Enhancement of compressive strength and cytocompatibility using apatite coated hexagonal mesoporous silica/poly(lactic acid-glycolic acid) microsphere scaffolds for bone tissue engineering

Abstract

In this study, hexagonal mesoporous silica (HMS) was composited with poly(lactic acid-glycolic acid) (PLGA) to enhance the compressive strength of pure PLGA microsphere scaffolds. The compressive strength of the HMS/PLGA composite scaffolds was significantly higher than that of pure PLGA scaffolds, which was more suitable for bone repair. However, the proliferation of mouse mesenchymal stem cells (MSCs) cultured on HMS/PLGA was inhibited when the HMS content was high. To promote the cytocompatibility of HMS/PLGA scaffolds, apatite were deposited on the scaffold's surface by an in vitro biomineralization process. After biomineralization, the cytocompatibility of mineralized scaffolds was significantly improved and without cytotoxicity. And the compressive strength of the mineralized scaffolds was close to that of the original scaffolds. Considering the suitable compressive strength and good cytocompatibility, the mineralized HMS/PLGA microsphere scaffolds would open a new door to major bone tissue engineering.

---

1. Introduction

In recent years, interest in bone tissue engineering (BTE) and its solutions have increased considerably. In particular, scaffolds have become fundamental tools in bone graft substitution. Heat sintered microsphere scaffolds, which are generic porous scaffolds, are fabricated easily and always exhibit excellent initial mechanical properties. The use of polymer based scaffolds in BTE is constantly increasing, since these biodegradable materials which are biocompatible and always meet minimal mechanical requirements have become fundamental tools to help the body rebuild damaged or diseased tissues. PLGA has been used most popularly as matrix material for preparation of BTE scaffolds because of its excellent biocompatibility and physical/mechanical properties, such as regulated degrading profiles.^1–3

However, poly(lactic acid-glycolic acid) (PLGA) that lack cell-affinitive moieties may be demanded as substrate materials for cell attachment. Therefore, modifications of polymer based on scaffolds are required. Hybridizing PLGA matrix with inorganic active materials to improve the cytocompatibility is a direct and simple modification method. The candidate inorganic materials could be calcium phosphate cements, hydroxyapatite, bioactive glass and mesoporous silica and so on.^4–7 Among them, silicon is a ubiquitous environmental element, which is known to play an important role in connective tissue metabolism especially in bone.^8–11 It was reported that silica-based materials could activate bone-related gene expression and stimulate osteoblast proliferation and differentiation.^12,13 Hexagonal mesoporous silica (HMS), a typical silica-based materials which shows great physical properties,¹⁴ was selected to hybridize with PLGA in this study. In addition, PLGA scaffolds which developed via solvent casting/particulate leaching, microspheres sintering and thermally induced phase separation have showed to be low mechanical strength, hybridizing HMS with PLGA will enhance the compressive strength of PLGA scaffolds.

However, HMS/PLGA scaffolds which are developed via microspheres sintering method, may have limited promotion towards cytocompatibility. For example, only part of HMS appears on the scaffolds surface. If HMS content is too high, these composite microspheres will be difficult to sinter together because there are few PLGA present on the surface. Depositing apatite layer on the HMS/PLGA surface to further improve the cytocompatibility could be a good idea. Since bio-glass can form a chemical bond to the bone by virtue of forming a calcium-apatite layer on the surface,¹⁵ coating of bone implants with apatite has been successfully exploited to induce osseointegration at the interface.¹⁶ It is also reported that presence of apatite enhances protein adsorption, improves osteoblast function¹⁷ and confers osteoconductivity¹⁸ to scaffolds. However, since conventional methods of incorporating ceramic always requires high temperature treatment, biomineralization is a moderate way to incorporate apatite within the polymer matrix at physiological temperature and pressure compared with conventional methods which involves apatite's controlled nucleation and growth on the surfaces of materials after immersion in simulated body fluid (SBF, whose ionic concentrations are roughly equal to that of plasma).¹⁹ Composite bone scaffolds containing this form of apatite may be more amenable for BTE.

In this study, we first fabricated and characterized HMS/PLGA microsphere scaffolds. And then we used SBF to fabricate the apatite/HMS/PLGA scaffolds. There has not been study of documenting the HMS/PLGA microsphere scaffolds and biomineralization towards HMS/PLGA microsphere scaffolds in five times SBF (5 × SBF). The mineralized scaffolds was characterized by different techniques such as scanning electron microscope (SEM) and Alizarin Red S (ARS) staining. Furthermore, we investigated the cytocompatibility essay in vitro of scaffolds before and after biomineralization. This work would serve as a foundation for further research in BTE applications.

2. Experiments

2.1 Materials

Tetraethyl orthosilicate (TEOS), ethyl alcohol (EtOH) and methylene chloride were purchased from Chemical Reagent Factory (Guangzhou, China). Dodecylamine (DDA) was supplied by SSS Reagent (Shanghai, China). Poly(lactic-co-glycolic acid) with a ratio of lactic to glycolic acid monomer units of 50 : 50 was purchased from Daigang Biomaterials (Jinan, China). This copolymer has an average molecular weight of 31 000 g mol⁻¹ with an inherent viscosity of 0.30 dl g⁻¹ in chloroform at 30 °C. Poly(vinyl alcohol) (PVA) was obtained from Sigma-Aldrich (Singapore). Several inorganic salts such as NaCl, KCl, MgCl₂·6H₂O, K₂HPO₄·3H₂O, NaHCO₃, CaCl₂ and Na₂SO₄, were obtained from Chengdu Kelong Co. Ltd (Chengdu, China). All cell-culture related reagents were purchased from Gibco, (Carlsbad, CA, USA). CCK-8 was obtained from Dojindo, (Kumamoto, Japan).

2.2 Synthesis methods

2.2.1 Preparation of HMS. HMS was synthesized based on the traditional process.²⁰ Briefly, a templating solution was prepared by dissolving 10 g of the dodecylamine (DDA) in 70 ml EtOH and 80 ml distilled water at room temperature, sonicating for a few min and magnetically stirring for 1 h. A second solution was prepared by mixing 44.6 ml hot tetraethyl orthosilicate (TEOS) and 40 ml EtOH under magnetic stirring at room temperature for about 30 min. The two solutions were then mixed at room temperature with stirring for about 18 h. The mixture was then aged for 30 min, washed with 800 ml distilled water, extracted with 600 ml HCl–EtOH for 4 h, followed by drying at 80 °C for 4 h; the dried product was calcinated.

2.2.2 Preparation of PLGA and HMS/PLGA microspheres scaffolds. PLGA microspheres were prepared using a single emulsion solvent evaporation technique (water/oil/water). One grams of PLGA was dissolved in 4 ml methylene chloride while the mixture was stirred as the oil phase. The oil phase was added dropwise to a 1.0% PVA aqueous solution, and the mixture was stirred for 12 h at 400 rpm, allowing the complete evaporation of the solvent. PLGA microspheres were isolated by vacuum filtration, and washed five times with deionized water.

HMS/PLGA microspheres were prepared using a single emulsion solvent evaporation method. Briefly, 1 g PLGA and HMS particles (0.1, 0.2, 0.25, 0.3, 0.4 and 0.5 g) were dissolved in 8 ml methylene chloride, and the mixture was sonicated for 1 min. The resultant mixture was then poured into a 1.0% PVA aqueous solution and stirred for 12 h, allowing the complete evaporation of the solvent. HMS/PLGA microspheres were isolated and washed five times with deionized water.

PLGA and HMS/PLGA microsphere sintered scaffolds were fabricated by pouring PLGA and HMS/PLGA microspheres into cylindrical molds, and then PLGA microspheres were sintered at 60 °C for 2 h and HMS/PLGA microspheres were sintered at 70 °C for 2 h.

2.2.3 Preparation of mineralized microspheres scaffolds. In vitro biomineralization study were conducted to fabricate mineralized HMS/PLGA scaffolds (HMS content was raised from 0% to 25%). 5 × Simulated Body Fluid (SBF) was prepared according to F. Barrere's recipe.²¹ Scaffolds (cylindrical scaffolds, diameter = 10 mm, height = 5 mm) were laid down in the 5 × SBF system at 37 °C for 2 days. The solution was replaced every 24 h. At the end of the incubation time, the samples were rinsed with deionized water and then dried under vacuum at room temperature. All the samples were stored at 4 °C for further use.

2.3 Characterization of HMS and microsphere scaffolds

2.3.1 Morphology of HMS and composite HMS/PLGA microspheres scaffolds. The morphology of HMS, PLGA and HMS/PLGA microspheres scaffolds before and after the process of biomineralization, were characterized by a scanning electron microscopy (SEM). The morphology of HMS was also characterized using a HRTEM (JEM2011) with an accelerating voltage of 50–200 kV, a point resolution of 0.23 nm and a lattice resolution of 0.14 nm. The energy dispersive X-ray spectrometer (EDX) (30XLFEG, Philips, The Netherlands) was used to analyze element content of these scaffolds. N₂ adsorption–desorption isotherms were obtained on a Micromeritics Tristar 3000 pore analyzer under continuous adsorption condition. Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods were used to determine the surface area, the pore size distribution and the pore volume.

2.3.2 Compressive testing. Cylindrical scaffolds (n = 6) with a length to diameter ratio of 2 : 1 (10 mm in length and 5 mm in diameter) were used for compressive testing. Compressive testing to failure was conducted using an Instron mechanical testing machine (Instron model 5544, Canton, MA) with a crosshead speed of 5 mm min⁻¹ at ambient temperature and humidity. Maximum compressive strength of scaffolds were determined using Merlin software associated with the Instron machine.

2.3.3 Quantification of biomineralization using Alizarin Red S (ARS). The biomineralization of scaffolds was quantified using a modification of a published protocol.²² Alizarin Red S (Sigma, MO) was dissolved in deionized (DI) water to a final concentration of 40 mM and pH was adjusted to 4.1 using 1 M NaOH. The solution was then filtered through a 0.8 μm mesh to remove any particulates and stored in the dark. After two week incubation, scaffolds were rinsed thrice for 10 min each in DI water, transferred into a new 24-well plate and fixed in 3.7% buffered formaldehyde for 30 min. The scaffolds were stained with an excess of Alizarin Red S (ARS) solution for 20 min on an orbital shaker, following which the scaffolds were rinsed repeatedly in DI water till all unbound dye was washed off (as evidenced by appearance of a clear wash solution). At this time, the scaffolds were transferred to a new plate and the bound dye solubilized with 10% acetic acid solution. After 30 min, the solubilized stain was pipetted out, 10% NaOH was added and the absorbance at 550 nm measured in a 96-well plate using a microplate reader.

2.4 Cell culture on HMS and microspheres scaffolds

Mouse mesenchymal stem cells (MSCs) were purchased from American Type Culture Collection (ATCC, Manassas, VA). MSCs were enzymatically lifted from culture dishes with trypsin/EDTA (0.125% and 0.53 mM, respectively), and centrifuged for 5 min at 250 rpm.

For HMS, MSCs were plated into 96-well plates at a density of 2.0 × 10³ cells per well in 200 μl culture medium and allowed to attach for 24 h. Afterwards, the growth media was removed and the cells were washed with PBS, then each well was added 200 μl a-MEM supplemented with 10% FBS medium with HMS at different concentrations of 50, 100, 200 and 400 μg ml⁻¹, and maintained in an incubator at 37 °C with 5% CO₂ and 95% humidified air for 1, 3 and 7 days.

Scaffolds (diameter = 10 mm, height = 5 mm) for in vitro cell study were sterilized by being immersed in 70% ethanol for 24 h, washed three times with PBS water for 30 min, and exposed to UV light for 30 min on each side. Cells were seeded onto the microsphere scaffolds before and after biomineralization at a density of 1 × 10⁵ cells per scaffold. Cells were cultured in a-MEM supplemented with 10% FBS.

2.5 Cell proliferation

Cell proliferation of HMS and HMS/PLGA (1 : 5) scaffolds before and after mineralized were quantitatively analyzed by utilizing CCK-8 assay. In brief, at designated time intervals, the medium was removed and the cells were washed with PBS (pH = 7.2) two times. CCK-8 solution was added to each well and the plate was incubated for 2 h in the incubator. The absorbance was measured at 450 nm using a microplate reader.

2.6 Statistical analysis

Experiments were repeated three times and results were expressed as means ± standard deviations (n = 6). Statistical significance was calculated using one-way analysis of variance (one-way ANOVA). Comparison between the two means was determined using the Tukey test and statistical significance was defined as p < 0.05.

3. Results and discussion

3.1 Characterization of HMS

The morphology and structure of HMS were investigated by SEM illustrated in Fig. 1A. HMS particles had an irregular sheet-shaped surface morphology and their sizes were about 200 nm. HRTEM images of the HMS given in Fig. 1B presenting uniform and homogeneously distributed mesopores. The N² ad–desorption isotherms and pore size distributions of the HMS are shown in Fig. 2. Brunauer–Emmett–Teller (BET) specific surface area of 1198.5 m² g⁻¹, an average pore size of 2.44 nm and a adsorption total volume at P/P₀ = 0.97 which was 0.668 cm³ g⁻¹, were obtained. The type IV isotherms, indicate the p6mm mesoporous structure of the HMS, which agreed well with the TEM.

Fig. 1 SEM (A) and HRTEM (B) images of HMS.

Fig. 2 The N₂ adsorption–desorption isotherms and the corresponding pore size distributions images of HMS.

Particle-induced cytotoxicity was measured with the CCK-8 assay, an approach widely used to measure the mitochondria activity to quantify the cell growth or cell death. According to the CCK-8 measurement, as shown in Fig. 3, compared with the untreated control (cells cultured on TCPS without materials), after high concentration of HMS (400 and 200 μg ml⁻¹) treatment, the cell viability of MSCs significantly decreased, in particular after 3 days and 7 days exposure. The inhibition efficiency was enhanced, when HMS concentration was increased. The 100 and 50 μg ml⁻¹ of HMS treated cell viabilities were higher than the untreated control in all cases, which suggested that they had good cytocompatibility. When the concentration of HMS were decreased or increased compared to 100 μg ml⁻¹, the cell viabilities were decreased. It means that the concentration of HMS below 200 μg ml⁻¹ have no cytotoxicity, and under this condition, suitable concentration of HMS can promote MSCs proliferation effectively than the lower concentration.

Fig. 3 CCK-8 assay for cell proliferation. (*) Indicates a significantly higher (p < 0.05) cell number on HMS than control ground.

3.2 Morphological and compressive characterization, cytocompatibility of HMS/PLGA scaffolds

The SEM images showed the HMS/PLGA microspheres scaffolds (HMS to PLGA ratio from 0 to 1 : 2) in Fig. 4A–G. All scaffolds had a three-dimensional network structure, which mimicked the structure of the natural extracellular matrix. Increasing in the HMS to PLGA ratio from 0 to 1 : 2 resulted in a more abundant distribution of HMS on the scaffolds or microspheres surface, which enhanced their surface roughness. Under the same heat sintering condition, microsphere scaffolds with higher HMS content had worse fusion between microspheres. As shown in Fig. 4E–G, there are abundant HMS dispersed on the high HMS content microspheres (HMS to PLGA ratio from 3 : 10 to 1 : 2) surface, and these kind of microspheres could not be fused together.

Fig. 4 SEM images of PLGA (A), HMS/PLGA (1 : 10) (B), HMS/PLGA (1 : 5) (C), HMS/PLGA (1 : 4) (D), sintered microspheres scaffolds and HMS/PLGA (3 : 10) (E), HMS/PLGA (2 : 5) (F) and HMS/PLGA (1 : 2) (G) microsphere.

Compressive strength of the PLGA and HMS/PLGA 1 : 10, 1 : 5 and 1 : 4 composite scaffolds are shown in Fig. 5. All HMS/PLGA scaffolds' compressive strength were significantly higher than the pure PLGA scaffolds. Indicating that HMS can effectively enhance the PLGA scaffolds' compressive strength.

Fig. 5 Compressive strength evaluation in compression for original and mineralized scaffolds. (*) Indicate statistical significance when compared with PLGA (p < 0.05).

MSCs proliferation on HMS/PLGA scaffolds was demonstrated via CCK-8 assay (Fig. 8). The viability of MSCs was thus assessed after 1, 3 and 7 days of culture. Compared with the PLGA, the cell viability of MSCs cultured on relative high HMS concentration HMS/PLGA (1 : 4) was significantly decreased at each time point. The cell viability of MSCs cultured on relative low HMS concentration HMS/PLGA (1 : 10) and (1 : 5) were higher than the PLGA in all cases, which suggested that they had good cytocompatibility. Compared with the (1 : 10) group, the (1 : 5) group have better cytocompatibility.

As a temporary extra cellular matrix, a BTE scaffolds should achieve the balance of mechanical properties between the load-bearing site and the scaffolds. However, PLGA is low mechanical strength and too flexible characteristic, which makes it not able to be utilized as the load-bearing applications alone. The composite HMS/PLGA scaffolds we developed in this study have compressive strength in the range of the cancellous bone (compressive strength is 2–180 MPa),²³ which ensures that the scaffolds will be mechanically suitable for load-bearing bone regeneration. And microsphere scaffolds with similar structure have been confirmed to be able to facilitate cell ingrowth, proliferation and differentiation.^24,25 However, when the HMS contents was improved from 9.1% to 20%, the compressive strength decreased significantly from 23.09 ± 2.11 to 7.5 ± 0.72 MPa. This is because when more HMS was added, more HMS concentrated on the surface of the composite microspheres, which led to worse fusion between microspheres by heat sintering, and these kinds of microsphere scaffolds were more easily destroyed under smaller stress.

Although PLGA scaffolds have been used extensively as graft substitutes in BTE, it is essential to reassess the biocompatibility of these scaffolds now that they have been composited with HMS. Therefore, preliminary in vitro studies using MSCs were carried out in order to investigate the cellular response in contact with the HMS/PLGA scaffolds. The proliferation of MSCs cultured on HMS/PLGA (1 : 4) was inhibited. This is mainly because that the cytocompatibility of HMS/PLGA scaffolds were attribute to the HMS. That is, HMS have cytotoxicity with high concentration, and suitable concentration of HMS can promote MSCs proliferation effectively than the lower concentration (Fig. 3). Abundant HMS appeared on the HMS/PLGA (1 : 4) scaffolds surfaces when HMS content is high, which results in the cytotoxicity. In addition, the HMS/PLGA (1 : 4) scaffolds with relative high surface roughness (Fig. 4D), which may retard cells proliferation.²⁶ HMS/PLGA scaffolds (1 : 10 and 1 : 5) with relatively low HMS content have suitable surface roughness, and a suitable amount of HMS on the scaffolds surfaces’, which could promote MSCs proliferation effectively.

3.3 Morphological, compositional and compressive characterization, cytocompatibility of mineralized scaffolds

After 2 days immersion in 5 × SBF, all scaffolds exhibited mineral deposition in a leaf-like shape were observed (Fig. 6A3, B3, C3 and D3). More apatite deposited on the scaffolds surface as HMS content increased (Fig. 6A1–2, B1–2, C1–2 and D1–2). EDX analysis showed Ca and P peaks after the 5 × SBF immersion, suggesting the Ca–P-containing particles deposited on the surface of scaffolds. The ratios of Ca–P of Ca–P-containing particles on PLGA, HMS/PLGA (1 : 10), HMS/PLGA (1 : 5) and HMS/PLGA (1 : 4) scaffolds were 1.18, 1.22, 1.23 and 1.33, respectively, indicating an calcium-deficient apatite.²⁷ Alizarin Red S (ARS) is an anthraquinone derivative that forms a water-insoluble salt with calcium, and has been routinely used as a qualitative assay for confirming the presence of calcium deposits. Our study adapted this technique to study mineral content in microspheres scaffolds. The intensity of staining seen in HMS/PLGA scaffolds (Fig. 7A–D) represents increasing amounts of HMS which results in greater mineral deposits. And the quantified results shown in Fig. 7E also indicate that staining intensity of ARS (measured by absorbance) increased as a function of amount of HMS added to polymer scaffolds, agree well with the results of SEM, EDX and ARS staining.

Fig. 6 SEM and inserted EDX images of microspheres scaffolds after incubating in 5 × SBF for 2 days, PLGA (A1–3); HMS/PLGA (1 : 10) (B1–3); HMS/PLGA (1 : 5) (C1–3); HMS/PLGA (1 : 4) (D1–3).

Fig. 7 Pictures of sintered microspheres scaffolds after incubating in 5 × SBF for 2 d and stained with Alizarin Red to visually follow the biomineralization, PLGA (A); HMS/PLGA (1 : 10) (B); HMS/PLGA (1 : 5) (C). HMS/PLGA (1 : 4) (D). A plot of absorbance (at 550 nm) as a function of differential HMS loading in PLGA scaffolds (E). (*) Indicate statistical significance when compared with PLGA scaffolds (p < 0.05).

After the biomineralization process for 2 days, the compressive strength of mineralized scaffolds were similar with the original scaffolds (Fig. 5). It means that this biomineralization process has less effect on the compressive strength of scaffolds.

After 1, 3 and 7 days of culture, MSCs cultured on the mineralized scaffolds exhibited better cell viability than original scaffolds (Fig. 8). As more apatite deposited on the scaffolds, better cell viability was observed. And the mineralized HMS/PLGA (1 : 4) scaffolds have the best cytocompatibility.

Fig. 8 CCK-8 assay for cell proliferation. (*) Indicates a significantly different (p < 0.05) cell number on original scaffolds than on original PLGA scaffolds. (#) Indicates a significantly higher (p < 0.05) cell number on mineralized scaffolds than on corresponding original scaffolds.

The in vitro biomineralization of HMS/PLGA scaffolds were achieved by immersing them in 5 × SBF. SBF was often used to form a biomimetic bone-like mineral on BTE materials' surface to mimic the process of biomineralization in native bone and to predict the in vivo bioactivity of the materials.²⁸ However, this biomimetic process is rather slow and normally up to a few weeks from conventional SBF. Shortening immersion period should be particularly significant to degradable polymer such as PLGA because it could be degraded significantly during the long incubation period. In addition, HMS has been reported that they had poor apatite mineralization ability in SBF.²⁵ And the traditional PLGA scaffolds modified via the biomineralization method where PLGA scaffolds are always surface-hydrolyzed with alkali liquor.²⁹ However, the surface-hydrolyzed may accelerate the degradation of PLGA. To make sure the apatite could deposited on HMS/PLGA scaffolds effectively in a short time, 5 × SBF were applied in this study and without surface-hydrolyzed treatment, to try not to damage the pristine scaffolds. Interestingly, we found that more apatite were deposited after 2 days on the scaffolds surface as the HMS content increased. This is because that the induction time was shorter than the dipping time, the apatite crystals are formed homogeneously in the 5 × SBF with the escape of gassy CO₂ due to the decomposition of HCO₃⁻, continue to grow in the solution and finally attached onto scaffolds surfaces.³⁰ And the increased surface roughness of HMS/PLGA scaffolds had a better ability for the apatite attachment and bonding.

The bone-mineral produced on the scaffolds surface is considered to provide favorable substrate conditions for the cell adhesion and growth, a required initial stage for BTE. In this study, the cytocompatibility of HMS/PLGA scaffolds were successfully promoted by depositing the apatite layers. The better cytocompatibility of mineralized HMS/PLGA scaffolds could be attributed to their chemical composition that offers the advantage of ionic products of apatite dissolution known for their beneficial role on the stimulation of cells proliferation (i.e. Ca and P ions). All the ionic products (Ca and P) of apatite were important elements in human bone tissues and their concentrations from these mineralized scaffolds, at which level they were non-cytotoxic to surrounding tissues. When more apatite deposited on the scaffolds surface, more Ca and P ions were released and could better stimulate the proliferation of MSCs. In particular, the densest apatite layers deposited on the HMS/PLGA scaffolds (1 : 4) could effectively prevent MSCs contact with scaffolds surface with abundant HMS, which made HMS/PLGA scaffolds (1 : 4) have the best cytocompatibility, conversely. However, more works such as animal experiments should be done to further test the biocompatibility of mineralized HMS/PLGA scaffolds in the future.

4. Conclusion

In this study, we developed apatite/HMS/PLGA microsphere scaffolds for BTE. The HMS significantly improved the compressive strength of PLGA scaffold and promoted apatite deposited on the scaffolds in 5 × SBF for 2 days. Also, after biomineralization process, the cytocompatibility of HMS/PLGA scaffolds was significantly improved. To sum up, mineralized HMS/PLGA scaffolds with great compressive strength and cytocompatibility, could be attractive in the future to be utilized as BTE scaffolds.

Acknowledgements

This study was supported by National Basic Research Program of China (A00102110400) and China Postdoctoral Science Foundation funded project (2013M540787).

References

1. J. M. Anderson and M. S. Shive, Adv. Drug Delivery Rev., 1997, 28, 5–24.
2. J. F. Nelson, H. G. Stanford and D. E. Cutright, Oral Surg., Oral Med., Oral Pathol., 1977, 43, 836–843.
3. J. O. Hollinger, J. Biomed. Mater. Res., 1983, 17, 71–82.
4. D. Chenglong, G. Han, L. Jingxiong, S. Jianlin, W. Jie and L. Changsheng, Biomaterials, 2011, 32, 8506–8517.
5. M. P. Ginebra, T. Traykova and J. A. Planell, J. Controlled Release, 2006, 113, 102–110.
6. M. Stigter, J. Bezemer, K. de Groot and P. Layrolle, J. Controlled Release, 2004, 99, 127–137.
7. W. Xia and J. Chang, J. Controlled Release, 2006, 110, 522–530.
8. K. Schwarz and D. B. Milne, Nature, 1972, 239, 333–334.
9. E. M. Carlisle, Science, 1970, 167, 279–280.
10. E. M. Carlisle, Science, 1972, 178, 619–621.
11. E. M. Carlisle, J. Nutr., 1980, 110, 1046–1056.
12. K.-H. Shin, J.-H. Sung, Y.-H. Koh, J.-H. Lee, W.-Y. Choi and H.-E. Kim, Mater. Lett., 2010, 64, 1539–1542.
13. H. Zhou, X. Wu, J. Wei, X. Lu, S. Zhang, J. Shi and C. Liu, J. Mater. Sci.: Mater. Med., 2011, 22, 731–739.
14. P. T. Tanev and T. J. Pinnavaia, Science, 1995, 267, 865–867.
15. L. L. Hench, J. Am. Ceram. Soc., 1991, 74, 1487–1510.
16. N. C. Geurs, R. L. Jeffcoat, E. A. McGlumphy, M. S. Reddy and M. K. Jeffcoat, The International journal of oral & maxillofacial implants, 2002, 17, 811.
17. K. M. Woo, J. Seo, R. Zhang and P. X. Ma, Biomaterials, 2007, 28, 2622–2630.
18. H. Yuan, Y. Li, J. De Bruijn, K. De Groot and X. Zhang, Biomaterials, 2000, 21, 1283–1290.
19. B. Marelli, C. E. Ghezzi, D. Mohn, W. J. Stark, J. E. Barralet, A. R. Boccaccini and S. N. Nazhat, Biomaterials, 2011, 32, 8915–8926.
20. P. T. Tanev, M. Chibwe and T. J. Pinnavaia, Nature, 1994, 368, 321–323.
21. F. Barrere, C. A. van Blitterswijk, K. de Groot and P. Layrolle, Biomaterials, 2002, 23, 2211–2220.
22. C. A. Gregory, W. G. Gunn, A. Peister and D. J. Prockop, Anal. Biochem., 2004, 329, 77–84.
23. S. A. Goldstein, J. Biomech., 1987, 20, 1055–1061.
24. C. T. Laurencin, J. Tao and W. I. Abdel-Fattah, Biomaterials, 2006, 27, 4894–4903.
25. X. Shi, Y. Wang, L. Ren, N. Zhao, Y. Gong and D.-A. Wang, Acta Biomater., 2009, 5, 1697–1707.
26. T. Jiang, W. I. Abdel-Fattah and C. T. Laurencin, Biomaterials, 2006, 27, 4894–4903.
27. J. G. Li, H. H. Liao and M. Sjostrom, Biomaterials, 1997, 18, 743–747.
28. T. Kokubo and H. Takadama, Biomaterials, 2006, 27, 2907–2915.
29. E. Jabbarzadeh, L. S. Nair, Y. M. Khan, M. Deng and C. T. Laurencin, J. Biomater. Sci., Polym. Ed., 2007, 18, 1141–1152.
30. Q. Cai, Q. Xu, Q. Feng, X. Cao, X. Yang and X. Deng, Appl. Surf. Sci., 2011, 257, 10109–10118.

---

Footnotes

† These authors contributed equally to this manuscript.

‡ Present address: Room 333, Building B12, South China University of Technology, Panyu District, Guangzhou City, Guangdong Province, China.

---