O-Acrylamidomethyl-2-hydroxypropyltrimethyl ammonium chloride chitosan and silk modified mesoporous bioactive glass scaffolds with excellent mechanical properties, bioactivity and long-lasting antibacterial activity

Abstract

Mesoporous bioactive glass (MBG) is a promising scaffold in bone tissue engineering because its large specific surface area facilitates bioactive behavior and allows mesopores to be loaded with osteogenic agents for promoting the formation of new bone. In the present study, a biocompatible MBG-based scaffold for bone regeneration applications with long-lasting antibacterial activity were fabricated via surface modification with O-acrylamidomethyl-2-hydroxypropyltrimethyl ammonium chloride chitosan (NMA-HACC) and silk. The NMA-HACC–silk (NHS) modified MBG scaffolds were prepared by evaporation-induced self-assembly using polyurethane sponges and the P123 surfactant as co-templates. The microstructure of the scaffold was characterized using scanning electron microscopy (SEM) and synchrotron radiation microcomputer tomography (SRμCT). Fourier transform infrared spectroscopy (FTIR), SEM, X-ray diffraction (XRD), and mechanical experiments were used to analyze the product's composition, inner microstructure, morphology, and mechanical strength before and after its surface modification. These methods were also used to assess the degree to which minerals became deposited on the scaffold after soaking in simulated body fluid (SBF). Confocal laser scanning microscopy (CLSM) was used to evaluate the antibacterial properties and biocompatibility of the scaffolds at various time intervals. Finally, biocompatibility was demonstrated by studying the in vitro proliferation and viability of human mesenchymal stem cells (hMSCs). The results showed that the fabricated scaffolds possessed well-ordered, three-dimensional structures and the NMA-HACC–silk modification rendered the pore network more uniform and continuous, leading to the significant improvement in bioactivity and in hMSC attachment, cell spreading and cell proliferation. Furthermore, the NMA-HACC–silk modification significantly prolonged the antibacterial activity of the MBG scaffold.

---

1. Introduction

The development of bioactive materials capable of replacing hard tissue has been an important advance in the field of bone grafting.^1–4 One problem that continues to challenge orthopedists is the treatment of infected segmental bone due to the seriousness of the bone infection, which is a considerable risk in bone filling and replacement.^5–8 There is a high incidence of osteomyelitis and bone necrosis, accompanying a high mortality rate.⁹ Generally, acute and chronic bacterial osteomyelitis are treated by surgically removing the necrotic bone tissue and reconstructing or repairing that tissue using porous, biodegradable scaffolds. Nonetheless, either these scaffolds are loaded with antibiotics or the patient must take high systemic doses of antibiotic drugs.^10–12

After removal of the necrotic bone, the remaining defects must be filled. This is usually accomplished with bioactive materials, which serves as a structural support for cell adhesion, cell proliferation, and cell differentiation within the defects.^13,14 Therefore, the chemical, physical, and biological properties of the biomaterials used in bone regeneration are highly relevant. Mesoporous bioactive glass (MBG) with hierarchical porous structure possesses a large amount of specific surface area and a large total volume, which facilitate the bioactive behavior of MBG. It is reported that the MBG scaffolds can be functionalized to enhance bone tissue regeneration by grafting amino or carboxylic groups onto scaffolds, which contributes the advantages of novel MBG scaffolds,¹⁵ and this provides a new strategy for bone tumor therapy in orthopedics. For example, Zhang and Zhu et al. revealed multifunctional Fe₃O₄/MBG/PCL composite scaffolds with enhanced bone regeneration, drug delivery and hyperthermia activity.¹⁶ Moreover, the novel fabrication technique also endows MBG scaffolds with promising advantages. For instance, a strontium-containing MBG scaffold was reported to increase in vivo bone regeneration, in which a three-dimensional (3-D) printing technique was used to control the architecture of MBG scaffolds.¹⁷ However, like other inorganic scaffolds, MBG scaffolds are very brittle, mechanically weak, and degrade too fast if coupled with an unstable surface/interface. This impairs cell attachment and growth.^18,19 MBG/polymer composite scaffolds have better physiochemical and biological properties than those of the ordinary MBG scaffolds.^20,21

An alternative strategy to systemic antibiotic therapy was in situ implantation of a local antibiotic delivery system, involving incorporation of the antibiotics into scaffolds. For example, Phetnin et al. fabricated antibacterial mesoporous bioactive glass microspheres incorporated with silver,²² while Ye et al. prepared a copper-containing mesoporous bioactive glass coating on implants for improving antibacterial activity.²³ Some other research also revealed antibacterial meso–macroporous glass scaffolds enriched with ZnO which presents good cytocompatibility.²⁴ Several other studies have shown that MBG scaffolds that release antibiotics can prolong the duration of delivery during bone repair.⁴ The complications associated with this type of treatment (in situ implantation of a local antibiotic delivery system), however, necessitate large amounts of antibiotics taken over a long period of time, which can foster the development of antibiotic-resistant bacteria.^25,26 Therefore, effective antimicrobial agents suitable for this type of treatment must be developed, especially agents that remain effective against antibiotic-resistant organisms.

One water-soluble chitosan derivative, 2-hydroxypropyltrimethyl ammonium chloride chitosan (HACC), was produced through the reaction of chitosan with glycidyl trimethylammonium chloride. This chitosan derivative possesses excellent antibacterial ability against Staphylococcus aureus (S. aureus), Escherichia coli, and Candida albicans, etc.^25,27–30 HACC can be further modified through reaction with N-methylolacrylamide. This produces a fiber-reactive chitosan derivative called O-acrylamidomethyl-HACC (NMA-HACC) (Fig. 1).³⁰ The acrylamidomethyl group is fiber-reactive group. It can form covalent bonds with cellulose under alkaline conditions. Recently, a series of implant materials and coatings with resisting infection and promoting bone growth have been developed for orthopedic surgery.⁵ Silk is currently being reassessed for use as a scaffold material due to its excellent biodegradability and biocompatibility, high mechanical property, and versatility.^31–33 For example, silk films and fiber matrices have a wide range of potential biomedical applications such as attaching human mesenchymal stem cells (hMSCs).³²

Fig. 1 Synthesis of HACC, NMA-HACC, and NMA-HACC–silk (NHS).

Herein, in order to facilitate bone regeneration (by MBG) and to resist bacteria for a long time (by NMA-HACC–silk), we prepared an NMA-HACC–silk modified MBG scaffold (abbreviated as MBG-NHS), which possesses better mechanical properties, slower degradation rate, significantly longer antibacterial activity, and better cell adhesion property comparing with the HACC or HACC–silk modified MBG scaffolds.

2. Material and methods

2.1. Materials

Triethyl phosphate (TEP), tetraethyl orthosilicate (TEOS), and HCl were obtained from Shanghai Lingfeng Chemical Reagent. Ca(NO₃)₂·4H₂O was purchased from Sinopharm Chemical Reagent. P123 (EO20-PO70-EO20) (M_w = 5800), chitosan, which had a molecular weight of 20.0 × 10⁴ and 90% N-deacetylation and O-acrylamidomethyl-N-[(2-hydroxy-3-trimethylammonium)propyl (48 wt% aq solution), was purchased from Aldrich Chemical Co. The glycidyl trimethylammonium chloride (GTMAC) was purchased from Shandong Sangong Chemical Co. Ltd.

2.2. Synthesis of HACC and NMA-HACC

HACCs with various degrees of substitution and 26% quaternary ammonium were prepared by chitosan and GTMAC as described previously.²⁸ Deacetylated chitosan (6.00 g, 37.0 mmol) was dispersed in 60 mL of isopropyl alcohol at 80 °C. Then GTMAC (111 mmol, 21.3 mL) was added in three portions (7.1 mL each) at 2 h intervals. After 10 h, the clear and yellowish solution was poured into cold acetone (200 mL) and stored in a refrigerator overnight. Then the acetone was decanted and the remaining gel-like raw product was dissolved in methanol (100 mL). The resulting solution was precipitated in 4 : 1 acetone–ethanol (250 mL). The white product was collected by filtration and further purified by washing with hot ethanol with a Soxhlet extractor. The final product was dried at 70 °C overnight.

O-Acrylamidomethyl-HACC (NMA-HACC) was synthesized according to a literature method.³⁰ The HACC (1.00 g, 3.20 mmol) was dissolved in 48 wt% aq N-methylolacrylamide solution (5.00 mL, 25.5 mmol). This solution contained 4-methoxyphenol (0.010 g) to inhibit polymerization. Then 0.680 g of NH₄Cl (12.8 mmol) was added to the solution. The solutions were allowed to react for 16 min at 140 °C. Methanol (15 mL) was then added to the solution and stirred for 10 s. The product was precipitated in acetone (100 mL), washed thoroughly with a mixture of 1 : 1 acetone–ethanol and then washed again with ether. The white product was dried at 40 °C under vacuum for 2 days.

2.3. Synthesis of NMA-HACC–silk (NHS)

The pad-dry-cure method was used to attach the NMA-HACC to the silk (Fig. 1). The silk was washed with distilled water before reaction with NMA-HACC. Distilled water containing 1.00 g L⁻¹ non-ionic detergent (Kierlon NB-MFB, BASF) was added. Then the silk was rinsed three times with distilled water and dried for 30 min under vacuum at 70 °C. The pad solutions were prepared by dissolving an alkaline catalyst (NaHCO₃, 1.0 wt%) and NMA-HACC in distilled water. Silk was padded with the finished solutions at 100% WPU with a laboratory padder (Werner Mathis AG). The padded samples were dried at 70 °C for 4 min, then cured in a laboratory oven (Werner Mathis AG) at 150 °C for 5 min. The treated silk (NMA-HACC–silk) was washed with distilled water to neutral pH and air-dried at room temperature.

2.4. Preparation of HACC, HACC–silk, and NMA-HACC–silk-modified MBG scaffolds

2.4.1 Preparation of MBG scaffolds. Porous mesopore bioglass (MBG) scaffolds were prepared using P123 and polyurethane sponges as described previously.²⁰ In a typical procedure, 6.7 g of TEOS, 4.0 g of P123, 1.4 g of Ca(NO₃)₂·4H₂O, 0.73 g of TEP, and 1.0 g of 0.5 M HCl were dissolved in 60 g of ethanol (Si/Ca/P = 80 : 15 : 5, molar ratio). This mixture was stirred at room temperature for 24 h. The polyurethane sponges (20 ppi) were immersed in this solution for 10 min. This mixture was then transferred to a Petri dish. Then excess solution was removed. The mixture was allowed to evaporate at room temperature for 12 h. This process was repeated for three times. The samples were dried completely and then calcined at 650 °C for 5 h to afford the MBG scaffolds.

2.4.2 Preparation of HACC modified MBG scaffolds (MBG-H). After the scaffolds were calcined, they were immersed in aqueous HACC solution (10.0% w/v) for 3 min and then centrifuged at 1000 rpm to remove excess HACC solution. The scaffolds were then air-dried under a fume hood for 24 h. They were then dried at 40 °C for 24 h. The HACC-modified scaffolds are labeled as MBG-H.

2.4.3 Preparation of HACC–silk-modified MBG (MBG-HS) and NMA-HACC–silk-modified MBG (MBG-NHS) scaffolds. Silk protein and NMA-HACC–silk solutions were prepared as described previously.³⁴ MBG-H scaffolds were immersed in silk to prepare HACC–silk. MBG scaffolds were immersed in NMA-HACC–silk solution (10%) and used to modify MBG scaffolds (MBG-HS and MBG-NHS). The procedure used to modify the scaffolds was the same as the one used to produce MBG-H.

2.5. Characterization

2.5.1 Physical–chemical properties of the scaffolds. Small-angle X-ray diffraction (SAXRD, Rigaku D/Max-2550 V diffractometer, Cu-Kα radiation) and transmission electron microscopy (TEM, JEOL 2010) were used to characterize the mesopore-channel microstructure of the scaffolds. Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) analysis were used to assess nano-pore size, nano-pore distribution, and specific surface area using N₂ adsorption–desorption isotherms and a Micromeritics Tristar 3000 pore analyzer at 77 K. The structure of the pores and the surface microstructure of the walls of the pores in the scaffolds were also assessed using a scanning electron microscope (SEM). Chemical composition was analyzed using Fourier transform infrared spectra (FTIR). Compression tests were performed using a universal material testing machine (Instron 5567, Instron Corp., U.S.). This machine was equipped with a model 2519-104 force transducer at a crosshead speed of 0.5 mm min⁻¹. SRμCT measurements were performed at beamline BL13W of SSRF (Shanghai, China). A monochromatic beam at 30 keV and a sample-to-detector distance of 1.5 m were used. A 4000 × 2500 CCD detector with a 6 mm pixel size was used to record images. Some 1200 projections were taken within an angular range of 180°. Total exposure lasted 8 s per projection. A filtered back-projection was used to reconstruct the 3D structure. These images were redigitized using 8 bit data format. This format was proportional to the measured attenuation coefficients of the voxels. Then VG Studio MAX 2.0 software (Volume Graphics, Heidelberg, Germany) was used to image the tomographic data. Thermal behavior of the samples were investigated using thermogravimetric (TG) using a SETARAM Model (no. LabsysTM TG-DT16, SETARAM Instrumentation).

2.5.2 In vitro bioactivity test. Simulated body fluid (SBF) solution was prepared as described previously.³⁵ Three samples were taken from each scaffold and soaked in 50 mL of SBF solution at 37 °C with shaking at 100 rpm. After 3 days, the specimens were removed and gently rinsed with distilled water. They were then dried overnight at 60 °C. The samples were then sputter-coated with gold. The morphology of the apatite formed on the material surface was determined using an environmental scanning electron microscope (FEI Quanta 250).

2.5.3 Antibacterial assays. The ability of the scaffolds to repel S. epidermidis (ATCC35984) and Staphylococcus aureus (ATCC43300) was assessed. The inocula were prepared by adjusting the concentration of an overnight bacterial broth to 1 × 10⁶ CFUs mL⁻¹ in TSB, as described by McFarland. One milliliter of suspension was added to a 48-well plate (Costar3548, U.S.) containing four kinds of scaffolds. It was incubated at 37 °C and constantly agitated at 100 rpm for 1 day and 7 days. The TBS was changed every day. After 1 day and 7 days, the scaffolds were removed with sterile forceps and placed in empty 24-well plates. They were then washed three times with PBS and placed in new 48-well plates, where they were stained with 0.5 mL of combination dyes (LIVE/DEAD BacLight Bacteria Viability Kits, Molecular Probes, L13152). They were then analyzed via confocal laser scanning microscopy (CLSM, Nikon A1R, Japan). The viable bacteria appeared green and the nonviable bacteria were red. The images were taken at random positions on the surfaces of the scaffolds.

Total bacterial growth was used as a negative indicator of scaffold bactericidal properties. Bacteria were incubated with the scaffolds at a concentration of 1 × 10⁶ CFUs mL⁻¹. The bacteria remaining on the substrates were dislodged using mild ultrasonication and rapid vortex mixing. The total bacterial population was assessed using serial dilution of 0.1 mL portions of the bacterial suspension. After 1, 3 and 7 days, the number of total bacteria was calculated. The MBG control suspension served as a standard.

2.5.4 Cell morphology, viability, and proliferation. First, 1.0 mL of the cell suspension was seeded into a 48-well plate (Costar3548, U.S.). Cell density was 2 × 10⁴ viable hMSCs, and the mixture contained four different kinds of scaffolds. The mixture was incubated at 37 °C in a humidified atmosphere of 5% CO₂ and 95% air.

The proliferation of hMSCs on the substrates was evaluated using a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay after 1, 3, and 7 days of culture. At specific points in time, 0.1 mL of MTT solution was added and the specimens were incubated at 37 °C, at which point they formed formazan, which was then dissolved using DMSO. Optical density (OD) was measured at 570 nm using an automated plate reader (Synergy HT Multidetection Microplate). The mean absorbance recorded from the medium-control well was determined using the test absorbance values.

After 24 h of incubation with the substrates, the cells on the surfaces of those substrates were washed three times with phosphate-buffered saline (PBS) and then fixed using 3.7% formaldehyde for 15 min at room temperature. The cells were permeabilized with 0.1% Triton X-100 in PBS for another 15 min. They were then re-washed three times with PBS and incubated with FITC-phalloidin (Abcam, Sigma-Aldrich) for 1 h. They were then re-washed once with PBS. The cell nuclei were stained with DAPI. Cell morphology and spreading were recorded using CLSM.

Cell viability was evaluated using a live/dead assay kit (Abcam, Sigma-Aldrich). This kit was used in accordance with the manufacturer's standard protocol. The cells and scaffold constructs were washed twice with PBS and then incubated in standard working solution at room temperature for 10 min. The constructs were then washed twice with PBS and observed using CLSM.

2.5.5 Statistical analysis. One-way analysis of variance (ANOVA) was used to evaluate significant differences between means in the measured data. Scheffe's multiple comparison test was used to determine the significance of the standard deviations in the measured data collected from each specimen under different experimental conditions. P < 0.05 was considered indicative of statistical significance.

3. Results

3.1. Characterization of MBG by BET and TEM

The nitrogen adsorption–desorption isotherm of MBG is shown in Fig. 2. This isotherm can be attributed to type IV behavior, and it is characteristic of capillary condensation in mesopores. The present sample shows type-H1 hysteresis loops in the mesopore range. These loops have cylindrical pores. The inset in Fig. 2 shows the distribution of pore size as calculated from the adsorption branch using the Barrett–Joyner–Halenda (BJH) model. The BET-specific surface area of the scaffold was 684.8 m² g⁻¹ and the total pore volume was 0.49 cm³. A narrow peak was observed in the BJH pore-size distribution curve around 5.94 nm, indicating that the mesopores were uniform in size.

Fig. 2 N₂ sorption isotherms and pore size distribution (inset) of synthesized mesoporous bioactive glass (MBG).

The mesoporous structure of MBG was further evaluated by TEM. Fig. 3 showed the TEM images of as-prepared MBG particles, revealing the ordered pore arrays in MBG. A uniform, homogeneously distributed 2D-hexagonal channel was observed with approximately 5 nm in diameter, which contributes strong adsorption properties as potential carrier material for scaffold fabrication in next step. This is consistent with the N₂ adsorption isotherm analyses.

Fig. 3 TEM images of the prepared MBG scaffolds showing their highly ordered mesopore-channel structure. (A) Lower magnification (B) higher magnification.

3.2. Characterization of NHS modified scaffolds

SRμCT imaging showed the average pore size of the MBG-NHS scaffolds was about 300 μm. As shown in Fig. 4, the synthesized MBG-NHS scaffold was made up of well-interconnected pore channels, which facilitates the ingrowth of cells and transportation of nutrients, leading to better bone healing.

Fig. 4 Macroporous network and strut microstructure of MBG-NHS scaffolds as shown using SRμCT. (A) Macroporous network of MBG-NHS, (B) electronic slice of MBG-NHS.

The FTIR spectra for scaffolds are shown in Fig. 5. Unlike those of MBG, spectra for the group containing HACC (MBG-H, MBG-HS, and MBG-NHS) showed characteristic absorption peaks for C–H of the trimethylammonium group at 1480 cm⁻¹. Absorption was also detected at 1587 cm⁻¹, which was in this case attributable to the vibration of N–H bonds. These results demonstrate that HACC had been grafted onto the surfaces of MBG-H, MBG-HS, and MBG-NHS. FTIR analysis of the MBG-HS and MBG-NHS scaffolds indicated two peaks characteristic of silk at 1550 cm⁻¹ and 1650 cm⁻¹. This confirmed that the surfaces of the MBG scaffolds had been successfully coated with silk and NMA-HACC–silk.

Fig. 5 FTIR spectra of silk, MBG-H, MBG-HS and MBG-NHS.

Simultaneous TG thermograms were recorded up to 950 °C (Fig. 6). Based on these TG curves, the thermal decomposition (total weight loss) of the MBG is 16%. The weight loss of MBG-HS can be divided into two stages, the first stage (40–220 °C) is 9.6% and the second stage (290–600 °C) is 11.8%, indicating 9.6% of HACC and 11.8% of silk in the MBG-HS. The total weight losses of the MBG-H and MBG-NHS were 43.6% and 44.8%, respectively, which indicated 43.6% of HACC and 44.8% of NHS in MBG-H and MBG-NHS, respectively.

Fig. 6 TGA of MBG, MBG-H, MBG-HS and MBG-NHS.

3.3. Mechanical properties

As shown in Fig. 7, the compressive strengths of pure MBG, MBG-H, MBG-HS, and MBG-NHS scaffolds were 62.8 kPa, 148.5 kPa, 285.6 kPa, and 302.8 kPa, respectively, suggesting that HACC, silk, and NMA-HACC–silk modifications can significantly increase the mechanical strength of MBG scaffolds. The strengths of these scaffolds were also assessed after soaking in SBF at 37 °C for 3 days. The compressive strengths of pure MBG, MBG-H, MBG-HS, and MBG-NHS scaffolds were 74.3 kPa, 75.6 kPa, 234.6 kPa, and 284.6 kPa, respectively after soaking in SBF. Compared with the scaffolds before soaking in SBF, the compressive strengths of the MBG scaffolds after soaking increased slightly, but that of MBG-H scaffolds decreased significantly, and that of the MBG-HS and MBG-NHS scaffolds decreased slightly. This is because water-soluble HACC was hydrated from MBG-H scaffolds during soaking. This caused a significant decrease in the mechanical strength of MBG-H scaffolds. However, silk, and NMA-HACC–silk coating preserved the mechanical strength of MBG-HS and MBG-NHS scaffolds.

Fig. 7 Mechanical properties of MBG, MBG-H, MBG-HS, and MBG-NHS scaffolds before and after soaking in SBF for 3 days.

3.4. Surface morphologies of scaffolds before and after soaking in SBF

The morphologies of the scaffolds are shown in Fig. 8A1, B1, C1 and D1. Pure MBG scaffolds showed many collapsed and discontinuous pore networks because of the brittleness of this material. The pore walls of the MBG-H, MBG-HS, and MBG-NHS scaffolds remained intact. This indicated that HACC, silk, and NMA-HACC–silk could render the MBG less brittle.

Fig. 8 Surface microstructure of (A1, A2, and A3) MBG, (B1, B2, and B3) MBG-H, (C1, C2, and C3) MBG-HS, and (D1, D2, and D3) MBG-NHS scaffolds at different magnification levels. SEM images of (A4) MBG, (B4) MBG-H, (C4) MBG-HS, and (D4) MBG-NHS scaffolds were obtained after soaking in SBF for 7 days.

Higher magnification SEM images showed smooth pore walls in the MBG-H, MBG-HS, and MBG-NHS scaffolds and no obvious microparticle attachment (Fig. 8B3, C3 and D3). Fig. 8A4, B4, C4 and D4 show the formation of hydroxyapatite (HA) on the surface of each scaffold, but there were more and better ordered HA on the surfaces of MBG and MBG-H scaffolds than on those of MBG-HS and MBG-NHS. This is because HACC is readily water-soluble and can dissolve quickly in SBF without having any significant impact on the growth of HA on the surface of the scaffold. However, silk and NMA-HACC–silk dissolve slowly, and undissolved silk and NMA-HACC–silk on the surface of the scaffold were found to hinder the growth and arrangement of HA.

The pores of the scaffold in the SEM images are consistent with that by SRμCT imaging (Fig. 4), suggesting that these channels may facilitate cell ingrowth and the transportation of nutrients for better bone healing.

3.5. Antibacterial efficacy of different scaffolds

MBG-H, MBG-HS, and MBG-NHS can inhibit the adherence of S. epidermidis (ATCC35984) and Staphylococcus aureus (ATCC43300) to a greater degree than MBG (Fig. 9) after 1 day of incubation in the bacterial broth.

Fig. 9 Confocal laser scanning microscopy (CLSM) analysis of bacterial viability on the surfaces of different MBG-based scaffolds. Bacteria were stained with green fluorescent SYTO 9 and red fluorescent propidium iodide, producing in live cells that appeared green and dead cells that appeared red under CLSM. Magnification, ×400.

CLSM was used to evaluate the number of S. epidermidis (ATCC35984) and Staphylococcus aureus (ATCC43300) on the surface of the four types of scaffolds after 1 day and 7 days of incubation. The results show that the antibacterial effects of MBG-H, MBG-HS, and MBG-NHS were significantly more pronounced than those of MBG at 24 h for both S. epidermidis (ATCC35984) and Staphylococcus aureus (ATCC43300) (Fig. 9). No significant differences in the density of the fluorescence of the number of bacteria were detected among the surfaces of the MBG-H, MBG-HS, or MBG-NHS scaffolds after 24 h.

The areas with the most intense fluorescence were bacterial foci, suggesting more bacteria. Areas of less intense fluorescence on the surfaces of MBG-H, MBG-HS, and MBG-NHS surface at the 24 h mark indicated less bacteria.

However, after 7 days, the antibacterial effects of MBG-NHS were significantly more pronounced than those of MBG, MBG-H, and MBG-HS (Fig. 9). There were very few bacterial colonies distributed on the surface of MBG-NHS after 24 h and after 7 days, indicating that there were few bacteria on the scaffold's surface and that the antibacterial effects of MBG-NHS could be maintained over long periods of time.

In these assays, MBG exhibited barely any antibacterial activity. Although the antibacterial activity of MBG-H and MBG-HS was equivalent to that of MBG-NHS at 24 h, it had decreased significantly by day 7, at that time MBG-H and MBG-HS, like MBG, had no effective antibacterial action. However, the antibacterial activity of MBG-NHS remained very high at both 24 h and 7 days. This indicates that although HACC can significantly increase the antibacterial activity of MBG, this antibacterial effect does not last a long period of time.

The number of ATCC35984 bacteria in suspension was further determined on day 1, day 3, and day 7 and the number of total bacteria was calculated and normalized to that of the MBG control suspension (Fig. 10). On day 1, MBG-H, MBG-HS, and MBG-NHS showed significantly less bacteria than that on the MBG surface. On day 3, MBG-HS and MBG-NHS remained highly effective against bacteria but MBG-H showed less antimicrobial activity than it had earlier. The antibacterial effect of MBG-NHS was found to be significantly more pronounced than those of MBG, MBG-H, or MBG-HS. These results suggest that the MBG-NHS has long-lasting antibacterial activity.

Fig. 10 The number of live ATCC35984 bacteria in suspension after contacting with four MBG-based scaffolds for 1 day, 3 days, and 7 days. The number of cells is expressed relative to the number detected after contacting with MBG. The data are representative of the results of three independent experiments and are expressed as the means ± SD. (#) denotes a significant difference from MBG (P < 0.05).

3.6. Biocompatibility of scaffolds

Fig. 11 shows the hMSCs on four MBG-based scaffolds. These cells exhibited polygonal and fusiform morphology. Those on the surface of the MBG-HS and MBG-NHS also showed clustering, confluence, and multi-layering morphology (Fig. 11). There were more cells on the surfaces of the MBG-NHS and MBG-HS than on those of MBG and MBG-H, suggesting excellent biocompatibility of MBG-HS.

Fig. 11 The cytoskeletal morphology of the hMSCs on the (A) MBG, (B) MBG-H, (C) MBG-HS, and (D) MBG-NHS scaffolds. Representative images of cells stained with FITC-phalloidin for actin filaments (green) and nuclei counterstained with DAPI (blue).

The cell proliferation data collected after 1, 3, and 7 days were normalized to data of MBG collected on day 1. Cells on the MBG-NHS scaffold showed a proliferation rate notably higher than that of cells on other scaffolds at all time intervals (Fig. 12). On day 3, the cells on the MBG and MBG-H scaffolds showed lower proliferation rates than those on the MBG-HS and MBG-NHS scaffolds. On day 7, the proliferation rate of cells on the MBG-NHS scaffold became significantly higher than those of cells on the MBG, MBG-H, and MBG-HS scaffolds. The relative proliferation rate of the hMSCs on the MBG, MBG-MBG-HS, and MBG-NHS scaffolds tended to increase from day 1 to day 7.

Fig. 12 hMSC proliferation on MBG, MBG-H, MBG-HS, and MBG-NHS scaffolds. *P < 0.05 compared with the MBG group.

Furthermore, hMSCs cells were viable 3 days after being seeded onto all four kinds of scaffolds, as indicated by a visualized fluorescent live/dead assay (Fig. 13). Live hMSCs were stained green and appeared to have adhered, and dead cells were stained red. A few dead cells (red spots) were observable on the MBG and MBG-H scaffold, indicating the presence of several dead cells with damaged membranes. Almost no dead cells were visible on the MBG-HS and MBG-NHS scaffolds.

Fig. 13 Fluorescence staining of hMSCs with a LIVE/DEAD kit 3 days after seeding on (A) MBG, (B) MBG-H, (C) MBG-HS, and (D) MBG-NHS scaffolds.

4. Discussion

Infection can seriously compromise the healing of bone fractures.³⁶ It is especially common in limb-saving surgery involving external fixation after excision of skeletal tumors. This is because the patients are immunocompromised from postoperative chemotherapy.³⁷ Most cases of chronic osteomyelitis can be treated only through a combination of surgical debridement and antibiotic therapy.³⁸ In this way, a system designed to manage infected segmental defects must incorporate the elimination of bacteria together and the regeneration of bone. Conventional antibiotic therapy has many drawbacks when used to treat osteomyelitis. It is extremely difficult to produce adequate concentrations of antibiotics at the site of infection through parenteral administration alone. Usually, high serum concentrations must be used. This involves considerable risk of toxicity.³⁹ These antibiotics are also highly expensive, which taxes the health care system. Worse, much of the dose is excreted before it can affect the diseased bone. Monitoring intravenous antibiotic therapy is difficult in both hospital and outpatient settings.⁴⁰ Within the past few years, in situ implantation of local antibiotic delivery systems has been proposed as an alternative to systemic antibiotic therapy. This system first eliminates bacteria and later reduces the dead space.^41,42 Local antibiotics can reach considerably higher concentrations at the contaminated site than systemic parenteral treatment. This increases the effectiveness of the treatment of the infection. It can also decrease the risk of systemic toxicity. However, overuse of antibiotics has led to the emergence of antibiotic-resistant bacteria. Two of the most notable of these are the methicillin-resistant Staphylococcus aureus (MRSA) and Staphylococcus epidermidis (MRSE).⁴³ This shows that more effective antimicrobial agents and better methods of fostering bone regeneration must be investigated.

In the present study, pins with antibacterial activity were built by coating NMA-HACC–silk onto the surface of MBG. Then scaffolds were prepared using P123 and polyurethane sponges. TEM and BET analysis showed the MBG scaffolds to have an ordered channel structure with channels approximately 5 nm in length. This profoundly increased total surface area and total pore volume, as indicated by the increases in degradation, apatite-mineralization, and in vivo bioactivity.^44,45 HACC can enhance antibacterial activity.^25,29,30 Tang found HACC to be only slightly cytotoxic and that it did not interfere with the proliferation or osteogenic differentiation of human mesenchymal stem cells.²⁸ To increase the MBG's mechanical strength, to extend the duration of antibacterial activity, and to decelerate degradation, silk was treated with HACC in the presence of an alkaline catalyst. NMA-HACC–silk was then coated on the surface of MBG. NMA-HACC–silk modification was found to render the MBG less brittle (Fig. 7). It also increased the mechanical strength of MBG scaffolds, and it retained this mechanical strength even after immersion in SBF. The mechanical strength of the MBG-HACC scaffolds decreased considerably (Fig. 7). This may be because NMA-HACC–silk linked the inorganic phase and retained its strength because of its poor solubility in water. SRμCT imaging showed the MBG scaffolds modified with NMA-HACC–silk to be highly porous. They exhibited interconnected macroporous networks with a modal interconnected pore diameter of 200–400 μm (Fig. 4). This suggested that NMA-HACC–silk increased the mechanical strength of MBG scaffolds without affecting porosity. Highly porous microstructures with large surface areas and interconnected pores foster tissue ingrowth. NMA-HACC–silk, HACC/silk, and HACC were coated on the surfaces of MBGs. This produced significant antibacterial activity as detected on day 1. However, antibacterial activity of MBG-H and MBG-HS had declined by day 3. The NMA-HACC–silk modification may have slowed the rate of degradation, facilitating better cell adhesion. Overall, the present results demonstrated that NMA-HACC–silk modification of MBG successfully inhibited bacterial adherence. Unexpectedly, NMA-HACC–silk modification of MBG scaffolds was also found to promote adherence and proliferation of hMSCs. The present results indicate that MBG-NHS has both antimicrobial and in vitro bioactivity. This makes it a promising therapeutic material suitable for use in orthopedic surgery.

5. Conclusions

MBG scaffolds modified with HACC, HACC–silk, and NMA-HACC–silk are successfully prepared. HACC, silk/HACC, and NMA-HACC–silk modification rendered the pore network more continuous and increased the mechanical strength of MBG scaffolds. The scaffolds were soaked in SBF for 24 h, which decreased the mechanical strength of MBG-HACC scaffolds considerably, but that of the MBG-HS and MBG-NHS scaffolds decreased only slightly. MBG was not found to have any antibacterial activity, and MBG modified with HACC or HACC/silk showed only short-term antibacterial activity, but MBG modified with NMA-HACC–silk showed both short-term and long-term antibacterial activity. Human mesenchymal stem cells (hMSCs) were found to adhere to the walls of the pores of all the four types of scaffolds. However, HACC/silk and NMA-HACC–silk coating was found to promote the attachment and proliferation of hMSCs.

This may be why they degraded more slowly. MBG-NHS shows significant promise as a material for use in bone repair in that it has an interconnected pore structure, good mechanical strength, minimal susceptibility to degradation, and long-lasting antibacterial activity.

All in all, the developed MBG-NHS scaffolds presented appropriate physical structure, good bioactivity and biocompatibility and proper degradation rate. Moreover, the scaffolds were found to show both good mechanical strength and long-lasting antibacterial activity. These results indicate that the NMA-HACC–silk modified MBG scaffolds system may be clinically useful for bone regeneration in the case of large bone defects.

Acknowledgements

We acknowledge the support of the Key Projects for Basic Research Shanghai Committee of Science and Technology (No. 12441902803), the Research Plan of Shanghai Municipal Commission of Health and Family Planning (No. 201540380), National Natural Science Foundation of China (No. 81171794 and No. 81571887) and the Science Foundation for Post Doctorate Research from Secondary Military Medical University and China Postdoctoral Science Foundation Grant (No. 2016M592936).

References

1. S. M. M. Roosa, J. M. Kemppainen, E. N. Moffitt, P. H. Krebsbach and S. J. Hollister, J. Biomed. Mater. Res., Part A, 2010, 92, 359–368.
2. N. Arya, V. Sardana, M. Saxena, A. Rangarajan and D. S. Katti, J. R. Soc., Interface, 2012, 9, 3288–3302.
3. M. Wang and B. Duan, J. R. Soc., Interface, 2010, 7(5), S615–S629.
4. L. Xia, X. Wang, L. Zhang, H. Chen and J. Shi, J. Biomed. Mater. Res., Part B, 2009, 89, 148–154.
5. M. B. Nair, J. D. Kretlow, A. G. Mikos and F. K. Kasper, Curr. Opin. Biotechnol., 2011, 22, 721–725.
6. Z. Panyu, X. Yan, C. Xiaosong, W. Panfeng, X. Yang and X. Shuogui, Biomaterials, 2014, 35, 10033–10045.
7. A. Trampuz and W. Zimmerli, Injury, 2006, 37(2), S59–S66.
8. D. J. Stinner, J. A. Keeney, J. R. Hsu, J. K. Rush, M. S. Cho, J. C. Wenke and J. R. Ficke, J. Surg. Orthop. Adv., 2010, 19, 49–53.
9. T. G. Kwon, S. Y. Choi, B. C. Ahn and C. H. An, Journal of Oral and Maxillofacial Surgery, Medicine, and Pathology, 2013, 25, 214–220.
10. L. Bing, K. V. Brown, J. C. Wenke and S. A. Guelcher, J. Controlled Release, 2010, 145, 221–230.
11. H. Winkler, K. Kaudela, A. Stoiber and F. Menschik, Cell Tissue Banking, 2006, 7, 319–323.
12. M. B. Toma, K. M. Smith, C. A. Martin and R. P. Rapp, Orthopedics, 2006, 29, 497–501.
13. Y. Shikinami, K. Okazaki, M. Saito, M. Okuno, S. Hasegawa, J. Tamura, S. Fujibayashi and T. Nakamura, J. R. Soc., Interface, 2006, 3, 805–821.
14. A. Gloria, T. Russo, U. D'Amora, S. Zeppetelli, T. D'Alessandro, M. Sandri, M. Bañobre-López, Y. Piñeiro-Redondo, M. Uhlarz and A. Tampieri, J. R. Soc., Interface, 2013, 10, 20120833.
15. X. Zhang, D. Zeng, N. Li, J. Wen, X. Jiang, C. Liu and Y. Li, Sci. Rep., 2016, 6, 19361.
16. J. Zhang, S. Zhao, M. Zhu, Y. Zhu, Y. Zhang, Z. Liu and C. Zhang, J. Mater. Chem. B, 2014, 2, 7583–7595.
17. J. Zhang, S. Zhao, Y. Zhu, Y. Huang, M. Zhu, C. Tao and C. Zhang, Acta Biomater., 2014, 10, 2269–2281.
18. C. Wu, Y. Zhang, Y. Zhu, T. Friis and Y. Xiao, Biomaterials, 2010, 31, 3429–3438.
19. H. Yun, S. Kim, Y. Hyun, S. Heo and J. Shin, Chem. Mater., 2007, 19, 6363–6366.
20. M. Zhu, L. Zhang, Q. He, J. Zhao, L. Guo and J. Shi, J. Mater. Chem., 2011, 21, 1064–1072.
21. C. G. Zalavras, M. J. Patzakis and P. Holtom, Clin. Orthop. Relat. Res., 2004, 427, 86–93.
22. R. Phetnin and S. T. Rattanachan, J. Sol-Gel Sci. Technol., 2015, 75, 279–290.
23. J. Ye, J. He, C. Wang, K. Yao and Z. Gou, Biotechnol. Lett., 2014, 36, 961–968.
24. S. Sánchez-Salcedo, S. Shruti, A. J. Salinas, G. Malavasi, L. Menabue and M. Vallet-Regí, J. Mater. Chem. B, 2014, 2, 4836.
25. P. Coates, S. Vyakrnam, E. A. Eady, C. E. Jones, J. H. Cove and W. J. Cunliffe, Br. J. Dermatol., 2002, 146, 840–848.
26. H. Tan, Z. Peng, Q. Li, X. Xu, S. Guo and T. Tang, Biomaterials, 2012, 33, 365–377.
27. Z. X. Peng, L. Wang, L. Du, S. R. Guo, X. Q. Wang and T. T. Tang, Carbohydr. Polym., 2010, 81, 275–283.
28. P. Zhou, Y. Xia, J. Wang, C. Liang, L. Yu, W. Tang, S. Gu and S. Xu, J. Mater. Chem. B, 2013, 1, 685–692.
29. H. Tan, S. Guo, S. Yang, X. Xu and T. Tang, Acta Biomater., 2012, 8, 2166–2174.
30. S. H. Lim and S. M. Hudson, Carbohydr. Polym., 2004, 56, 227–234.
31. S. Sofia, M. B. McCarthy, G. Gronowicz and D. L. Kaplan, J. Biomed. Mater. Res., 2001, 54, 139–148.
32. K. Y. Lee, S. J. Kong, W. H. Park, W. S. Ha and I. C. Kwon, J. Biomater. Sci., Polym. Ed., 1998, 9, 905–914.
33. K. H. Lam, A. J. Nijenhuis, H. Bartels, A. R. Postema, M. F. Jonkman, A. J. Pennings and P. Nieuwenhuis, J. Appl. Biomater., 1995, 6, 191–197.
34. K. Tadashi and T. Hiroaki, Biomaterials, 2006, 27, 2907–2915.
35. M. P. Ginebra, T. Traykova and J. A. Planell, J. Controlled Release, 2006, 113, 102–110.
36. T. Shirai, H. Tsuchiya, T. Shimizu, K. Ohtani, Y. Zen and K. Tomita, J. Biomed. Mater. Res., Part B, 2009, 91, 373–380.
37. L. G. Howard and D. G. Anderson, et al., Arch. Surg., 1950, 60, 112–124.
38. D. S. Kahn and K. P. Pritzker, Clin. Orthop. Relat. Res., 1973, 96, 12–19.
39. T. I. Braun and B. Lorber, in Orthopedic Infection, ed. D. Schlossberg, Springer, New York, NY, 1988, pp. 9–20,  DOI:10.1007/978-1-4612-3870-6_2.
40. C. T. Laurencin, T. Gerhart, P. Witschger, R. Satcher, A. Domb, A. E. Rosenberg, P. Hanff, L. Edsberg, W. Hayes and R. Langer, J. Orthop. Res., 1993, 11, 256–262.
41. S. P. Mohanty, M. N. Kumar and N. S. Murthy, J. Orthop. Surg., 2003, 11, 73–79.
42. J. D. Blaha, J. H. Calhoun, C. L. Nelson, S. L. Henry, D. Seligson, J. L. Esterhai Jr, R. B. Heppenstall, J. Mader, R. P. Evans and J. Wilkins, et al., Clin. Orthop. Relat. Res., 1993, 295, 8–12.
43. S. Zhilong, N. Koon Gee, K. En-Tang, P. Chyekhoon and W. Wilson, Tissue Eng., Part A, 2009, 15, 417–426.
44. T. J. Wadas, M. Eiblmaier, A. Zheleznyak, C. D. Sherman, R. Ferdani, K. Liang, S. Achilefu and C. J. Anderson, J. Nucl. Med., 2008, 49, 1819–1827.
45. X. Yan, X. Huang, C. Yu, H. Deng, Y. Wang, Z. Zhang, S. Qiao, G. Lu and D. Zhao, Biomaterials, 2006, 27, 3396–3403.

---

Footnote

† These authors contributed equally to this work.

---