Journal of Materials Science & Technology  2019 , 35 (9): 1959-1965 https://doi.org/10.1016/j.jmst.2019.04.027

Orginal Article

Bioactive glass nanotube scaffold with well-ordered mesoporous structure for improved bioactivity and controlled drug delivery

Jian Xiaoa, Yizao Wanab, Zhiwei Yangb, Yuan Huanga, Fanglian Yaoc, Honglin Luoab*

a School of Materials Science and Engineering, Tianjin University, Tianjin, 300072, China
b Institute of Advanced Materials, School of Materials Science and Engineering, East China Jiaotong University, Nanchang, 330013, China
c Key Laboratory of Systems Bioengineering of Ministry of Education, School of Chemical Engineering and Technology, Tianjin University, Tianjin, 300072, China

Corresponding authors:   *Corresponding author at: School of Materials Science and Engineering, TianjinUniversity, Tianjin, 300072, China.E-mail address: hlluotju@126.com (H. Luo).

Received: 2019-01-28

Revised:  2019-04-14

Accepted:  2019-04-24

Online:  2019-09-20

Copyright:  2019 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

In this study, a novel mesoporous bioactive glass nanotube (MBGN) scaffold has been fabricated via template-assisted sol-gel method using bacterial cellulose (BC) as template and nonionic block copolymer (P123) as pore-directing agent. The scaffold was characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), Fourier-transform infrared (FTIR) spectroscopy, and N2 adsorption-desorption analysis. Furthermore, simvastatin was used to evaluate the loading efficiency and release kinetics of the scaffold. The obtained scaffold displays nanofiber-like morphology, ordered mesopores on the tube walls, and interconnected three-dimensional (3D) network structure that completely replicates the BC template. In addition, it shows dual pore sizes (16.2 and 3.3 nm), large specific surface area (537.2 m2 g-1) and pore volume (1.429 cm3 g-1). More importantly, the scaffold possesses excellent apatite-forming ability and sustainable drug release as compared to the counterpart scaffold without mesopores. This unique scaffold can be considered a promising candidate for drug delivery and bone tissue regeneration.

Keywords: Bioactive glass ; Mesopore ; Nanotube ; Bioactivity ; Drug delivery

0

PDF (3411KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Jian Xiao, Yizao Wan, Zhiwei Yang, Yuan Huang, Fanglian Yao, Honglin Luo. Bioactive glass nanotube scaffold with well-ordered mesoporous structure for improved bioactivity and controlled drug delivery[J]. Journal of Materials Science & Technology, 2019, 35(9): 1959-1965 https://doi.org/10.1016/j.jmst.2019.04.027

1. Introduction

Mesoporous bioactive glasses (MBGs) have received tremendous interest in drug delivery and bone tissue engineering [[1], [2], [3], [4], [5], [6]]. Compared to traditional bioactive glasses (BGs), MBGs possess large specific surface area and pore volume due to their well-ordered mesoporous structure, which endow them with enhanced in vitro bioactivity [1,7]. Furthermore, the mesopores can provide reservoirs for effective drug loading. Wu et al. used MBGs as carrier to load dexamethasone (DEX) and doxorubicin (DOX) and sustained release was realized [8,9]. Several other groups loaded MBGs with ampicillin to achieve anti-bacterial property [10].

These studies indicate that MBGs have potential in drug delivery. However, previous reports mainly concentrated on MBG powders or granules, rather than fibrous MBGs, which mimic natural extracellular matrix (ECM). Unlike powders and granules, fibrous MBGs can act as scaffolds to support cell migration, nutrient delivery and bone ingrowth due to their interconnected porous and three-dimensional (3D) network structure [[11], [12], [13], [14], [15]]. In general, MBG fibers are constructed using electrospinning technique and subsequent calcination [[16], [17], [18], [19]]. Although significant advancements have been made, the reported MBG fibers (solid or hollow) possess diameters at the submicron scale [20]. There is no report on the MBG hollow fibers with a diameter down to the nano-scale, which is desirable since collagen fiber diameter within natural ECM ranges from 50 to 500 nm [21].

In the present work, we report the preparation of a novel 3D mesoporous bioactive glass nanotube (MBGN) scaffold with an ultrafine tube diameter of around 40 nm and a large specific surface area of 537.2 m2 g-1. Such unique MBGN with well-ordered mesoporous structure on the walls was fabricated via the template-assisted sol-gel method using bacterial cellulose (BC) as template and nonionic block copolymer (P123) as pore-directing agent. Simvastatin (SIM) was loaded into the MBGN scaffold and release profiles were investigated. In addition, the in vitro bioactivity of the MBGN scaffold was evaluated.

2. Experimental

Tetraethyl orthosilicate (TEOS, 98%) and simvastatin (SIM, 98%) were of analytical grade and purchased from Beijing Innochem Science & Technology Co., Ltd (Beijing, China). Absolute ethyl alcohol, tertiary butanol, calcium nitrate tetrahydrate (Ca(NO3)2·4H2O, 99%) and triethylphospharm (TEP, 99.8%) were of analytical grade and purchased from Tianjin Kemiou Chemical Reagent Co., Ltd (Tianjin, China). Nonionic block copolymer EO20PO70EO20 (P123, average molar mass M = 5800 g mol-1) was purchased from Sahn chemical technology Co., Ltd (Shanghai, China). Bacterial cellulose (BC) was prepared in our lab using the previously reported procedures [[22], [23], [24], [25], [26]].

MBGN scaffold was prepared via the template-assisted sol-gel method followed by calcination using BC as the template and P123 as the pore-directing agent. In a typical process, 4 g of P123 was completely dissolved in 50 mL of absolute ethyl alcohol at room temperature followed by adding 9 mL of TEOS, 1.75 g of Ca(NO3)2·4H2O, and 0.85 mL of TEP under vigorous stirring for 3 h. Subsequently, 125 mg of dried BC was added under continuous stirring for 24 h. The obtained products of SiO2/CaO/P2O5@BC hydrogel were washed with absolute ethanol and soaked in the mixture of C2H5OH/H2O (9:1; vol.%) for 24 h to allow complete hydrolysis and poly-condensation. Finally, the obtained specimens were freeze-dried for 24 h and then heated from ambient temperature to 600 °C at a heating rate of 1 °C min-1 and held in air for 5 h to obtain the final MBGN scaffold. As a control group, BGN scaffold was synthesized by the same process without adding P123.

The porosity of BGN and MBGN scaffolds was assessed based on Archimedes’ principle: BGN and MBGN scaffolds with a size of 5 mm× 5 mm× 3 mm were used in the measurements and water was the liquid medium. The porosity (P*) was calculated using the following equation:

P*=(Wsat-Wdry)/(Wsat-Wsus) × 100% (1)

where Wdry is the dry weight of scaffolds, Wsus and Wsat are the scaffold weights suspended in water and saturated with water, respectively.

The in vitro bioactivity of BGN and MBGN scaffolds was assessed by immersing in simulated body fluid (SBF). Typically, 30 mg of scaffold was soaked in 30 mL of SBF in a polyethylene bottle at 36.5 °C for different periods ranging from 4 to 24 h. Afterwards, the specimens were collected from SBF, rinsed with absolute ethanol followed by distilled water and then dried.

BGN and MBGN scaffolds were dispersed into ethanol solution of SIM (1 mg mL-1). The suspension was shaken in a sealed vessel at a constant rate (110 rpm) for 24 h followed by freeze-drying to obtain SIM loaded scaffolds. To assess the release kinetics of SIM, 10 mg of SIM-loaded scaffolds was placed in a cover-sealed bottle with 20 mL of phosphate buffered saline (PBS), and then the bottle was fixed on the shaking bed with a shaking speed of 90 rpm at 37 °C. The release medium was withdrawn at the predetermined time intervals (1, 3, 6, 12, 24 h and 3, 7, 11, 14 d) and replaced with the same volume of fresh PBS each time. The SIM released from the SIM-loaded scaffolds was evaluated by ultraviolet-visible (UV) analysis at a wavelength of 238 nm. Before determination, a calibration curve was recorded by NANODROP 2000 Spectrophotometer.

The surface morphology of BGN and MBGN scaffolds were assessed by scanning electron microscopy (SEM, Hitachi S4800) from which the fiber diameter was calculated by measuring at least 100 randomly selected fiber segments as previously reported [[27], [28], [29]], and the standard deviation is used for estimating the error bars. Elemental analysis was carried out with energy dispersive spectroscopy (EDS) attached to SEM. Transmission electron microscopy (TEM) was recorded on JEOL JEM-2100 F microscope at 200 kV. X-ray diffraction (XRD) patterns were obtained on Bruker D8 advanced XRD diffractometer with Cu Ka irradiation. Brunauer-Emmett-Teller (BET) surface area of BGN and MBGNs scaffolds was evaluated from nitrogen adsorption isotherms at 77 K using a surface area analyzer (NOVA 2200e). Attenuated total reflection Fourier-transform infrared (ART-FTIR; Varian 640-IR) spectroscopic analysis was performed to analyze the chemical structure over a range of 400-4000 cm-1 at a resolution of 4 cm-1. Thermogravimetric (TG) analysis was obtained on Netzsch STA-409C thermogravity analyzer with a heating rate of 10 °C min-1 from ambient temperature to 800 °C in N2 atmosphere.

3. Results and discussion

3.1. Morphologies and physicochemical properties

SEM was employed to evaluate the morphologies of BGN and MBGN scaffolds and digital photographs were used to prove the 3D nature. Both BGN (Fig. 1(a)) and MBGN (Fig. 1(c)) scaffolds exhibit 3D interconnected porous structure, which is important in bone tissue engineering [[29], [30], [31]]. Based on SEM images, the diameters of BGN and MBGN nanofibers were measured. The average fiber diameters of BGN (Fig. 1(b)) and MBGN (Fig. 1(d)) scaffolds are approximately 27 ± 6 nm and 37 ± 8 nm, respectively, suggesting an effect of P123 on the fiber size of MBGN scaffold.

Fig. 1.   SEM images of BGN (a) and MBGN (c), and fiber diameter distributions of BGN (b) and MBGN (d). Insets of (a) and (c) show the digital photographs of the corresponding samples, respectively.

TEM was used to determine the nanotube structure of BGN and MBGN scaffolds. As shown in Fig. 2(a) and (c), both scaffolds have nanotube morphology with a wall thickness of around 5-8 nm. Furthermore, the walls of BGN scaffold are solid (Fig. 2(b)) while MBGNs possess well-ordered mesopores throughout their walls (Fig. 2(d)). According to EDS analysis (Fig. 2(e) and (f)), the chemical compositions of BGN and MBGN are the same, being composed of Ca, Si, P, and O without other elements, suggesting high purity. The insets in Fig. 2(a) and (c) reveal the amorphous nature of BGN and MBGN.

Fig. 2.   TEM images of BGN (a) and MBGN (c), and HRTEM images of BGN (b) and MBGN (d), EDS spectra of BGN (e) and MBGN (f) scaffolds. Insets of (a) and (c) show SAED patterns of the corresponding samples, respectively.

The amorphous nature of BGN and MBGN is further verified by wide-angle XRD patterns (Fig. 3(a)) because there are no diffraction peaks except a wide SiO2 peak at 2θ ≈ 25°. These results are in accordance with selected area electron diffraction (SAED) patterns (Fig. 2(a) and (c)). Small-angle XRD patterns display an apparent diffraction peak at around 1.2° for MBGN (Fig. 3(b)), and two weak peaks between 1.5° and 1.7° (inset in Fig. 3(b)). These results confirm uniformly distributed 2D-hexagonal pore structure in MBGN, which is similar to that of MBG powders reported by Yan et al. [32]. However, no diffraction peaks are visible in BGN (Fig. 3(b)), suggesting the absence of mesopores in BGN, consistent with the high resolution TEM (HRTEM) result shown in Fig. 2(d).

Fig. 3.   Wide-angle (a) and small-angle (b) XRD patterns of BGN and MBGN scaffolds. Inset of (b) shows the enlarged region of MBGN scaffold at around 1.4°-1.9°.

The pore characteristics of BGN and MBGN scaffolds were examined by BET method. N2 adsorption-desorption isotherms and corresponding pore size distributions of BGN and MBGN scaffolds are shown in Fig. 4. The isotherms of BGN and MBGN scaffolds possess type IV isotherm with H3-type hysteresis loop (Fig. 4(a)), which is a typical characteristic of mesoporous structure. The pore size distributions, obtained from the adsorption branch using BJH model, reveal dual pores (16.2 and 3.3 nm) of MBGN and sole pore (16.0 nm) of BGN (Fig. 4(b) and Table 1). The pore size of 3.3 nm agrees well with HRTEM result (Fig. 2(d)). The pores with a size of 16.2 nm in MBGN and pores of 16.0 nm in BGN are formed by neighboring nanotubes through stacking. The formation of these pores is similar to the pores formed in electrospun scaffolds. The pore volumes of BGN and MBGN scaffolds are 0.385 cm3 g-1 and 1.429 cm3 g-1, respectively, and the specific surface areas of BGN and MBGN scaffolds calculated from the linear part of the BET plots are 196.9 m2 g-1 and 537.2 m2 g-1, respectively. These larger pore volume and specific surface area of MBGN than that of BGN scaffold are due to the presence of mesopores in MBGN. These mesopores on the walls of MBGN scaffold are of crucial role in loading and releasing biomolecules since they are connected to the cavities within the nanotubes of MBGN scaffold, thus providing channels for drug loading and releasing.

Fig. 4.   (a) N2 adsorption-desorption isotherms of BGN and MBGN scaffolds and (b) pore size distribution curves of BGN and MBGN scaffolds. Inset of (b) shows the enlarged portion from 0 to 18 nm.

Table 1   Structural parameters of BGN and MBGN scaffolds.

SampleSpecific surface area (m2 g-1)Pore volume (cm3 g-1)Pore size
(nm)
BGN196.90.38516.0
MBGN537.21.4293.3/16.2

New window

3.2. In vitro bioactivity

MBGN scaffold is supposed to possess excellent bioactivity due to its large specific surface area. Hence, the in vitro bioactivity was assessed by immersing samples into SBF for various lengths of time. The changes in morphology, crystalline phase, and chemical structure were investigated. SEM images (Fig. 5) demonstrate significant changes in morphology after soaking in SBF as compared to the SEM images before soaking (Fig. 1(a) and (c)). Note that there is significant difference in the growth rate of minerals on BGN and MBGN scaffolds at each soaking time, the latter (Fig. 5(d)-(f)) showing faster growth than the former, based upon the degree of mineral coverage (Fig. 5(a) vs. (c), (b) vs. (e), (c) vs. (f)).

Fig. 5.   SEM images of BGN (a-c) and MBGN (d-f) scaffolds after soaking in SBF for 8 h (a, d), 16 h (b, e) and 24 h (c, f).

To determine the crystalline structure of minerals on BGN and MBGN scaffolds, XRD analysis was performed (Fig. 6). At 4 h, a weak peak at 2θ ≈ 32° is noted on MBGN (Fig. 6(a)) and BGN (Fig. 6(b)) scaffolds. The peak number and intensity increase with immersion time and several peaks at 26°, 32°, 39°, 46°, 49° and 53°, which can be assigned to the (002), (211), (310), (222), (213) and (004) reflections of hydroxyapatite (HAp) phase [33], are noted on two scaffolds. Notably, BGN exhibits weaker peaks than MBGN at each soaking time, consistent with SEM results, further indicating the higher in vitro bioactivity of MBGN scaffold.

Fig. 6.   XRD patterns (a, b) and FTIR spectra (c, d) of MBGN (a, c) and BGN (b, d) scaffolds before and after soaking in SBF for different times.

The structure of minerals deposited on two scaffolds was further characterized by FTIR analysis. Before soaking in SBF, BGN and MBGN scaffolds exhibit similar spectra in which the bending and stretching vibrations of Si-O-Si bonds appear at 1040, 800, and 470 cm-1 [34], respectively (Fig. 6(c) and (d)). After soaking in SBF for 4 h, MBGN scaffold shows an obvious P—O vibrational band at 566 cm-1 (Fig. 6(c)), suggesting the formation of amorphous product. At 8 h, the vibrational peak near 566 cm-1 splits into double peaks at 566 and 606 cm-1, respectively, corresponding to crystalline structure [1]. Further increasing the soaking time leads to enhanced intensity of vibrational peak at 566 and 606 cm-1, and the appearance of other vibrational peaks of HAp. At 24 h, well-resolved vibrational peaks at 1487, 1414 and 873 cm-1, due to C-O vibration bands and the vibrational peaks at 1040, 963, 606 and 566 cm-1, due to the crystalline P—O bands, are detected, which confirms the presence of carbonated hydroxyapatite (HCA) [35,36]. Compared to MBGN scaffold, BGN scaffold shows identical but weaker bands at each soaking time (Fig. 6(d)). Together with the SEM and XRD results, the FTIR analysis evidences the better in vitro bone forming bioactivity of MBGN scaffold than that of BGN scaffold due to its larger surface area and the presence of mesopores. The results were consistent with those reported by Zhao et al. [2]. Besides, Vallet-Regí et al. also reported that MBGs possessed enhanced bone-forming bioactivity than BGs [37], since a larger specific surface area increases the contact area between scaffold and SBF, thus resulting in higher bioactivity.

3.3. Loading and release of simvastatin

Besides improvement of bioactivity, mesopores can act as reservoirs to store drugs and a controlled drug release profile is expected.

In the present work, SIM, a derivative of statin, was loaded into BGN and MBGN scaffolds. To quantify the amount of SIM loaded into MBGN and BGN scaffolds, TG analysis was performed (Fig. 7(a)). The weight loss of SIM-loaded MBGN (26.2%) is much larger than that of SIM-loaded BGN (12.7%). The calculated loading capacities of SIM in BGN and MBGN are 7.1% and 21.8%, respectively (Fig. 7(b)). The higher SIM loading capacity of MBGN over BGN is ascribed to its mesopores, larger specific surface area and pore volume since these structural parameters have a significant effect on the loading efficiency and releasing kinetics of drugs [2,37,38].

Fig. 7.   TG curves (a) and loading efficiency (b) of SIM in BGN and MBGN scaffolds.

Fig. 8(a) depicts the release profiles of SIM from BGN and MBGN scaffolds in PBS. BGN and MBGN scaffolds exhibit similar release behavior: an initial rapid release during the first 24 h followed by a much slower release. The initial release approaches 32.7% of the total amount of SIM for MBGN while this value is 45.7% for BGN, which suggests a quicker burst release of BGN than MBGN. After that, MBGN also shows a significantly slower release as compared to BGN. The better sustainable release of MBGN than that of BGN can be elucidated by the illustration in Fig. 8(b). For both BGN and MBGN, SIM molecules are adsorbed on the outer surfaces of BG nanotubes via hydrogen-bond interaction between hydroxyl groups in SIM and Si-OH and POH groups in BGN [37]. In addition, MBGN loads SIM molecules in the inner cavities of nanotubes and the window of mesoporous channels [37,38], which results in a larger drug loading capacity than BGN. The burst release of BGN and MBGN during the initial period is due to the release of SIM molecules from the surfaces of nanotubes. The more sustainable release of SIM from MBGN can be attributed to the long-distance transportation of SIM molecules across inner cavities and mesopores in MBGN. Another possible reason is the larger specific surface area of MBGN which further retards the SIM release from the surface due to the hydrogen-bond interaction.

Fig. 8.   (a) Release profiles of SIM from BGN and MBGN scaffolds and (b) proposed mechanisms of loading and release of SIM molecules from BGN and MBGN scaffolds.

4. Conclusion

A novel 3D MBGN scaffold with ultrafine tubes and well-ordered mesopores on the tube walls has been successfully prepared by template-assisted sol-gel method. The obtained MBGN scaffold possesses an ultrafine tubular (around 40 nm) structure, a higher specific surface area, a larger pore volume, and enhanced in vitro bioactivity as compared to the BGN scaffold. Additionally, the MBGN scaffold shows a higher SIM loading efficiency and more sustainable drug release than the BGN scaffold without mesopores on the nanotube walls. It is believed that this novel MBGN scaffold possesses potential in drug delivery and bone tissue regeneration applications.

Acknowledgments

The work was supported financially by the National Natural Science Foundation of China (Nos. 51572187 and 30660264) and the Youth Science Foundation of Jiangxi Province (No. 20181BAB216010).

The authors have declared that no competing interests exist.


/