KBr at a ratio of 1/100 (g/g) and pressing it into a translucent circular sheet with a diameter of approximately 12 mm. IR spectra were obtained by detecting changes of translucent sheets in transmittance (or absorption) intensity as a function of frequency in an FTIR spectrometer (Nicolet iS5, Thermo Fisher Scientific, Waltham, MA, USA).

Working/Setting Time, Injectability, and Dispersibility
When the bone cement was mixed and the slurry was gradually hardened, no mutual adhesion was found between the powders in the slurry; i.e., the two slurries could not be aggregated into a block, which was an indicator that could be used to measure the working time. The setting time was measured according to a dental phosphoric acid standard, ISO 9917-1, by pressing down the sample vertically with a 400 g Gillmore needle at 37 • C and 60-70% humidity until no visible indentation was found on the surface; i.e., the hardening time was recorded.
The test of injectability and disintegration was performed by putting the bone cement after mixing for 1 min into a 5 mL syringe within 3 min, and immediately applying 250 N vertically to the syringe barrel to inject the bone cement into 37 • C ddH 2 O. The change of bone cement after injection into the water was observed by taking pictures. A successful injection meant that it can be injected. If no disintegration phenomenon was observed after 15 min of immersion, then it has disintegration resistance.

Compressive Strength and Fracture Surface Observation
The sample mold used in this study had a diameter of 6 mm and a height of 12 mm, and was tested according to ASTM F451-16. The sample was immersed in Tris-buffer solution artificial body fluid at a ratio of 1 g/10 mL. It was placed at 37 • C for 1 day and then taken out. The compressive strength of the wet specimens was measured using a universal testing machine (HT-2402, Hung Ta, Taichung, Taiwan) at a crosshead speed of 1 mm/min.
The morphology of composite bone cement after compression was observed using a scanning electron microscope (SEM; S-3000N, HITACHI, Tokyo, Japan) equipped with an energy dispersive spectrometer (EDS) to observe the distribution of Si, and to analyze whether the MBG was well compounded into the CPC.

Antibacterial Abilities
Staphylococcus aureus (S. aureus; ATCC No. 25923) and Escherichia coli (E. coli; ATCC No. 10798) were cultured in tryptic soy broth (TBS). The bacterial suspension was diluted to achieve an optical density of 0.2 (equivalent to~1.0 × 10 7 cells/mL on average) at 595 nm (OD 595 ). This value was confirmed with an enzyme-linked immunosorbent assay (ELISA) reader (EZ Read-400, Biochrom, Holliston, MA, USA). The bacterial suspensions of S. aureus and E. coli were subsequently diluted to achieve an OD 595 value of 0.2. First, the bacterial liquid cultured with TBS was evenly coated on the surface of the agar, and Genta@mMBG and Genta@mMBG/CPC were used to form cylindrical samples with a diameter of 6 mm and a height of 3 mm. The samples were then attached to the plate and incubated at 37 • C for 24 h. The zone of inhibition was observed and measured. In terms of antibacterial quantification, 0.2 g of the sample was mixed with 2 mL of bacterial suspension and cultured at 37 • C for 1, 2, 3, and 4 days. Approximately 100 µL of bacterial suspension was removed at each time point, and an equal amount of TBS was added to the original shaker tube. The absorbance of OD 595 was measured, and the antibacterial effect of the samples was quantified.

In Vitro Cytotoxicity Tests
The L929 cell line of neonatal mouse fibroblasts was provided by the National Institutes of Health, Miaoli, Taiwan, and used for cytotoxicity assays. The testing procedure was performed according to ISO 10993-5:2009. L929 cells were cultured with culture medium in an incubator at 37 • C and 5% CO 2 , and subcultured when cell concentrations were between 0.8 × 10 6 and 1.0 × 10 6 cells/mL. The medium used was Minimal Essential Medium alpha medium (Gibco, Thermo Fisher Scientific Inc., Waltham, MA, USA) containing 10% horse serum, changed every 2 days of culture.
The samples prepared for cell culture were sterilized via autoclaving at 121 • C and 1.05 kg/cm 2 (15-20 psi; TOMIN, TM-328, Taipei, Taiwan). Sterilized mMBG/CPC composite, high-density polyethylene (HDPE) negative control, and 15 vol.% dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) were added into the medium. The extraction ratio of sample to medium was set to 1 g/5 mL, and the samples were placed in a 37 • C incubator for 24 h to extract the medium for cytotoxicity detection.
For quantitative cytotoxicity arrays, 100 µL of suspension containing L929 1 × 10 4 cells was transferred to a 96-well microplate and incubated at 37 • C for 24 h. Media was removed and 100 µL of sample extract was added, then L929 cells were incubated in a 37 • C incubator for 24 h and mixed with 50 µL of XTT Cell Proliferation Assay Kit (Biological Industries, Kibbutz Beit Haemek, Israel) for a 4 h extension reaction. Afterwards, they were processed with another ELISA reader (SPECTROstar Nano, BMG LABTECH, Offenburg, Germany). The measured OD 490 absorbance was proportional to the cell viability.
In the qualitative cytotoxicity test, the control group extract and medium were prepared in the same way as in the quantitative test. Approximately 1000 µL of cell suspension were collected, and L929 cells were seeded into a 48-well microplate at a cell concentration of 1 × 10 5 cells/well. The original medium was cultured in an incubator at 37 • C and 5% CO 2 for 1 day. Then, the medium was removed. Sample extract (100 µL) was added and cultured for 24 h, and cell morphology was observed under an inverted microscope (IVM-3AFL, SAGE VISION Co., LTD, New Taipei City, Taiwan).

Statistical Analysis
Analysis of variance (ANOVA) was performed using IBM SPSS Statistics (version 20, IBM Corp., Armonk, NY, USA) for the measurement of working/setting time, compressive strength, and antibacterial abilities. ANOVA was used to determine whether the differences among the means of multiple groups were significant. The estimates of two different variables were used to compare the differences.

Injectability, Dispersibility, and Working/Setting Time of Genta@mMBG/CPC Composite
The working time is indicated when the slurry gradually hardens and there is no further mutual adhesion between the slurries. The initial strength was tested with a Gillmore needle, no obvious indentation was found on the surface, and the setting time was recorded. Optical image comparison of the mMBG/CPC composite bone cement to measure the difference between the working and setting times is shown in Figure 1.

Composite
The working time is indicated when the slurry gradually hardens and there is no further mutual adhesion between the slurries. The initial strength was tested with a Gillmore needle, no obvious indentation was found on the surface, and the setting time was recorded. Optical image comparison of the mMBG/CPC composite bone cement to measure the difference between the working and setting times is shown in Figure 1. Considering that CPC bone cement has the advantage of being able to pipe the slurry before the CPC sets, it can be applied as a minimally operative procedure, and it would perfectly fit the repair site. Its anti-washout and anti-dispersion properties in solution are critical for clinical applications. The addition of fillers to the CPC may alter the properties of the CPC. They should be measured because they affect the handling properties or the acceptable range of cement work and setting time. The anti-extrusion force may hinder the application of CPC for the injection process. In this study, the required injection time, including mixing and loading times, was controlled within 3 min, and the injection results are shown in Figure 2. All groups could be injected smoothly without an obvious blockage of CPC residues, and no obvious powder dispersion phenomenon was observed after the CPC was injected into ddH2O. At the same time, the mMBG/CPC composite showed no obvious disintegration phenomenon after soaking for 15 min, indicating that the ratio of mMBG in CPC did not significantly hinder the original injection or anti-disintegration ability of CPC. Considering that CPC bone cement has the advantage of being able to pipe the slurry before the CPC sets, it can be applied as a minimally operative procedure, and it would perfectly fit the repair site. Its anti-washout and anti-dispersion properties in solution are critical for clinical applications. The addition of fillers to the CPC may alter the properties of the CPC. They should be measured because they affect the handling properties or the acceptable range of cement work and setting time. The anti-extrusion force may hinder the application of CPC for the injection process. In this study, the required injection time, including mixing and loading times, was controlled within 3 min, and the injection results are shown in Figure 2. All groups could be injected smoothly without an obvious blockage of CPC residues, and no obvious powder dispersion phenomenon was observed after the CPC was injected into ddH 2 O. At the same time, the mMBG/CPC composite showed no obvious disintegration phenomenon after soaking for 15 min, indicating that the ratio of mMBG in CPC did not significantly hinder the original injection or anti-disintegration ability of CPC.  Table 1. The mMBG addition shortened the working and setting times (p < 0.05), and the time decreased with the increasing content of mMBG added (p < 0.05). The working and setting times of the CPC composite bone cement range for each group were 3-9 and 8-13 min, respectively. The working time of injectable bone cement is the duration for which the cement remains at a more or less constant and ideal viscosity for delivery to in situ restoration and penetration of the restoration. Based on the phase change of the CPC solidification reaction, the setting time provides an initial strength for overcoming anti-rushing and dispersion before the strength reaches its peak. According to the expectations of consulting clinicians and literature [8,[19][20][21], the ideal working time of CPC composite bone cement needs to be controlled in the range of 4-10 min, and the setting time needs to be in the range of 10-20 min. The developed 10 wt.% mMBG/CPC composite with the highest mMBG content achieved the expected runtime.  Table 1. The mMBG addition shortened the working and setting times (p < 0.05), and the time decreased with the increasing content of mMBG added (p < 0.05). The working and setting times of the CPC composite bone cement range for each group were 3-9 and 8-13 min, respectively. The working time of injectable bone cement is the duration for which the cement remains at a more or less constant and ideal viscosity for delivery to in situ restoration and penetration of the restoration. Based on the phase change of the CPC solidification reaction, the setting time provides an initial strength for overcoming anti-rushing and dispersion before the strength reaches its peak. According to the expectations of consulting clinicians and literature [8,[19][20][21], the ideal working time of CPC composite bone cement needs to be controlled in the range of 4-10 min, and the setting time needs to be in the range of 10-20 min. The developed 10 wt.% mMBG/CPC composite with the highest mMBG content achieved the expected runtime.
Considering that the value of the compressive strength of the human trabecular bone is 0.2-16 MPa [22][23][24], the ideal strength of the implant material needs to be at least close to that of trabecular bone, to judge whether the implant material meets the needs of clinical applications. The compressive strength of the mMBG/CPC composite bone cement after immersion in Tris-buffer solution for 1 day was between 42 and 75 MPa, indicating that each group was suitable for clinical application. Figure 3 shows the fracture images and Si elemental mapping of mMBG/CPC composites with different mMBG ratios after soaking in Tris-buffer for 1 day. A coral reef-like crystal structure formed on the surface of each group; this was the product of apatite after the CPC reaction. Smooth lumps were observed in the mMBG-added CPC composite, and further Si mapping ensured that the Considering that the value of the compressive strength of the human trabecular bone is 0.2-16 MPa [22][23][24], the ideal strength of the implant material needs to be at least close to that of trabecular bone, to judge whether the implant material meets the needs of clinical applications. The compressive strength of the mMBG/CPC composite bone cement after immersion in Tris-buffer solution for 1 day was between 42 and 75 MPa, indicating that each group was suitable for clinical application. Figure 3 shows the fracture images and Si elemental mapping of mMBG/CPC composites with different mMBG ratios after soaking in Tris-buffer for 1 day. A coral reef-like crystal structure formed on the surface of each group; this was the product of apatite after the CPC reaction. Smooth lumps were observed in the mMBG-added CPC composite, and further Si mapping ensured that the main component was silicon. Considering that the distribution of mMBG among the groups was random rather than uniform, a further decrease in compressive strength was observed. According to the comprehensive consideration of working/setting time and compressive strength, the optimal addition amount of mMBG in this study was 10 wt.%. Therefore, subsequent analyses used this addition to impregnate Genta in mMBG (Genta@mMBG), which was then composited into CPC. observed. According to the comprehensive consideration of working/setting time and compressive strength, the optimal addition amount of mMBG in this study was 10 wt.%. Therefore, subsequent analyses used this addition to impregnate Genta in mMBG (Genta@mMBG), which was then composited into CPC.   Figure 4 shows the qualitative and quantitative antibacterial activities of the different concentrations of Genta-impregnated mMBG against S. aureus and E. coli. Results of the inhibition zone test in Figure 4a showed the obvious inhibition ability of each group on microbial growth, and the measured inhibition zone was similar. Results of the quantitative test in Figure 4b showed that the bacterial survival rates of each group were still similar, indicating that the drug concentration released by the 2Genta@mMBG group reached a good bacteriostatic ability that was similar to that of the positive control DMSO. Even if the concentration of Genta was increased to 4Genta@mMBG, the best bacteriostatic effect was obtained with 2Genta@mMBG. Therefore, the antibiotic concentration group,  Figure 4 shows the qualitative and quantitative antibacterial activities of the different concentrations of Genta-impregnated mMBG against S. aureus and E. coli. Results of the inhibition zone test in Figure 4a showed the obvious inhibition ability of each group on microbial growth, and the measured inhibition zone was similar. Results of the quantitative test in Figure 4b showed that the bacterial survival rates of each group were still similar, indicating that the drug concentration released by the 2Genta@mMBG group reached a good bacteriostatic ability that was similar to that of the positive control DMSO. Even if the concentration of Genta was increased to 4Genta@mMBG, the best bacteriostatic effect was obtained with 2Genta@mMBG. Therefore, the antibiotic concentration group, 2Genta@mMBG, with the least burden of a further application for clinical application, was selected for subsequent composite CPC experiments.   Figure 5a shows the microstructural images of mMBG before and after impregnation with Genta. The SEM analysis shows that the mMBG particles before drug loading are large and irregular in shape, and with a particle size of approximately 100-500 µm. Several large concave pores were obvious on the surface, and some particles showed interconnected pore structures, which were related to the inheritance of polyurethane sponge as a dipping template for sintering. After drug loading, the particles became significantly smaller, the average particle size was less than 100 µm, and the shape was broken and irregular. As mMBG was soaked and agitated for a day, particle collisions occurred, resulting in the refinement of the mMBG particles. Images from TEM showed that even Genta-impregnated mMBG had mesopores neatly arranged in a grid. Considering that Genta impregnation was not observed via SEM and TEM, the subsequent nitrogen adsorption/desorption was used for drug loading analysis. Figure 5b shows the nitrogen adsorption/desorption curves and pore size/pore volume distribution of mMBG before and after impregnation with Genta. All the curves of the groups are type IV isotherm curves, indicating that the two mMBGs have mesoporous structures before and after impregnation with the drug [25]. The type of hysteresis loop of mMBG before and after impregnation with Genta was the H1 type, indicating that the pore type was a neatly arranged cylindrical channel [26]. The pore size distribution map calculated using BJH theory was affected by the concave pores of different sizes on the surface. Thus, a pore size distribution with an average pore diameter of 126.7 Å (12.67 nm) before drug impregnation was wider.

SEM, TEM, and BET of MBG and Genta-Impregnated mMBG
3.2.2. SEM, TEM, and BET of MBG and Genta-Impregnated mMBG Figure 5a shows the microstructural images of mMBG before and after impregnation with Genta. The SEM analysis shows that the mMBG particles before drug loading are large and irregular in shape, and with a particle size of approximately 100-500 µm. Several large concave pores were obvious on the surface, and some particles showed interconnected pore structures, which were related to the inheritance of polyurethane sponge as a dipping template for sintering. After drug loading, the particles became significantly smaller, the average particle size was less than 100 µm, and the shape was broken and irregular. As mMBG was soaked and agitated for a day, particle collisions occurred, resulting in the refinement of the mMBG particles. Images from TEM showed that even Genta-impregnated mMBG had mesopores neatly arranged in a grid. Considering that Genta impregnation was not observed via SEM and TEM, the subsequent nitrogen adsorption/desorption was used for drug loading analysis. Figure 5b shows the nitrogen adsorption/desorption curves and pore size/pore volume distribution of mMBG before and after impregnation with Genta. All the curves of the groups are type IV isotherm curves, indicating that the two mMBGs have mesoporous structures before and after impregnation with the drug [25]. The type of hysteresis loop of mMBG before and after impregnation with Genta was the H1 type, indicating that the pore type was a neatly arranged cylindrical channel [26]. The pore size distribution map calculated using BJH theory was affected by the concave pores of different sizes on the surface. Thus, a pore size distribution with an average pore diameter of 126.7 Å (12.67 nm) before drug impregnation was wider. Table 2 shows the summary results of specific surface area, pore volume, and pore size obtained via BET testing. After impregnation of mMBG with Genta, the specific surface area and pore volume increased, but the mesopore size decreased. The particle refinement after mMBG impregnation increased the specific surface area, especially the mMBG surface, and the dispersed Genta was also exposed in the internal pores of mMBG, thereby increasing the total pore volume and reducing the average pore size.  Table 2. Analysis of specific surface area, pore volume, and pore size of Genta-free mMBG and 2Genta@mMBG.

FTIR and XRD of mMBG/CPC and Genta@mMBG/CPC Composites
The infrared spectra of mMBG before and after the impregnation of Genta and further composite CPC bone cement are shown in Figure 6a. Hydroxyl groups (OH − ) can be found in mMBG at 1452, 1635, and 3440 cm −1 [27], indicating that the prepared MBG was hydrophilic [28]. At 470, 800, and 1081 cm −1 , the non-stretching, symmetric, and asymmetric stretching vibration peaks of silicate (Si-O-Si) response are shown, respectively; these are the main absorption bands of MBG [29,30]. The N-H functional group in Genta should be displayed at 1635 cm −1 [31], but this cannot be verified in the 2GentaM group, because the N-H absorption band of Genta overlapped with broad Si-O-Si bands of mMBG. Then, after soaking the mMBG/CPC composite of 10 wt.% mMBG/CPC in Tris-buffer for 1 day, the O-P-O (ν4 mode in PO4 3-) bending vibrations of HA can be seen at 563 and 603 cm −1 . The asymmetric bending vibration of HA was found at 1041 cm −1 (ν3 mode in PO4 3− ) [32][33][34]. These bands indicated that the addition of mMBG does not affect the changes in CPC functional group vibrations. Figure 6b shows the XRD analysis of 10 wt.% Genta-free mMBG, 2Genta@mMBG, and mMBG further composite CPC. A broad SiO2 diffraction peak was observed between 20° and 30° (2θ) before and after mMBG impregnation with Genta, thereby indicating that mMBG is an amorphous structure [35,36]. After comparison with the database, mMBG observed the diffraction plane of CaSiO3 at 29.42° (JCPD 45-0156) [37,38]. Then, after soaking 10 wt.% mMBG/CPC in Tris-buffer for 1 day, the main diffraction planes of HA, such as (002), (211), (310), and (222), can be observed (JCPD 09-  Table 2 shows the summary results of specific surface area, pore volume, and pore size obtained via BET testing. After impregnation of mMBG with Genta, the specific surface area and pore volume increased, but the mesopore size decreased. The particle refinement after mMBG impregnation increased the specific surface area, especially the mMBG surface, and the dispersed Genta was also exposed in the internal pores of mMBG, thereby increasing the total pore volume and reducing the average pore size. Table 2. Analysis of specific surface area, pore volume, and pore size of Genta-free mMBG and 2Genta@mMBG.

Samples
Surface Area (m 2 /g) Pore Volume (cm 3 /g) Pore Size (nm) The infrared spectra of mMBG before and after the impregnation of Genta and further composite CPC bone cement are shown in Figure 6a. Hydroxyl groups (OH − ) can be found in mMBG at 1452, 1635, and 3440 cm −1 [27], indicating that the prepared MBG was hydrophilic [28]. At 470, 800, and 1081 cm −1 , the non-stretching, symmetric, and asymmetric stretching vibration peaks of silicate (Si-O-Si) response are shown, respectively; these are the main absorption bands of MBG [29,30]. The N-H functional group in Genta should be displayed at 1635 cm −1 [31], but this cannot be verified in the 2GentaM group, because the N-H absorption band of Genta overlapped with broad Si-O-Si bands of mMBG. Then, after soaking the mMBG/CPC composite of 10 wt.% mMBG/CPC in Tris-buffer for 1 day, the O-P-O (ν 4 mode in PO 4 3-) bending vibrations of HA can be seen at 563 and 603 cm −1 . The asymmetric bending vibration of HA was found at 1041 cm −1 (ν 3 mode in PO 4 3− ) [32][33][34]. These bands indicated that the addition of mMBG does not affect the changes in CPC functional group vibrations. Figure 6b shows the XRD analysis of 10 wt.% Genta-free mMBG, 2Genta@mMBG, and mMBG further composite CPC. A broad SiO 2 diffraction peak was observed between 20 • and 30 • (2θ) before and after mMBG impregnation with Genta, thereby indicating that mMBG is an amorphous structure [35,36]. After comparison with the database, mMBG observed the diffraction plane of CaSiO 3 at 29.42 • (JCPD 45-0156) [37,38]. Then, after soaking 10 wt.% mMBG/CPC in Tris-buffer for 1 day, the main diffraction planes of HA, such as (002), (211), (310), and (222), can be observed (JCPD 09-0432) [39], indicating that the addition of mMBG did not affect the phase transition of the CPC reaction into the HA phase.
Biomimetics 2022, 7, 121 12 of 18 0432) [39], indicating that the addition of mMBG did not affect the phase transition of the CPC reaction into the HA phase.   Figure 7a is the qualitative antibacterial test of 2Genta@mMBG and 10 wt.% 2Genta@mMBG/CPC composite. The inhibition zone after the 2Grenta@mMBG composite CPC was smaller than before compositing, indicating that combining Genta-impregnated mMBG into CPC reduced the initial antibiotic release of Genta. In addition, the antibacterial quantitative test for 1-4 days is shown in Figure 7b. The antibacterial effect of composite 10 wt.% 2Genta@mMBG/CPC was worse than that of 2Genta@mMBG-only, although it still had good antibacterial activity.

Antibacterial Abilities of Genta@mMBG/CPC Composite
3.3.3. Cytotoxicity of mMBG and Antibiotic-Impregnated 2Genta@mMBG and mMBG/CPC Composites Figure 8 shows the quantitative and qualitative analyses of cytotoxicity before and after mMBG was loaded with Genta, and the composite CPC extract was cultured with L929 cells for 1 day. Cell viability between 50% and 70% indicated slight cytotoxicity (Grade 1) according to ISO 10993-5 [40]. If the biomaterial extract showed no evidence of causing cell lysis or toxicity, and if the cell cultures exposed to the test item did not show greater than mild reactivity (Grade 2), then the biomaterial can be considered to be cytocompatible. Considering that 2Genta@mMBG belongs to Grade 1 cytotoxicity rather than other groups of Grade 0 (no reactivity), it must be added to other materials to improve biocompatibility if it is used in clinical restoration. Figure 8b shows the qualitative analysis of cytotoxicity in each group. The results of all the groups are the same as those of the quantitative analysis of cytotoxicity.  Figure 7a is the qualitative antibacterial test of 2Genta@mMBG and 10 wt.% 2Genta@mMBG/CPC composite. The inhibition zone after the 2Grenta@mMBG composite CPC was smaller than before compositing, indicating that combining Genta-impregnated mMBG into CPC reduced the initial antibiotic release of Genta. In addition, the antibacterial quantitative test for 1-4 days is shown in Figure 7b. The antibacterial effect of composite 10 wt.% 2Genta@mMBG/CPC was worse than that of 2Genta@mMBG-only, although it still had good antibacterial activity. 3.3.3. Cytotoxicity of mMBG and Antibiotic-Impregnated 2Genta@mMBG and mMBG/CPC Composites Figure 8 shows the quantitative and qualitative analyses of cytotoxicity before and after mMBG was loaded with Genta, and the composite CPC extract was cultured with L929 cells for 1 day. Cell viability between 50% and 70% indicated slight cytotoxicity (Grade 1) according to ISO 10993-5 [40]. If the biomaterial extract showed no evidence of causing cell lysis or toxicity, and if the cell cultures exposed to the test item did not show greater than mild reactivity (Grade 2), then the biomaterial can be considered to be cytocompatible. Considering that 2Genta@mMBG belongs to Grade 1 cytotoxicity rather than other groups of Grade 0 (no reactivity), it must be added to other materials to improve biocompatibility if it is used in clinical restoration. Figure 8b shows the qualitative analysis of cytotoxicity in each group. The results of all the groups are the same as those of the quantitative analysis of cytotoxicity.

Conclusions
Antibiotic-impregnated bioactive glass composite calcium phosphate bone cement was investigated. The addition of 5, 10, and 15 wt.% mMBG of mMBG/CPC composite bone cement did not affect the injectability of bone cement, and it still exhibited good anti-dispersion properties. However, considering that the working setting time and the compressive strength of the mMBG/CPC composites decreased with the increase of mMBG concentration, 10 wt.% mMBG was chosen to be added to mMBG/CPC. The working time was about 5 min, and the setting time was about 10 min. The FTIR and XRD results showed that the addition of mMBG did not hinder the formation of apatite. In addition, from the bacteriostatic test, in the mMBG impregnated with 2 mg/mL gentamicin, the minimum amount of Genta added in the 2Genta@mMBG group had good bacteriostatic activity, but less cytotoxicity. After 2Genta@mMBG composite CPC bone cement, 10 wt.% 2Genta@mMBG/CPC composite bone cement was observed to reduce the initial antibiotic burst release of Genta. Although 10 wt.% 2Genta@mMBG/CPC composite bone cement was not as effective as 2Genta@mMBG alone, it still had good antibacterial properties. Overall, 10 wt.% 2Genta@mMBG/CPC composite bone cement is antibacterial and does not significantly delay the CPC response, making it a good candidate for bone replacement materials. was partially supported by the research grant of Show Chwan Memorial Hospital (SRD108017). The authors acknowledge and appreciate the Precision Instrument Support Center of Feng Chia University, which provided the fabrication and measurement facilities. The assistance of the participants in this research is also acknowledged.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data is contained within the article.