Materials Science and Engineering C 36 (2014) 237–243
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
pH-responsive controlled-release system based on mesoporous bioglass materials capped with mineralized hydroxyapatite Chunyu Yang, Wei Guo, Liru Cui, Di Xiang, Kun Cai, Huiming Lin ⁎, Fengyu Qu ⁎ Key Laboratory of Design and Synthesis of Functional Materials and Green Catalysis, College of Chemistry and Chemical Engineering, Harbin Normal University, Harbin 150025, PR China
a r t i c l e
i n f o
Article history: Received 26 May 2013 Received in revised form 28 November 2013 Accepted 6 December 2013 Available online 16 December 2013 Keywords: Bioglass mesoporous materials Mineralized hydroxyapatite pH-responsive Controlled-release
a b s t r a c t A controlled release system with pH-responsive ability has been presented. Mesoporous bioglass (MBG) was used as the drug carrier and a spontaneous mineralization method was adopted to cap the pores of the carrier with hydroxyapatite (HAp) and to restrict the drug release. It is a simple and green method to realize the ingenious pH-sensitive controlled release. The model drug, metformin hydrochloride (MH), was loaded simultaneously with the mineralization process. Due to the degradation of HAp at acid environments, the system shows well pH-sensitive drug release ability. The release kinetics can be easily adjusted by the mineralization time and the ion concentration of media. The system is recommended as a promising candidate as a pH-sensitive vehicle for drug controlled release to low pH tissues, such as inflammatory sites and tumors. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The drug delivery has attracted considerable interests in the past few decades. The main advantage of drug delivery is to allow optimizing the security and efficacy of the treatments. At present, cancer has become one of the most serious diseases to threaten people's health. Among the various treatments, the medication therapy was considered as the most effective method. However, medication effect has been greatly limited due to the toxic effects of the nonspecific distribution of anticancer drugs. That is because most anticancer drugs cannot distinguish between cancerous and healthy cells, inducing collateral damages and adverse side effects. To address this formidable challenge, using the proper methods to realize the controlled release to increase the drug effect become the most important in the biopharmaceutical field. At present, many materials are used as the hosts to load drugs, showing the different drug release kinetics [1–4]. With uniform and tunable morphology and pore size, large surface area, high pore volume, modifiable surface property, and good biocompatibility and biodegradation, mesoporous silica nanomaterials (MSNs) have become one of the most important hosts for drug delivery. With the various functionalities, many organic materials are always used to realize the controllable release process. For example, PAH/PSS and ALG/CHI based on mesoporous silica nanotubes have been developed as pH-controlled drug delivery systems via the layer by layer self-assembly technique [5]. Yang et al. synthesized a pH responsive nanocarriers with chitosan-poly (methacrylic acid) shells and MSN cores on one step. The system showed well physicochemical and pH sensitive controlled release properties [6]. Vallet⁎ Corresponding authors. Tel.: +86 451 88060653. E-mail addresses: [email protected] (H. Lin), [email protected] (F. Qu). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.12.006
Regí et al. reported on the synthesis of (PEI/NIPAM)/magnetic mesoporous silica nanocarrier, which was able to release different cargos in a controlled manner in response to an external alternating magnetic field [7]. Recently, due to the cost and complexity of the use of organic components, some inorganic nanoparticles [8–10] have been tried as the capping agents outside MSNs. There are two release mechanisms for these drug loading systems: 1) the breaking of the link between the cap and the host. Lin et al. synthesized mesoporous silica capped with Fe3O4. The controlled-release is based on breaking the disulfide linkage between the Fe3O4 nanoparticle caps and the linker-MSN hosts by reducing agents such as DHLA [11]. 2) The cap collapse makes the drug release. Zhu et al. used ZnO QDs to seal the pores of mesoporous silica. ZnO QDs lid on MSNs can be efficiently dissolved in the acidic intracellular compartments of cancer cells, resulting in the release of the drug cargo from the pores of the MSNs into the cytosol [12]. Although these works used the inorganic nanoparticles as the cap which showed well responsive controlled release, they could not avoid the use of the organic components as the linkage between cap and host. And the synthetic methods are often multistep procedures or harsh synthetic conditions. So that, a simple, green, and effective approach to produce a pHresponsive drug delivery system is still a great challenge. With well bioactivity, mesoporous bioglass (MBG) has been widely used in bone tissue reparation and regeneration [13,14]. MBG can induce the growth of hydroxyapatite (HAp) in vivo and in vitro (SBF, 37 °C) as well [15]. It is known that, HAp is the major inorganic component of natural bones with nontoxic, nonirritant and well bioabsorbabitity ability [16]. In addition, it can be dissolved producing nontoxic ions (calcium and phosphate ions) in acidic cellular environments such as endosomes (pH = 5.0) and lysosomes (pH = 4.5) [17].
238
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
Based on the above considerations, we developed a simple and green method to produce a drug loaded MBG system with pHsensitive controlled release ability. As shown in Scheme 1, MBG was used as the drug carrier. Metformin hydrochloride (MH, with the similar hydrophilic property of the doxorubicin hydrochloride) was use as the model drug. After drug loading process, the pores of the carrier were filled with drug molecules. Then, the SBF was added and the HAp nanoparticle layer formed outside owing to the spontaneous induced growth of HAp on the MBG. HAp nanoparticles can block the pore channels as a cap to prevent the drug release. In acidic cellular environments, the degradation of HAp cap makes the drug molecules release. The pH-sensitive drug release performance of the system was investigated in detail. And the mineralization time and the concentration of the SBF can be used to control the release process. The system can be used as a pH-sensitive vehicle for in vivo delivery of therapeutic agents to low pH tissues, such as inflammatory sites and tumors. 2. Experimental 2.1. Materials Tetraethyl orthosilicate (TEOS, PA), Ca(NO3)2·4H2O, triethyl phosphate (TEP, 99.8%, PA), metformin hydrochloride (MH, pharmacopeia) and ethanol (PA) were purchased from Tianjin Chemical Corp. of China. Poly (ethylene)-block-poly (propylene glycol)-block-poly (ethylene) (P123, PA) was purchased from Aldrich. All the chemicals were used as received. Deionized water was used for all experiments. 2.2. Preparation of MBG scaffolds Porous mesopore-bioglass (MBG) scaffolds were prepared by the sol– gel processing of the previous publication [18]. Specifically, 4.0 g of P123 (Mw = 5800, Aldrich), 6.7 g of tetraethyl orthosilicate (TEOS), 1.4 g of Ca(NO3)2·4H2O, 0.73 g of triethyl phosphate (TEP, 99.8%), and 1.0 g of 0.5 M HCl were dissolved in 60 g of ethanol (Si/Ca/P = 80:15:5 molar ratio) and stirred at room temperature for 24 h. Afterwards, the solution was transferred to a Petri dish and then allowed to evaporate at room temperature for 3–5 days. When the samples were completely dry, they were calcined at 700 °C for 6 h to obtain the MBG scaffolds.
Table 1 Ion concentration and pH of SBF and 2SBF.
SBF 2SBF
Na+ (mM)
K+ (mM)
Mg2+ (mM)
Ca2+ (mM)
Cl−
HCO3− (mM)
HPO2− 4 (mM)
SO2− 4 (mM) pH
pH
142 142
5 5
1.5 1.5
2.5 5
103.0 103
27 27
1.0 2.0
0.5 0.5
7.40 6.80
2.3. Drug loading The loading of the drug was conducted as the previous report [19]. A typical loading procedure of MH in MBG was as follows: 0.206 g of MBG with 2.06 g of MH was suspended in 50 mL distilled water and stirring for 2 h at room temperature. The vial was sealed to prevent the evaporation of solvent. After the drug loading process, 50 ml xSBF was added and standing for a period at 37 °C. The system was separated from the solution by vacuum filtration and dried in vacuum. In the process of vacuum filtration, sintered glass funnel (G5, pore size 1.5–2.5 μm) was used to separate the drug loaded samples from the solutions. The obtained solution was appropriately diluted, and the residual MH amounts were determined by UV/Vis measurements at 233 nm. The loading efficiency (LE %) of MH can be calculated by using formula (1). The experiment repeated three times. LE% ¼
mðoriginal
MHÞ −mðresidual MHÞ
mðMBGÞ þ mðoriginal
MHÞ −mðresidual MHÞ
100%
ð1Þ
The solid sample is named as MBG-MH-xSBF (yh). The x was used to describe the media solution concentration × times as SBF (x = 1 or 2, Table 1). y means the mineralization time. 2.4. Drug release 150 mg of MBG-MH-xSBF (yh) was soaked in 500 mL of media solution (pH = 4.0 HCl aqueous, pH = 5.0 phosphate buffer solution, pH = 6.8 phosphate buffer solution, pH = 7.4 phosphate buffer solution) at 37 °C. At predetermined time intervals, 3 mL samples have been withdrawn and immediately replaced with an equal volume of new medium to keep the volume constant. The withdrawn samples were filtered (0.45 um), properly diluted and monitored by UV–vis
Scheme 1. Schematic illumination of the drug assembly and release process.
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
239
total volume of the dissolution medium. This experiment has been repeated for 3 times. 2.5. Physicochemical characterization
Fig. 1. Small-angle XRD patterns for MBG.
spectrophotometer at 233 nm. Calculation of the corrected concentration of the released MH is based on the following formula: Ctorr ¼ Ct þ
t−1 vX C V 0 t
Powder X-ray patterns (XRD) were recorded on a SIEMENS D5005 X-ray diffractometer with Cu Kα radiation (40 kV, 30 mA). The nitrogen adsorption/desorption, surface areas, and median pore diameters were measured using a Micromeritics ASAP 2010M sorptometer. Before measured at 77 K, the samples were degassed at 473 K for 12 h. Specific surface areas and pore size distributions were calculated using the Brunauer–Emmett–Teller (BET) and Density Functional Theory (DFT) models from the adsorption branches, respectively. The morphologies of the prepared samples were characterized using a scanning electron microscope (SEM, Hitachi S4800) at an accelerating voltage of 20 kV, and sample elements were analyzed by an energy dispersive spectrometer (EDS), associated with SEM. Fourier transform infrared (FTIR) spectra were recorded on a Perkin-Elmer 580B Infrared Spectrophotometer using the KBr pellet technique. A UV–vis spectrum which was used to describe the properties of the drug release was taken on a Lambda 45 spectrophotometer. Transmission electron microscopy (TEM) images were recorded on TECNAI F20. 3. Results and discussion
ð2Þ
where Ctorr is the corrected concentration at time t, Ct is the apparent concentration at time t, v is the volume of sample taken, and V is the
3.1. Morphology and structure Fig. 1 shows the small-angle XRD patterns of the MBG materials. It can be seen that there is a strong (100) diffraction peaks of MBG at
Fig. 2. SEM images of A) pure MBG, B) MBG-MH-1SBF (2 h), C) MBG-MH-1SBF (4 h), D) MBG-MH-1SBF (6 h), E) MBG-MH-2SBF (2 h), F) MBG-MH-2SBF (4 h), and G) MBG-MH-2SBF (6 h).
240
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
about 2θ = 1.2° [20], suggesting the ordered mesoporous structure. The SEM images of the MBG and mineralized samples with different mineralization time and ion concentration are shown in Fig. 2. Fig. 2A shows the SEM image of pure MBG, revealing the smooth surface of the pure MBG. After immerged in the SBF, the surface of the samples obviously becomes coarse. From Fig. 2B–D, the surface particles become more and more with the increase of the mineralization time. Comparing Fig. 2B–D with Fig. 2E–G, the high ion concentration is beneficial to the growth of HAp. From Fig. 2G, a lot of HAp particles cover on the surface of the MBG like a lot of flowers. In order to further reveal the growth of HAp, the wide angle XRD was tested. From Fig. S1, the mineralized MBGs reveal the typical (211) and (203) diffraction peaks, which was ascribed to the appearance of HAp (JCPDS No. 09-0432). As shown in Fig. S2, all samples are composed of Ca, P, Si, and O. The surface components of materials are summarized in Table 2 according to EDS. From Table 2, the Ca/P of pure MBG is 12.1, and the Ca/Ps of MBG-MH-1SBFs decrease to 8.80, 7.17, and 3.03 by the mineralization time 2, 4, and 6 h, respectively. When the MBG is mineralized in 2SBF, the Ca/P of the MBG-MH-2SBFs is lower than that of MBG-MH-1SBFs. With the mineralization time increases from 2 h to 6 h, the Ca/P decreases from 4.40 to 1.67 of the MBG-MH-2SBFs. The 2SBF systems show the lower Ca/P than that of MBG-MH-1SBFs. As we know, the Ca/P of HAp (Ca10(PO4)6(OH)2) is approximate 1.67 and it is believed that the decrease of the Ca/P is due to the HAp growth on the surface of the samples. That is also consistent with the SEM and EDS: the long mineralization time and the high ion concentration benefit to the growth of HAp. And there is no diffraction peak which can be assigned to MH (Fig. S4), suggesting the MH molecule is loaded in the pore of MBGs. From the above investigations, the growth of particles from SEM images is HAp, in accordance with the original report [21]. N2 adsorption–desorption analysis of the samples are shown in Fig. 3. All samples reveal the typical IV isotherm and H1 hysteresis loop, suggesting the cylindrical channel structure of the sample. From the adsorption–desorption isotherm, the adsorption amounts decrease with the increase of the mineralization time, further testifying the drug loading and the growth of HAp gain block the pore. These also can be proved by the change of the pore characters. From Table 2, the surface area, pore volume, and pore size decrease after drug loading and HAp mineralization. And with the longer mineralization time and higher ion concentration, the pore characters decrease more, suggesting the long mineralization time and high ion concentration benefit the HAp mineralization. That is also further testified the above conclusions. In addition, the drug loading mounts of these samples are similar to each other, illuminating the change of the pore characters of MBGMH-xSBF (yh) is ascribed to the HAp growth. TEM was also used to confirm mesoporous structure and the HAp growth. From Fig. 4A, MBG possesses the ordered straight mesoporous channel. After incubating in the SBF, the ordered straight mesoporous channel also can be found from their TEM images. And the surfaces of MBGs are successfully covered with a lot of HAp particles. Fig. S3 is the HRTEM image of Fig. 4B shows the typical HAp (100) plane with the d100 = 0.825 nm.
Fig. 3. Nitrogen adsorption–desorption isotherm and pore size distribution of the samples.
FTIR was used to provide evidence of the functional group. From Fig. 5B–G, the peaks at 1080 and 794 cm−1 are attributed to the antisymmetric absorption and symmetric absorption of Si\O\Si. Compare with pure MBG (Fig. 5B), the stronger peaks at 574 and 459 cm−1, which are attributed to the anti-symmetric angle changed peaks and symmetric angle changed peaks of PO3− 4 , can prove the HAp are successfully deposited on MBG. From Fig. 5C–G, the peak at 3173 cm−1 is attributed to the \NH group of MH, which testifies that the drug have been loaded into MBGs. From Table 2, the drug loading amounts for all systems are close to each other (25–29%), due to the same drug host and similar loading process. 3.2. Drug release profiles
Table 2 Characteristics of MBGs with different mineralization time and ion concentration. Samples
BET (m2/g)
Vp (mL/g)
Pore size (nm)
Drug loading efficiency (%)
Ca/Pa
MBG MBG-MH-1SBF (2 h) MBG-MH-1SBF (4 h) MBG-MH-1SBF (6 h) MBG-MH-2SBF (2 h) MBG-MH-2SBF (4 h) MBG-MH-2SBF (6 h)
302.574 301.807 300.254 298.213 272.307 269.364 257.845
4.34 4.32 4.05 3.84 3.87 3.84 3.63
0.376 0.335 0.301 0.277 0.288 0.269 0.242
25 26 28 27 28 29 26
12.12 8.83 7.17 3.03 4.44 1.72 1.67
a
Ca/P is calculated from EDS of SEM.
± ± ± ± ± ± ±
1.25 1.83 1.76 1.11 2.05 1.78 2.13
Due to the same drug loading mode and little influence of HAp for the drug loading efficacy, the drug efficiency doesn't have obvious distinction, as shown in Table 2. Fig. 6 shows the MH release profiles from pure MBG and MBG-MH-xSBF (yh)s at pH = 4.0 media. From Fig. 6, the pure MBG shows the fastest MH release. The release amounts reach approximately 61% at 0.5 h and it just takes 2 h to reach approximately 82%. Without the HAp growth, there is no cap to block the pore, so that the release process of pure MBG is just controlled by the mesopore. Comparing the release process of MBG-MH-1SBFs from different mineralization times, we found that the release rate decreases with the increase of the mineralization time. And with the same mineralization time, the MBG-MH-2SBF systems show the well control ability
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
241
Fig. 4. TEM images of A) pure MBG, B) MBG-MH-1SBF (2 h), C) MBG-MH-1SBF (4 h), D) MBG-MH-1SBF (6 h), E) MBG-MH-2SBF (2 h), F) MBG-MH-2SBF (4 h), and G) MBG-MH-2SBF (6 h).
than that of MBG-MH-1SBF. The growth of HAp nanoparticles blocks the mesoporous channels which hamper the drug release, making the release rate decrease. In the release media (pH = 4.0), the degradation of the HAp makes the drug release from the pore of the materials. From the above investigation, with the increase of incubation time
Fig. 5. FTIR spectra for A) metformin hydrochloride, B) MBG, C) the drug-loading (2 h) bio-glass in water, D) the drug-loading (2 h) bio-glass in simulated body fluid, E) the drug-loading (3 h) bio-glass in simulated body fluid, F) the drug-loading (4 h) bio-glass in simulated body fluid, and G) the drug-loading (6 h) bio-glass in simulated body fluid.
and ion concentration, more HAp particles grow to block the mesoporous channels to decrease the release rate. To investigate the pH-triggered drug release kinetics of mineralized MBGs, the release curves of MBG-MH-1SBF (4 h) performed at different pH media (Fig. 7) were investigated. From the release curves of pH = 4, it takes 0.5 h to reach approximately 50%, and the approximate 74%
Fig. 6. MH release profiles from MBG, MBG-MH-1SBF and MBG-MH-2SBF with various incubation time in pH = 4.0.
242
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
A
are ascribed to the dissolving of HAp. And the acid condition is beneficial for the dissolve of HAp cap, so that the fastest first-step release was occurring at pH = 4.0. Due to the dissolving of HAp, a lot of Ca2+ and PO3− 4 were dissolved into the medium. The HAp nanoparticles gradually lose of the blocking capacity, making the drug release. At the secondstep, when the dissolving of Ca2+ and PO3− from HAp-cap reached at 4 equilibrium, the drug release was mainly regulated by the mesopores. 4. Conclusion In summary, a pH-responsive controlled release system with different release kinetics by using MBG as host and HAp nanoparticles as capping agents. Considering the spontaneous mineralization of HAp on MBG in SBF, the drug loading and HAp-cap growth were carried out at one-step. The HAp nanoparticles block the pore of the MBG to regulate the drug release. It is a new, simple, and effective method to obtain the pH-responsive drug release system. Under adjusting the mineralization time and the ion concentration, the different drug release profiles were obtained. Because the degradation of HAp tends more easily at the acid media, the drug loading system shows the pH-responsive release process. The results make the system a promising candidate as a pHsensitive vehicle for in vivo delivery of therapeutic agents to low pH tissues, such as inflammatory sites and tumors. This synthetic approach could also provide a novel route to realize the spontaneous growth of the functional parts that is helpful to be used for other synthesis of the hybrid systems.
B
Acknowledgments
Fig. 7. (A) MH release profiles from MBG-MH-1SBF (4 h) under pH control and (B) the Higuchi plot for the release of MH from MBGs.
Financial support for this study was provided by the National Native Science Foundation of China (21171045, 21101046), Innovation special fund of Harbin Science and Technology Bureau of China (2010RFXXS055), Program for Scientific and Technological Innovation Team Construction in Universities of Heilongjiang Province (2011TD010), Research Fund for the Doctoral Program of Higher Education of China (20102329110002), and Technology Development Preproject of Harbin Normal University (12XYG-11). Appendix A. Supplementary data
release amount which just costs 6 h is ascribed to the fast degradation of HAp at this strong acid condition. From Fig. 7, with the increase of pH value, the releases become obviously slow. The burst release is approximately 50, 28, 22 and 16% at pH = 4.0, 5.0, 6.8, and 7.4 at the first 0.5 h, respectively. And at 24 h, the releases reach approximately 76, 64, 49 and 34% with pH from 4 to 7.4. It is known that, HAp is degradable at acid media, and tends steady at neutral and alkaline conditions. From Fig. 7, the different releases can ascribe to the different degradation abilities at the various media conditions. The drug loading system shows the well pH-sensitive drug release capability. Drug release kinetics from an insoluble, porous carrier matrix are frequently described by the Higuchi model [22,23], and the release rate can be described by Eq. (3): Q ¼k t
1=2
ð3Þ
where Q is the amount of drug released from the materials, t denotes time, and k is the Higuchi dissolution constant. According to the model, for a purely diffusion-controlled process, the linear relationship is valid for the release of relatively small molecules distributed uniformly throughout the carrier [23]. However, we observe that all releasing systems display (approximately) a two-step release based on the Higuchi model (Fig. 7B), which is in agreement with Andersson's report [24]. The MBG-MH-1SBF (4 h) in pH = 4.0 system only took 2 h to finish the first-step release, while the other three systems took more than 4 h for the first-step release. In the first-step, the drug release behaviors
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2013.12.006. References [1] K. Song, Y. Liu, H.M. Macedo, L. Jiang, C. Li, G. Mei, T. Liu, Mater. Sci. Eng. C 33 (2013) 1506–1513. [2] M.H. Xiong, Y. Bao, X.Z. Yang, Y.C. Wang, B.L. Sun, J. Wang, J. Am. Chem. Soc. 134 (2012) 4355–4362. [3] H.Y. Zhou, X.G. Chen, C.S. Liu, X.H. Meng, C.G. Liu, L.J. Yu, Biochem. Eng. J. 31 (2006) 228–233. [4] W. Lee, L.C. Loei, E. Widjaja, S.C.J. Loo, J. Control. Release 151 (2011) 229–238. [5] Y.J. Yang, X. Tao, Q. Hou, Y. Ma, X.L. Chen, J.F. Chen, Acta Biomater. 6 (2010) 3092–3100. [6] H.Y. Tang, J. Guo, Y. Sun, B.S. Chang, Q.G. Ren, W.L. Yang, Int. J. Pharm. 421 (2011) 388–396. [7] A. Baeza, E. Guisasda, E.R. Hemandez, M.V. Regi, Chem. Mater. 24 (2012) 517–524. [8] V. Escoto, J.L. Slowing, I.I. Wu, C. Lin, V.S-Y, J. Am. Chem. Soc. 131 (2009) 3462–3463. [9] E. Aznar, M.D. Marcos, R. Martinez-Manez, F. Sancenon, J. Soto, P. Amoro, P. Guillem, J. Am. Chem. Soc. 131 (2009) 6833–6843. [10] C.Y. Lai, B.G. Trewyn, D.M. Jeftinija, K. Jeftinija, S. Xu, S. Jeftinija, V.S.Y. Lin, J. Am. Chem. Soc. 125 (2003) 4451–4459. [11] S. Giri, B.G. Trewyn, M.P. stellmaker, V.S.Y. Lin, Angew. Chem. Int. Ed. 44 (2005) 5038–5044. [12] F. Muhammad, M. Guo, W.X. Qi, F. Sun, A.F. Wang, Y.J. Guo, G.S. Zhu, J. Am. Chem. Soc. 133 (2011) 8778–8781. [13] L.L. Hench, J. Eur. Ceram. Soc. 29 (2008) 1257–1265. [14] L.L. Hench, Biomaterials 19 (1998) 1419–1432. [15] S.V. Porozhkin, M. Epple, Angew. Chem. Int. Ed. 41 (2002) 3130–3146. [16] Y.P. Guo, Y. Zhou, D. Jia, H.X. Tang, Microporous Mesoporous Mater. 118 (2009) 480–488.
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243 [17] H.P. Rim, K.H. Min, H.J. Lee, S.Y. Jeong, S.C. Lee, Angew. Chem. Int. Ed. 50 (2011) 8853–8857. [18] Y. Zhu, C. Wu, Y. Ramaswamy, E. Kockrick, P. Simon, S. Kaskel, Microporous Mesoporous Mater. 112 (2008) 494–503. [19] F.Y. Qu, G.S. Zhu, S.Y. Huang, S.G. Li, J.Y. Sun, D.L. Zhang, S.L. Qiu, Microporous Mesoporous Mater. 92 (2006) 1–9.
243
[20] X. Yan, C.Z. Yu, X.F. Zhou, J.W. Tang, D.Y. Zhao, Angew. Chem. Int. Ed. 43 (2004) 5980–5984. [21] D. Arcos, J. Peña, M. Vallet-Regí, Chem. Mater. 15 (2003) 4132–4138. [22] T. Higuchi, J. Pharm. Sci. 50 (1961) 874–875. [23] T. Higuchi, J. Pharm. Sci. 52 (1963) 1145–1149. [24] J. Andersson, J. Rosenholm, S. Areva, M. LindOn, Chem. Mater. 16 (2004) 4160–4167.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
pH-responsive controlled-release system based on mesoporous bioglass materials capped with mineralized hydroxyapatite Chunyu Yang, Wei Guo, Liru Cui, Di Xiang, Kun Cai, Huiming Lin ⁎, Fengyu Qu ⁎ Key Laboratory of Design and Synthesis of Functional Materials and Green Catalysis, College of Chemistry and Chemical Engineering, Harbin Normal University, Harbin 150025, PR China
a r t i c l e
i n f o
Article history: Received 26 May 2013 Received in revised form 28 November 2013 Accepted 6 December 2013 Available online 16 December 2013 Keywords: Bioglass mesoporous materials Mineralized hydroxyapatite pH-responsive Controlled-release
a b s t r a c t A controlled release system with pH-responsive ability has been presented. Mesoporous bioglass (MBG) was used as the drug carrier and a spontaneous mineralization method was adopted to cap the pores of the carrier with hydroxyapatite (HAp) and to restrict the drug release. It is a simple and green method to realize the ingenious pH-sensitive controlled release. The model drug, metformin hydrochloride (MH), was loaded simultaneously with the mineralization process. Due to the degradation of HAp at acid environments, the system shows well pH-sensitive drug release ability. The release kinetics can be easily adjusted by the mineralization time and the ion concentration of media. The system is recommended as a promising candidate as a pH-sensitive vehicle for drug controlled release to low pH tissues, such as inflammatory sites and tumors. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The drug delivery has attracted considerable interests in the past few decades. The main advantage of drug delivery is to allow optimizing the security and efficacy of the treatments. At present, cancer has become one of the most serious diseases to threaten people's health. Among the various treatments, the medication therapy was considered as the most effective method. However, medication effect has been greatly limited due to the toxic effects of the nonspecific distribution of anticancer drugs. That is because most anticancer drugs cannot distinguish between cancerous and healthy cells, inducing collateral damages and adverse side effects. To address this formidable challenge, using the proper methods to realize the controlled release to increase the drug effect become the most important in the biopharmaceutical field. At present, many materials are used as the hosts to load drugs, showing the different drug release kinetics [1–4]. With uniform and tunable morphology and pore size, large surface area, high pore volume, modifiable surface property, and good biocompatibility and biodegradation, mesoporous silica nanomaterials (MSNs) have become one of the most important hosts for drug delivery. With the various functionalities, many organic materials are always used to realize the controllable release process. For example, PAH/PSS and ALG/CHI based on mesoporous silica nanotubes have been developed as pH-controlled drug delivery systems via the layer by layer self-assembly technique [5]. Yang et al. synthesized a pH responsive nanocarriers with chitosan-poly (methacrylic acid) shells and MSN cores on one step. The system showed well physicochemical and pH sensitive controlled release properties [6]. Vallet⁎ Corresponding authors. Tel.: +86 451 88060653. E-mail addresses: [email protected] (H. Lin), [email protected] (F. Qu). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.12.006
Regí et al. reported on the synthesis of (PEI/NIPAM)/magnetic mesoporous silica nanocarrier, which was able to release different cargos in a controlled manner in response to an external alternating magnetic field [7]. Recently, due to the cost and complexity of the use of organic components, some inorganic nanoparticles [8–10] have been tried as the capping agents outside MSNs. There are two release mechanisms for these drug loading systems: 1) the breaking of the link between the cap and the host. Lin et al. synthesized mesoporous silica capped with Fe3O4. The controlled-release is based on breaking the disulfide linkage between the Fe3O4 nanoparticle caps and the linker-MSN hosts by reducing agents such as DHLA [11]. 2) The cap collapse makes the drug release. Zhu et al. used ZnO QDs to seal the pores of mesoporous silica. ZnO QDs lid on MSNs can be efficiently dissolved in the acidic intracellular compartments of cancer cells, resulting in the release of the drug cargo from the pores of the MSNs into the cytosol [12]. Although these works used the inorganic nanoparticles as the cap which showed well responsive controlled release, they could not avoid the use of the organic components as the linkage between cap and host. And the synthetic methods are often multistep procedures or harsh synthetic conditions. So that, a simple, green, and effective approach to produce a pHresponsive drug delivery system is still a great challenge. With well bioactivity, mesoporous bioglass (MBG) has been widely used in bone tissue reparation and regeneration [13,14]. MBG can induce the growth of hydroxyapatite (HAp) in vivo and in vitro (SBF, 37 °C) as well [15]. It is known that, HAp is the major inorganic component of natural bones with nontoxic, nonirritant and well bioabsorbabitity ability [16]. In addition, it can be dissolved producing nontoxic ions (calcium and phosphate ions) in acidic cellular environments such as endosomes (pH = 5.0) and lysosomes (pH = 4.5) [17].
238
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
Based on the above considerations, we developed a simple and green method to produce a drug loaded MBG system with pHsensitive controlled release ability. As shown in Scheme 1, MBG was used as the drug carrier. Metformin hydrochloride (MH, with the similar hydrophilic property of the doxorubicin hydrochloride) was use as the model drug. After drug loading process, the pores of the carrier were filled with drug molecules. Then, the SBF was added and the HAp nanoparticle layer formed outside owing to the spontaneous induced growth of HAp on the MBG. HAp nanoparticles can block the pore channels as a cap to prevent the drug release. In acidic cellular environments, the degradation of HAp cap makes the drug molecules release. The pH-sensitive drug release performance of the system was investigated in detail. And the mineralization time and the concentration of the SBF can be used to control the release process. The system can be used as a pH-sensitive vehicle for in vivo delivery of therapeutic agents to low pH tissues, such as inflammatory sites and tumors. 2. Experimental 2.1. Materials Tetraethyl orthosilicate (TEOS, PA), Ca(NO3)2·4H2O, triethyl phosphate (TEP, 99.8%, PA), metformin hydrochloride (MH, pharmacopeia) and ethanol (PA) were purchased from Tianjin Chemical Corp. of China. Poly (ethylene)-block-poly (propylene glycol)-block-poly (ethylene) (P123, PA) was purchased from Aldrich. All the chemicals were used as received. Deionized water was used for all experiments. 2.2. Preparation of MBG scaffolds Porous mesopore-bioglass (MBG) scaffolds were prepared by the sol– gel processing of the previous publication [18]. Specifically, 4.0 g of P123 (Mw = 5800, Aldrich), 6.7 g of tetraethyl orthosilicate (TEOS), 1.4 g of Ca(NO3)2·4H2O, 0.73 g of triethyl phosphate (TEP, 99.8%), and 1.0 g of 0.5 M HCl were dissolved in 60 g of ethanol (Si/Ca/P = 80:15:5 molar ratio) and stirred at room temperature for 24 h. Afterwards, the solution was transferred to a Petri dish and then allowed to evaporate at room temperature for 3–5 days. When the samples were completely dry, they were calcined at 700 °C for 6 h to obtain the MBG scaffolds.
Table 1 Ion concentration and pH of SBF and 2SBF.
SBF 2SBF
Na+ (mM)
K+ (mM)
Mg2+ (mM)
Ca2+ (mM)
Cl−
HCO3− (mM)
HPO2− 4 (mM)
SO2− 4 (mM) pH
pH
142 142
5 5
1.5 1.5
2.5 5
103.0 103
27 27
1.0 2.0
0.5 0.5
7.40 6.80
2.3. Drug loading The loading of the drug was conducted as the previous report [19]. A typical loading procedure of MH in MBG was as follows: 0.206 g of MBG with 2.06 g of MH was suspended in 50 mL distilled water and stirring for 2 h at room temperature. The vial was sealed to prevent the evaporation of solvent. After the drug loading process, 50 ml xSBF was added and standing for a period at 37 °C. The system was separated from the solution by vacuum filtration and dried in vacuum. In the process of vacuum filtration, sintered glass funnel (G5, pore size 1.5–2.5 μm) was used to separate the drug loaded samples from the solutions. The obtained solution was appropriately diluted, and the residual MH amounts were determined by UV/Vis measurements at 233 nm. The loading efficiency (LE %) of MH can be calculated by using formula (1). The experiment repeated three times. LE% ¼
mðoriginal
MHÞ −mðresidual MHÞ
mðMBGÞ þ mðoriginal
MHÞ −mðresidual MHÞ
100%
ð1Þ
The solid sample is named as MBG-MH-xSBF (yh). The x was used to describe the media solution concentration × times as SBF (x = 1 or 2, Table 1). y means the mineralization time. 2.4. Drug release 150 mg of MBG-MH-xSBF (yh) was soaked in 500 mL of media solution (pH = 4.0 HCl aqueous, pH = 5.0 phosphate buffer solution, pH = 6.8 phosphate buffer solution, pH = 7.4 phosphate buffer solution) at 37 °C. At predetermined time intervals, 3 mL samples have been withdrawn and immediately replaced with an equal volume of new medium to keep the volume constant. The withdrawn samples were filtered (0.45 um), properly diluted and monitored by UV–vis
Scheme 1. Schematic illumination of the drug assembly and release process.
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
239
total volume of the dissolution medium. This experiment has been repeated for 3 times. 2.5. Physicochemical characterization
Fig. 1. Small-angle XRD patterns for MBG.
spectrophotometer at 233 nm. Calculation of the corrected concentration of the released MH is based on the following formula: Ctorr ¼ Ct þ
t−1 vX C V 0 t
Powder X-ray patterns (XRD) were recorded on a SIEMENS D5005 X-ray diffractometer with Cu Kα radiation (40 kV, 30 mA). The nitrogen adsorption/desorption, surface areas, and median pore diameters were measured using a Micromeritics ASAP 2010M sorptometer. Before measured at 77 K, the samples were degassed at 473 K for 12 h. Specific surface areas and pore size distributions were calculated using the Brunauer–Emmett–Teller (BET) and Density Functional Theory (DFT) models from the adsorption branches, respectively. The morphologies of the prepared samples were characterized using a scanning electron microscope (SEM, Hitachi S4800) at an accelerating voltage of 20 kV, and sample elements were analyzed by an energy dispersive spectrometer (EDS), associated with SEM. Fourier transform infrared (FTIR) spectra were recorded on a Perkin-Elmer 580B Infrared Spectrophotometer using the KBr pellet technique. A UV–vis spectrum which was used to describe the properties of the drug release was taken on a Lambda 45 spectrophotometer. Transmission electron microscopy (TEM) images were recorded on TECNAI F20. 3. Results and discussion
ð2Þ
where Ctorr is the corrected concentration at time t, Ct is the apparent concentration at time t, v is the volume of sample taken, and V is the
3.1. Morphology and structure Fig. 1 shows the small-angle XRD patterns of the MBG materials. It can be seen that there is a strong (100) diffraction peaks of MBG at
Fig. 2. SEM images of A) pure MBG, B) MBG-MH-1SBF (2 h), C) MBG-MH-1SBF (4 h), D) MBG-MH-1SBF (6 h), E) MBG-MH-2SBF (2 h), F) MBG-MH-2SBF (4 h), and G) MBG-MH-2SBF (6 h).
240
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
about 2θ = 1.2° [20], suggesting the ordered mesoporous structure. The SEM images of the MBG and mineralized samples with different mineralization time and ion concentration are shown in Fig. 2. Fig. 2A shows the SEM image of pure MBG, revealing the smooth surface of the pure MBG. After immerged in the SBF, the surface of the samples obviously becomes coarse. From Fig. 2B–D, the surface particles become more and more with the increase of the mineralization time. Comparing Fig. 2B–D with Fig. 2E–G, the high ion concentration is beneficial to the growth of HAp. From Fig. 2G, a lot of HAp particles cover on the surface of the MBG like a lot of flowers. In order to further reveal the growth of HAp, the wide angle XRD was tested. From Fig. S1, the mineralized MBGs reveal the typical (211) and (203) diffraction peaks, which was ascribed to the appearance of HAp (JCPDS No. 09-0432). As shown in Fig. S2, all samples are composed of Ca, P, Si, and O. The surface components of materials are summarized in Table 2 according to EDS. From Table 2, the Ca/P of pure MBG is 12.1, and the Ca/Ps of MBG-MH-1SBFs decrease to 8.80, 7.17, and 3.03 by the mineralization time 2, 4, and 6 h, respectively. When the MBG is mineralized in 2SBF, the Ca/P of the MBG-MH-2SBFs is lower than that of MBG-MH-1SBFs. With the mineralization time increases from 2 h to 6 h, the Ca/P decreases from 4.40 to 1.67 of the MBG-MH-2SBFs. The 2SBF systems show the lower Ca/P than that of MBG-MH-1SBFs. As we know, the Ca/P of HAp (Ca10(PO4)6(OH)2) is approximate 1.67 and it is believed that the decrease of the Ca/P is due to the HAp growth on the surface of the samples. That is also consistent with the SEM and EDS: the long mineralization time and the high ion concentration benefit to the growth of HAp. And there is no diffraction peak which can be assigned to MH (Fig. S4), suggesting the MH molecule is loaded in the pore of MBGs. From the above investigations, the growth of particles from SEM images is HAp, in accordance with the original report [21]. N2 adsorption–desorption analysis of the samples are shown in Fig. 3. All samples reveal the typical IV isotherm and H1 hysteresis loop, suggesting the cylindrical channel structure of the sample. From the adsorption–desorption isotherm, the adsorption amounts decrease with the increase of the mineralization time, further testifying the drug loading and the growth of HAp gain block the pore. These also can be proved by the change of the pore characters. From Table 2, the surface area, pore volume, and pore size decrease after drug loading and HAp mineralization. And with the longer mineralization time and higher ion concentration, the pore characters decrease more, suggesting the long mineralization time and high ion concentration benefit the HAp mineralization. That is also further testified the above conclusions. In addition, the drug loading mounts of these samples are similar to each other, illuminating the change of the pore characters of MBGMH-xSBF (yh) is ascribed to the HAp growth. TEM was also used to confirm mesoporous structure and the HAp growth. From Fig. 4A, MBG possesses the ordered straight mesoporous channel. After incubating in the SBF, the ordered straight mesoporous channel also can be found from their TEM images. And the surfaces of MBGs are successfully covered with a lot of HAp particles. Fig. S3 is the HRTEM image of Fig. 4B shows the typical HAp (100) plane with the d100 = 0.825 nm.
Fig. 3. Nitrogen adsorption–desorption isotherm and pore size distribution of the samples.
FTIR was used to provide evidence of the functional group. From Fig. 5B–G, the peaks at 1080 and 794 cm−1 are attributed to the antisymmetric absorption and symmetric absorption of Si\O\Si. Compare with pure MBG (Fig. 5B), the stronger peaks at 574 and 459 cm−1, which are attributed to the anti-symmetric angle changed peaks and symmetric angle changed peaks of PO3− 4 , can prove the HAp are successfully deposited on MBG. From Fig. 5C–G, the peak at 3173 cm−1 is attributed to the \NH group of MH, which testifies that the drug have been loaded into MBGs. From Table 2, the drug loading amounts for all systems are close to each other (25–29%), due to the same drug host and similar loading process. 3.2. Drug release profiles
Table 2 Characteristics of MBGs with different mineralization time and ion concentration. Samples
BET (m2/g)
Vp (mL/g)
Pore size (nm)
Drug loading efficiency (%)
Ca/Pa
MBG MBG-MH-1SBF (2 h) MBG-MH-1SBF (4 h) MBG-MH-1SBF (6 h) MBG-MH-2SBF (2 h) MBG-MH-2SBF (4 h) MBG-MH-2SBF (6 h)
302.574 301.807 300.254 298.213 272.307 269.364 257.845
4.34 4.32 4.05 3.84 3.87 3.84 3.63
0.376 0.335 0.301 0.277 0.288 0.269 0.242
25 26 28 27 28 29 26
12.12 8.83 7.17 3.03 4.44 1.72 1.67
a
Ca/P is calculated from EDS of SEM.
± ± ± ± ± ± ±
1.25 1.83 1.76 1.11 2.05 1.78 2.13
Due to the same drug loading mode and little influence of HAp for the drug loading efficacy, the drug efficiency doesn't have obvious distinction, as shown in Table 2. Fig. 6 shows the MH release profiles from pure MBG and MBG-MH-xSBF (yh)s at pH = 4.0 media. From Fig. 6, the pure MBG shows the fastest MH release. The release amounts reach approximately 61% at 0.5 h and it just takes 2 h to reach approximately 82%. Without the HAp growth, there is no cap to block the pore, so that the release process of pure MBG is just controlled by the mesopore. Comparing the release process of MBG-MH-1SBFs from different mineralization times, we found that the release rate decreases with the increase of the mineralization time. And with the same mineralization time, the MBG-MH-2SBF systems show the well control ability
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
241
Fig. 4. TEM images of A) pure MBG, B) MBG-MH-1SBF (2 h), C) MBG-MH-1SBF (4 h), D) MBG-MH-1SBF (6 h), E) MBG-MH-2SBF (2 h), F) MBG-MH-2SBF (4 h), and G) MBG-MH-2SBF (6 h).
than that of MBG-MH-1SBF. The growth of HAp nanoparticles blocks the mesoporous channels which hamper the drug release, making the release rate decrease. In the release media (pH = 4.0), the degradation of the HAp makes the drug release from the pore of the materials. From the above investigation, with the increase of incubation time
Fig. 5. FTIR spectra for A) metformin hydrochloride, B) MBG, C) the drug-loading (2 h) bio-glass in water, D) the drug-loading (2 h) bio-glass in simulated body fluid, E) the drug-loading (3 h) bio-glass in simulated body fluid, F) the drug-loading (4 h) bio-glass in simulated body fluid, and G) the drug-loading (6 h) bio-glass in simulated body fluid.
and ion concentration, more HAp particles grow to block the mesoporous channels to decrease the release rate. To investigate the pH-triggered drug release kinetics of mineralized MBGs, the release curves of MBG-MH-1SBF (4 h) performed at different pH media (Fig. 7) were investigated. From the release curves of pH = 4, it takes 0.5 h to reach approximately 50%, and the approximate 74%
Fig. 6. MH release profiles from MBG, MBG-MH-1SBF and MBG-MH-2SBF with various incubation time in pH = 4.0.
242
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243
A
are ascribed to the dissolving of HAp. And the acid condition is beneficial for the dissolve of HAp cap, so that the fastest first-step release was occurring at pH = 4.0. Due to the dissolving of HAp, a lot of Ca2+ and PO3− 4 were dissolved into the medium. The HAp nanoparticles gradually lose of the blocking capacity, making the drug release. At the secondstep, when the dissolving of Ca2+ and PO3− from HAp-cap reached at 4 equilibrium, the drug release was mainly regulated by the mesopores. 4. Conclusion In summary, a pH-responsive controlled release system with different release kinetics by using MBG as host and HAp nanoparticles as capping agents. Considering the spontaneous mineralization of HAp on MBG in SBF, the drug loading and HAp-cap growth were carried out at one-step. The HAp nanoparticles block the pore of the MBG to regulate the drug release. It is a new, simple, and effective method to obtain the pH-responsive drug release system. Under adjusting the mineralization time and the ion concentration, the different drug release profiles were obtained. Because the degradation of HAp tends more easily at the acid media, the drug loading system shows the pH-responsive release process. The results make the system a promising candidate as a pHsensitive vehicle for in vivo delivery of therapeutic agents to low pH tissues, such as inflammatory sites and tumors. This synthetic approach could also provide a novel route to realize the spontaneous growth of the functional parts that is helpful to be used for other synthesis of the hybrid systems.
B
Acknowledgments
Fig. 7. (A) MH release profiles from MBG-MH-1SBF (4 h) under pH control and (B) the Higuchi plot for the release of MH from MBGs.
Financial support for this study was provided by the National Native Science Foundation of China (21171045, 21101046), Innovation special fund of Harbin Science and Technology Bureau of China (2010RFXXS055), Program for Scientific and Technological Innovation Team Construction in Universities of Heilongjiang Province (2011TD010), Research Fund for the Doctoral Program of Higher Education of China (20102329110002), and Technology Development Preproject of Harbin Normal University (12XYG-11). Appendix A. Supplementary data
release amount which just costs 6 h is ascribed to the fast degradation of HAp at this strong acid condition. From Fig. 7, with the increase of pH value, the releases become obviously slow. The burst release is approximately 50, 28, 22 and 16% at pH = 4.0, 5.0, 6.8, and 7.4 at the first 0.5 h, respectively. And at 24 h, the releases reach approximately 76, 64, 49 and 34% with pH from 4 to 7.4. It is known that, HAp is degradable at acid media, and tends steady at neutral and alkaline conditions. From Fig. 7, the different releases can ascribe to the different degradation abilities at the various media conditions. The drug loading system shows the well pH-sensitive drug release capability. Drug release kinetics from an insoluble, porous carrier matrix are frequently described by the Higuchi model [22,23], and the release rate can be described by Eq. (3): Q ¼k t
1=2
ð3Þ
where Q is the amount of drug released from the materials, t denotes time, and k is the Higuchi dissolution constant. According to the model, for a purely diffusion-controlled process, the linear relationship is valid for the release of relatively small molecules distributed uniformly throughout the carrier [23]. However, we observe that all releasing systems display (approximately) a two-step release based on the Higuchi model (Fig. 7B), which is in agreement with Andersson's report [24]. The MBG-MH-1SBF (4 h) in pH = 4.0 system only took 2 h to finish the first-step release, while the other three systems took more than 4 h for the first-step release. In the first-step, the drug release behaviors
Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2013.12.006. References [1] K. Song, Y. Liu, H.M. Macedo, L. Jiang, C. Li, G. Mei, T. Liu, Mater. Sci. Eng. C 33 (2013) 1506–1513. [2] M.H. Xiong, Y. Bao, X.Z. Yang, Y.C. Wang, B.L. Sun, J. Wang, J. Am. Chem. Soc. 134 (2012) 4355–4362. [3] H.Y. Zhou, X.G. Chen, C.S. Liu, X.H. Meng, C.G. Liu, L.J. Yu, Biochem. Eng. J. 31 (2006) 228–233. [4] W. Lee, L.C. Loei, E. Widjaja, S.C.J. Loo, J. Control. Release 151 (2011) 229–238. [5] Y.J. Yang, X. Tao, Q. Hou, Y. Ma, X.L. Chen, J.F. Chen, Acta Biomater. 6 (2010) 3092–3100. [6] H.Y. Tang, J. Guo, Y. Sun, B.S. Chang, Q.G. Ren, W.L. Yang, Int. J. Pharm. 421 (2011) 388–396. [7] A. Baeza, E. Guisasda, E.R. Hemandez, M.V. Regi, Chem. Mater. 24 (2012) 517–524. [8] V. Escoto, J.L. Slowing, I.I. Wu, C. Lin, V.S-Y, J. Am. Chem. Soc. 131 (2009) 3462–3463. [9] E. Aznar, M.D. Marcos, R. Martinez-Manez, F. Sancenon, J. Soto, P. Amoro, P. Guillem, J. Am. Chem. Soc. 131 (2009) 6833–6843. [10] C.Y. Lai, B.G. Trewyn, D.M. Jeftinija, K. Jeftinija, S. Xu, S. Jeftinija, V.S.Y. Lin, J. Am. Chem. Soc. 125 (2003) 4451–4459. [11] S. Giri, B.G. Trewyn, M.P. stellmaker, V.S.Y. Lin, Angew. Chem. Int. Ed. 44 (2005) 5038–5044. [12] F. Muhammad, M. Guo, W.X. Qi, F. Sun, A.F. Wang, Y.J. Guo, G.S. Zhu, J. Am. Chem. Soc. 133 (2011) 8778–8781. [13] L.L. Hench, J. Eur. Ceram. Soc. 29 (2008) 1257–1265. [14] L.L. Hench, Biomaterials 19 (1998) 1419–1432. [15] S.V. Porozhkin, M. Epple, Angew. Chem. Int. Ed. 41 (2002) 3130–3146. [16] Y.P. Guo, Y. Zhou, D. Jia, H.X. Tang, Microporous Mesoporous Mater. 118 (2009) 480–488.
C. Yang et al. / Materials Science and Engineering C 36 (2014) 237–243 [17] H.P. Rim, K.H. Min, H.J. Lee, S.Y. Jeong, S.C. Lee, Angew. Chem. Int. Ed. 50 (2011) 8853–8857. [18] Y. Zhu, C. Wu, Y. Ramaswamy, E. Kockrick, P. Simon, S. Kaskel, Microporous Mesoporous Mater. 112 (2008) 494–503. [19] F.Y. Qu, G.S. Zhu, S.Y. Huang, S.G. Li, J.Y. Sun, D.L. Zhang, S.L. Qiu, Microporous Mesoporous Mater. 92 (2006) 1–9.
243
[20] X. Yan, C.Z. Yu, X.F. Zhou, J.W. Tang, D.Y. Zhao, Angew. Chem. Int. Ed. 43 (2004) 5980–5984. [21] D. Arcos, J. Peña, M. Vallet-Regí, Chem. Mater. 15 (2003) 4132–4138. [22] T. Higuchi, J. Pharm. Sci. 50 (1961) 874–875. [23] T. Higuchi, J. Pharm. Sci. 52 (1963) 1145–1149. [24] J. Andersson, J. Rosenholm, S. Areva, M. LindOn, Chem. Mater. 16 (2004) 4160–4167.
