Polyelectrolyte-mediated hierarchical mesoporous

Showcasing collaborative research from the Western Seoul Center, Korea Basic Science Institute and the Department of Chemical and Biological Engineering, Hanbat National University in Korea.

As featured in:

Title: Polyelectrolyte-mediated hierarchical mesoporous calcium silicates: a platform for drug delivery carrier with ultrahigh loading capacity and controlled release behaviour Mesoporous calcium silicate hydrates (PAH-CS) with a large specific surface area and pore volume have been developed by a poly(allylamine hydrochloride)-assisted synthetic method. The very high cationic charge of PAH-CS due to an excess of Ca2+ and NH3+ existing on its surface resulted in an extremely high loading capacity for anionic drugs and proteins.

See Won San Choi, Ha-Jin Lee et al., J. Mater. Chem. B, 2015, 3, 1001.

www.rsc.org/MaterialsB Registered charity number: 207890

Journal of

Materials Chemistry B View Article Online

Published on 18 December 2014. Downloaded by Korea Basic Science Institute on 29/01/2015 04:35:28.

PAPER

View Journal | View Issue

Cite this: J. Mater. Chem. B, 2015, 3, 1001

Polyelectrolyte-mediated hierarchical mesoporous calcium silicates: a platform for drug delivery carrier with ultrahigh loading capacity and controlled release behavior† Md. Shahinul Islam,a Ha Neul Choi,a Won San Choi*b and Ha-Jin Lee*a We have developed a facile method for the poly(allylamine hydrochloride) (PAH)-assisted synthesis of mesoporous calcium silicate hydrates (PAH-CS) with a large specific surface area (BET ¼ 348.4 m2 g1) and pore volume (Vp ¼ 1.42 cm3 g1). Tetraethyl orthosilicate (TEOS) was employed as a silicon source, which was rapidly hydrolyzed and reacted with the amine groups of PAH to form spherical SiO2 nanoparticles (PAH-Si). Subsequently, Ca2+ ions reacted with the silicate anions produced during the dissolution of SiO2 in basic media, leading to the formation of the highly porous 3D networks of PAH-CS that were synthesized only under optimized reaction conditions. The PAH-CS containing an excess of Ca2+ and NH3+ enriched the surfaces with a very high cationic charge (z ¼ +65.66 mV)and resulted in an extremely high loading capacity for anionic drugs and proteins. Ibuprofen (IBU) and FITC-labeled bovine albumin (FITC-Albumin) were chosen as a model drug and model protein, respectively, to test the loading and delivery efficiencies of the PAH-CS carriers. The ultrahigh drug loading capacities (DLC) and their release patterns were investigated under controlled pH conditions. Strikingly, the highest DLC reported to date (IBU or FITC-Albumin/carrier (3.35 g or 1 g g1) was achieved in this work. The PAH-CS

Received 19th November 2014 Accepted 18th December 2014

had no cytotoxic effect on osteoblast-like MC3T3-E1 cell lines evaluated by the LDH (Lactate dehydrogenase) assay in supernatant medium. Furthermore, the PAH-CS carriers could be entirely

DOI: 10.1039/c4tb01911c

transformed to hydroxyapatite after releasing the drug in simulated body fluid (SBF), indicating good

www.rsc.org/MaterialsB

bioactivity and biodegradability of the PAH-CS carriers.

1. Introduction The synthesis of novel mesoporous materials with well-tailored properties, including a large surface area, high drug loading efficiency, potential for hybridization with other organic/inorganic materials, and biocompatibility, is a major challenge in biomedical research on the development of drug delivery carriers, diagnostic agents, sensing probes, and tracking labels.1,2 These materials can be synthesized using various templates such as surfactants, carbons, silicates, titanias, and phosphates.3–8 The cavities of mesoporous structures are easily produced by removing the templates using an acid treatment or calcination at high temperature, resulting in the formation of porous or hollow structures and facilitating the adsorption as well as controlled release of drugs.9 However, residual a

Western Seoul Center, Korea Basic Science Institute, 150 Bugahyun-ro, Seoudaemun-gu, Seoul, 120-140, Republic of Korea. E-mail: [email protected]

b

Department of Chemical and Biological Engineering, Hanbat National University, San 16-1, Dukmyoung dong, Yuseong-gu, Daejeon, 305-719, Republic of Korea. E-mail: [email protected] † Electronic supplementary 10.1039/c4tb01911c

information

(ESI)

This journal is © The Royal Society of Chemistry 2015

available.

See

DOI:

surfactants or the introduced organic groups may cause cytotoxicity in clinical applications and reduce the drug loading capacities (DLCs).10 Template-based polyelectrolyte (PE) microcapsules consisting of biocompatible anionic and cationic PEs have also attracted considerable attention for their use as drug carriers.11–13 However, decomposition of the templates may affect the integrity and the properties of the capsules, resulting in decreased drug loading and release efficiency.14,15 In addition, degradable polymer-based drug delivery systems have been mostly used for the specic and targeted delivery of therapeutic agents, antibiotics, and siRNA.16–18 They do not exhibit cytotoxicity to cells; therefore, some examples of applications in drug delivery systems have been reported with their low DLCs.19 Surfactant or organic-based drug carriers may cause cytotoxicity, and carriers based on biocompatible polymer show low DLCs. To overcome these disadvantages, the development of new biocompatible mesoporous materials with high DLCs as drug carriers is needed. Clinically used inorganic bioceramics, such as tricalcium phosphate, bioactive glass, hydroxyapatite (Ca10(PO4)6(OH)2), and calcium silicate (CaSiO3), have attracted considerable attention in

J. Mater. Chem. B, 2015, 3, 1001–1009 | 1001

View Article Online


Journal of Materials Chemistry B

Paper

recent years for potential applications in bone tissue engineering, drug delivery, and diagnostics.20–25 In particular, calcium silicate hydrates (CSH), which have been mainly regarded as simple components of glass, have been recently considered as excellent candidates for applications in bone regeneration and as drug carriers because of their good bioactivity, biodegradability, and high DLCs.26–29 To improve the bioactivity and biodegradability, it is important to synthesize mesoporous CSHs without using any surfactants. The Ca2+ cations on the pore-wall surface of the CSH can be effectively used for capturing the drug molecules, including acid group such as ibuprofen, aspirin, and amoxicillin. In addition, the combination of CaO in SiO2 structures can secure high bioactivity and clinical safety.30,31 It has been reported that a large specic surface area, a large pore volume, and a 3D hierarchical structure with an interconnected pore network are important factors for achieving a high DLC.32 Nevertheless, calcium silicate (CS) nanomaterials that satisfy these requirements are usually difficult to prepare. Only a few studies have been reported using the sonochemical method.33 The development of new approaches is necessary for synthesizing hierarchical CS nanostructures with high DLCs. Herein, we report a facile method for the poly(allylamine hydrochloride) (PAH)-assisted synthesis of mesoporous calcium silicate hydrates (denote as PAH-CS) with a large specic surface area and large pore volume. The mesoporous PAH-CS carriers exhibited remarkably enhanced DLCs (3.35 g IBU was loaded per gram of carrier) due to their large surface area (SBET ¼ 348.4 m2 g1), large pore volume (Vp ¼ 1.42 cm3 g1), and very high cationic charge (z ¼ +65.66 mV) compared to other CS materials. pH triggered protein-loading and releasing behaviors of the mesoporous PAH-CS were also investigated at various pH conditions. FITC-labeled bovine albumin (FITC-Albumin) was selected as a model protein and the PAH-CS showed high loading capacity with switchable releasing behavior under various pH (4.5 to 2.5) conditions (FITC-Albumin/PAH-CS, 1 g g1).

PAH-CS carriers. The structural features of the as-synthesized PAH-CS (TEOS : PAH : Ca2+ ¼ 1 : 15 : 65 (v/v)) composed 3D hierarchical networks, which were markedly different from the PAH-Si constructed by reacting the cationic polyelectrolyte PAH and TEOS (Fig. 1a and b). Initially, TEOS was employed as a silicon source and was rapidly hydrolyzed and reacted with the amine groups of PAH to form spherical PAH-Si with a relatively smooth surface (Fig. 1a). Subsequently, the PAH-Si spheres were converted into PAH-CS composed of spherically assembled nanoplates by introducing Ca2+ in basic media (pH 12.0) (Fig. 1b and 2). The PAH-CS with a relatively uniform size of 0.9 mm presented creased or folded surface morphologies and a low density at the particle edges (Fig. 2a–c). The surface charge of the PAH-CS was found to be +65.66 mV, which was remarkably increased compared to that of PAH-Si (26 mV), indicating the incorporation of Ca2+ (Fig. S1†). EDX analysis showed the presence of N (meaning PAH), Si (meaning CS), and Ca (meaning CS) peaks (Fig. 2d), conrming that PAH-CS was successfully synthesized from PAH-Si. PAH (pKa ¼ 8.7 in salt-

2.

(a–c) UHR-FESEM images and (d) corresponding EDX data for the PAH-CS carrier prepared by mixing TEOS and Ca2+ in the presence of PAH. (a and b) SE mode, (c) TE mode. The Al peak in the EDX data was from the sample holder.

Results and discussion

2.1. Characterization of hierarchical mesoporous PAH-CS Fig. 1 shows the procedure for synthesizing the PAH-CS carriers as well as the loading/releasing processes of drug on/from the

Fig. 1

Fig. 2

Schematic illustrations of the synthesis of the PAH-CS carrier and of the drug loading and release processes.

1002 | J. Mater. Chem. B, 2015, 3, 1001–1009


View Article Online


Paper

free solution) is one of the most popular polyelectrolytes and has previously been used for the drug delivery because of its biocompatibility.34,35 In an aqueous solution of PAH, water molecules are associated with the ammonium ions of PAH through van der Waals interactions. The addition of silicate species, such as TEOS, causes the displacement of the water molecules and establishes a network of properly organized silicate, leading to the formation of a silica-polymer framework.36 In our case, the ammonium ions of PAH can interact with the silicate ions through electrostatic interactions as well as through hydrogen bonding to form spherical PAH-Si. Subsequently, aer increasing the pH to 12 with NaOH and adding Ca2+, the PAH-Si spheres can be dissolved into silicate anions that further react with calcium ions (Ca2+) at the surface of the PAH-Si spheres to form calcium silicates (PAH-CS). In particular, when the pH was over 10, the positive charge of PAH in the PAH-Si structures decreased by up to 30% due to deprotonation of the PAH. The basic pH also induced an entangled conformation of PAH due to the decrease in repulsive forces. Thus, Ca2+ was easily able to approach PAH-Si and react with the silicate anions released from the PAH-Si, leading to the formation of a network structure of PAH-CS with a large cationic surface area. Although the size of PAH-Si was approximately 250–300 nm, the size of the nal products (PAH-CS) was increased four-fold (over 1 mm) aer the addition of Ca2+ at highly alkaline conditions (Fig. 1a and b). This increase is because of a large amount of Ca2+ aggregated with the silicate anions to form nanoplates (thickness 80 nm), which were the building blocks for the PAH-CS. When the amount of PAH was increased (TEOS : PAH : Ca2+ ¼ 1 : 30 : 65) (Fig. S2b and d†) or the amount of Ca2+ was decreased (TEOS : PAH : Ca2+ ¼ 1 : 15 : 32.5) (Fig. S2e and f†) compared to the standard conditions, an irregular size of PAH-CS in the range of 400–700 nm with a relatively smooth surface morphology was obtained. These results suggest that hierarchical 3D networks of PAH-CS can be formed using an optimized ratio of Ca2+ to PAH. The role of PAH was not only providing the template for the assembly of Ca2+, but also increasing the net cationic surface charge (almost double of CS, Fig. S3†) due to the coexistence of both the Ca2+ and NH3+ in mesoporous PAH-CS. Actually, the specic surface area (BET) of PAH-CS was increased around 1.8 fold higher compare to PAH-Si (BET ¼ 195.6 m2 g1). The high specic surface area and high charge density remarkably enhance the uptake of oppositely charged drug molecules (IBU or FITCAlbumin). PAH-CS carriers with various morphologies could be prepared by varying the reaction time (Fig. 3). In the beginning of the reaction (1 hour), the shape of the PAH-CS was irregular (Fig. 3a). Aer 2 hours, it became more spherical and relatively well-dened in structure (Fig. 3b). As the reaction time increased up to 3 hours, the surface morphology of the PAH-CS showed more compact structures composed of spherically assembled nanoplates that were thicker and more clearly distinguished compared to the other cases (1 h and 2 h) (Fig. 3c). Aer 6 hours, however, the surface morphology of the PAH-CS was blurred, and the thickness of the nanoplate building blocks of the PAH-CS was decreased (Fig. 3d).



Fig. 3 UHR-FE-SEM images of PAH-CS carriers prepared using different reaction times. The scale bars represent 500 nm. (a) 1 h, (b) 2 h, (c) 3 h, and (d) 6 h.

Therefore, a reaction time of 3 hours was chosen as the optimized condition for synthesizing the PAH-CS carriers. The specic surface area, pore volume, and pore size of the PAH-CS carriers were investigated by the Brunauer–Emmett– Teller (BET) and the Barrett–Joyner–Halenda (BJH) measurements. The nitrogen adsorption–desorption isotherm curve of the PAH-CS showed a type H3 with a hysteresis loop (according to the International Union of Pure and Applied Chemistry; IUPAC), deriving from plate-like-particle aggregates (hierarchical structure) with slit shaped pores, which is consistent with the SEM micrographs (Fig. 4a).37 Most of the pores were within the range of 4–40 nm (the inset in Fig. 4a). From the BJH measurements, the average pore size was determined to be 24.2 nm, which is larger than the typical pores in MCM-41 or CSH prepared in the absence of any swelling agent or calcination. The BET specic surface area and the BJH desorption cumulative pore volume (Vp) were 348.4 m2 g1 and 1.42 cm3 g1, respectively (Table 1). These values are considerably larger than those of previously reported CS materials without any polymer.33 The specic surface area and pore volume are closely related to the drug loading ability.

J. Mater. Chem. B, 2015, 3, 1001–1009 | 1003

View Article Online



Paper

(a) N2 adsorption–desorption isotherm and BJH-desorption pore size distribution curves of PAH-CS carriers loaded with different amounts of IBU. (b) FT-IR data verifying IBU loading onto the PAH-CS carriers. (c) Amount of IBU released over time. (d–f) SEM images after loading different amounts of IBU. (d) 1 mg mg1, (e) 2 mg mg1, and (f) 3.6 mg mg1 (IBU mg per carrier mg).

Fig. 4

Table 1 Summary of the PAH-CS-IBU systems. SBET is the BET surface area. The average pore sizes were calculated using the BJH method. V is the total pore volume. * The charges neutralized by IBU were determined by calculating the zeta potential differences before and after IBU loading. Here, 65.66 mV was considered as 100%

Drug delivery system

SBET (m2 g1)

ddc(average) (nm)

Vp (cm3 g1)

Zeta potential (mV)

*Charges neutralized by IBU (%)

*DLC (wt%)

Loading eff. (%)

PAH-CS PAH-CS-IBU1.0 PAH-CS-IBU2.0 PAH-CS-IBU3.6

348.4 121.6 42.98 10.79

24.2 16.54 10.91 4.2

1.42 0.49 0.18 0.031

65.66 46.73 27.84 2.41

— 28.83 62.17 96.32

— 0.93 1.86 3.35

— 93 93 94.56

2.2. Cytotoxicity evaluation of PAH-CS To evaluate the cytotoxicity of mesoporous PAH-CS materials, LDH (Lactate dehydrogenase) assay in supernatant medium was used in this study. The cytotoxic effect of PAH-CS on osteoblastlike MC3T3-E1 cells is shown in Fig. S4.† It can be seen that there were no cytotoxic effects on MC3T3-E1 cells for PAH-CS materials from a low concentration (5 mg mL1) to a high extract concentration (20 mg mL1) aer 24 h. In the case of higher concentration of PAH-CS (20 mg mL1), the calculated cell viability percentage (%) was over 98% even aer 3 days (data not shown). These results indicate that our PAH-CS materials are in the feasible levels. 2.3. Loading of model drug IBU “The Ca2+ cations on the pore-wall surface of the CSH can be effectively used for capturing drug molecules containing acidic groups such as ibuprofen, aspirin, and amoxicillin. Moreover,

1004 | J. Mater. Chem. B, 2015, 3, 1001–1009

the CaO in SiO2 structures can secure high bioactivity and clinical safety.30,31 Therefore, ibuprofen (IBU) was chosen as a model drug to test the loading and delivery efficiencies of the PAH-CS carriers.” The ibuprofen (IBU) drug loading capacity onto the PAH-CS carriers was monitored by BET surface area and zeta potential measurements (Fig. S3† and Table 1). The DLCs of the PAH-CS carriers are summarized in Table 1. The subscripts indicate the amount (mg) of IBU loaded per mg of PAH-CS carrier. The specic surface area of the PAH-CS carriers decreased as the amount of IBU loaded increased. For PAH-CSIBU3.6, the specic surface area of the PAH-CS carriers was greatly decreased to 10.79 m2 g1 aer IBU loading. The zeta potential values also decreased to 2.41 mV, indicating that 96.32% of the charge of bare PAH-CS carriers was neutralized. These results indicate that IBU was successfully loaded into the PAH-CS carriers. The calculated DLC (wt%) was 3.35 mg with a loading efficiency of 94.56%. No more signicant changes were observed aer using any higher amount of IBU (4 mg per mg of


View Article Online


Paper

the carrier). Therefore, the maximum DLC for IBU was determined to be 3.35 mg per mg of PAH-CS carrier, which is higher than the values reported in the studies (Table S1†).33,38 To further conrm interactions between the drug molecules and the carrier, FTIR measurements of the PAH-CS carriers were performed before and aer IBU loading (Fig. 4b). A peak for the PAH-CS carriers at 1503 cm1 was assigned to the stretching vibration of –NH2 bending in PAH, demonstrating the successful interaction of amino groups with the silica surface (green line). Acid–base reactions between the carboxylic acid groups of IBU and the amino groups of PAH took place during IBU loading onto the PAH-CS carriers. For PAH-CS-IBU, a blue shi of the Si–O stretching vibration from 948 to 1095 cm1 and a red shi of the C]O stretching vibration of IBU from 1763 to 1557 cm1 occurred, indicating chemical interactions between the –Si–O–Ca– groups of the PAH-CS carrier and the –COOH groups of the IBU drug molecules, respectively (red line).39 Fig. 4d–f shows the morphological changes of the PAH-CS carriers aer loading various amounts of IBU in hexane. The surface morphology of the PAH-CS carriers became relatively smooth as the amount of IBU increased. The UV-Vis absorption spectra of the IBU hexane solution before and aer loading IBU onto the PAH-CS carriers are shown in Fig. S5a.† The absorption spectra of IBU3.6 in hexane showed a characteristic peak at 264 nm.40 Aer reacting with the PAH-CS carrier for 24 h, the absorbance of the characteristic peak at 264 nm was decreased to 6% of the original maximum absorbance. The changes in morphology and the signicant decrease in absorbance indicate that the drug was loaded onto the PAH-CS carriers at a high DLC (Fig. 4d–f and S5a†). 2.4. In vitro release of IBU and formation of hydroxyapatite Because IBU molecules form good linkages with the PAH-CS carriers by electrostatic interactions between anions (–COO groups) and cations (Ca2+ or NH3+) under basic circumstances and these linkages are gradually broken under neutral conditions, the release prole of IBU was analyzed in neutral simulated body uid medium (SBF, pH 7.4). Fig. S5b–d† show timedependent UV-Vis spectra of SBF solution containing PAH-CS carriers loaded with various amounts of IBU. The absorbance at 264 nm increased as time increased, indicating that the concentration of IBU gradually increased due to the sustained release of IBU from the PAH-CS carriers. Kinetic release curves were plotted by monitoring the absorption peak at 264 nm over time (Fig. 4c). The results clearly indicate evident differences in the drug-release rates for the different drug delivery systems. For the PAH-CS-IBU1.0 and the PAH-CS-IBU2.0, 81.1% and 85.4% of the loaded IBU was released over a long period of 252 h and 228 h, respectively. These two cases showed sustained release of IBU. This could be explained by the formation of hydrogen bonds between –COOH groups of IBU and –NH2 groups of PAH as well as –OH groups of hydroxyapatite (HA; Ca10(PO4)6(OH)2) inside the pore-walls of the carrier, which holds back the release of the IBU. It could prolong the drug effect and encourage it as being a promising candidate of PAH-CS for bone implantable drug-delivery system. Moreover, PAH-CS-IBU3.6 initially



displayed a burst-release, releasing 80% of the loaded drug within 36 h, and the release was almost complete aer 72 h (93.3%). The rapid drug release for this group can be attributed to the IBU molecules being adsorbed onto the external surfaces and in the mesopores.41 This type of release prole could be useful when an immediate high dosage is required. Fig. S6† shows the representative SEM images of PAH-CS aer release of IBU3.6 in SBF. EDX data conrms the existence of P, O, and Ca peaks and the transformation of calcium silicate into hydroxyapatite (Ca10(PO4)6(OH)2). X-ray diffraction (XRD) measurements were made to identify changes in the chemical composition of the PAH-CS-IBU3.6 carriers in SBF solution (Fig. 5). Aer immersing the carriers in SBF solution for 72 h, a hydroxyapatite (HA) peak was clearly observed at 31.77 (blue line), and this peak was not observed in the PAH-CS-IBU3.6 before immersion in SBF (black line). This result indicates that the PAH-CS carrier was entirely transformed to HA, indicating good bioactivity and biodegradability of the PAH-CS carrier. Aer prolonged immersion of the PAHCS-IBU1.0 for 252 h, the intensities of the HA peaks increased, indicating increases in the HA contents (red line). The slow drug release of the PAH-CS-IBU2.0 and PAH-CS-IBU1.0 groups is associated with the formation of hydroxyapatite that gradually restrains the release of drug from the mesopore channels.41,42 Moreover, it is important to maintain stability of a drug during the loading and release processes. The shapes of the IBU absorption curves were essentially the same in all cases, indicating that the properties of IBU were not changed during the loading and release processes (Fig. S5†). 2.5. Loading of protein and in vitro pH-triggered release In addition, protein loading efficiency as well as pH-triggered releasing behavior of PAH-CS carriers were monitored under low pH conditions (4.5 to 2.5). The isoelectric point of albumin is pH 5.0. Albumin has a negative charge above pH 5.0 with an expanded structure, and in acidic pH conditions its charge is positive.43,44 Therefore, the negatively charged FITC-Albumin can be loaded onto the positively charged PAH-CS carrier using electrostatic interactions in PBS (pH ¼ 7.4). Therefore, another

Wide angle XRD patterns of PAH-CS-IBU before (black line) and after immersion in SBF for 252 h (red line) and 72 h (blue line). H ¼ hydroxyapatite.

Fig. 5

J. Mater. Chem. B, 2015, 3, 1001–1009 | 1005

View Article Online



model drug (protein), FITC-Albumin (M.W. ¼ 66.5 kD) labeled with uorescein isothiocyanate (FITC), was tested. The protein was successfully incorporated into the PAH-CS carrier at pH ¼ 7.4 in PBS buffer solution, and its pH-triggered controlled release was investigated. Fig. 6a–c show confocal laser scanning microscopy (CLSM) images of the PAH-CS carriers aer loading 0.5 and 1 mg of FITC-Albumin. For 1 mg of FITC-Albumin, the distribution of green uorescence observed within the PAH-CS carriers was more homogenous compared to the case of 0.5 mg of FITC-Albumin (Fig. 6a–c). The absorption spectra of FITCAlbumin showed a characteristic peak at 496 nm (black line, Fig. 6d). Aer adsorption of the FITC-Albumin onto the PAH-CS carriers for 24 h, the absorbance of the FITC-Albumin solution was decreased to 10% of the original maximum absorbance (red line, Fig. 6d). This signicant decrease in the absorbance indicates that 90% of the protein was loaded onto the PAH-CS carrier. The amount of FITC-Albumin loaded onto the PAH-CS carriers was calculated from the weight difference of the carrier before and aer FITC-Albumin loading (Fig. 6e). It was found that the maximum loading weight% was 88.6 for FITC-Albumin (1 mg mL1). The loading weight% was almost unchanged, even when a FITC-Albumin concentration greater than 1 mg mL1 was used (data not shown). By considering the neutralized zeta

Paper

potential values before (65.6 mV) and aer the (7.2 mV) loading of FITC-Albumin onto the PAH-CS carrier, it is apparent that most of the FITC-Albumin was successfully loaded (Fig. S1 and S7†). Because the FITC-Albumin is a bulk material (M.W. ¼ 66.5 kD) compared to a low molecular weight-drug (IBU), most of the FITC-Albumin can be attached on the surfaces rather than inside pore-walls of the carrier. Therefore, its loading density was lower and releasing rate was faster than that of the IBU. Acidic pH conditions that are lower than an isoelectric point of FITC-Albumin were used for proling the releasing patterns of FITC-Albumin from the PAH-CS carrier. In this low pH conditions, the net charge of the FITC-Albumin was changed to a positive. Positive charges of the PAH-CS carrier became enriched due to the protonation of PAH (NH3+). As the result, repulsion forces between the same charge groups of the FITCAlbumin and the PAH-CS carrier became superior, which drives their release from the carrier. The absorbance at 496 nm increased over time for all of the different pH conditions tested (4.5, 3.5, and 2.5) (Fig. 6f and S8†). There was a burst release at the early stage, followed by a sustained release stage. At pH 2.5, 50% of the protein was released aer 20 h, and the release was completed aer nearly 60 h. In contrast, at pH 4.5 it took 54 h to reach 50% release. The release rate for albumin was higher at

Fig. 6 (a–c) CLSM images and corresponding fluorescent intensity profiles for FITC-Albumin-loaded PAH-CS carriers in PBS (pH 7.4). (a and b) 1 mg L1 and (c) 0.5 mg L1 of FITC-Albumin was used. The inset of (a) shows a corresponding UHR-FESEM micrograph (the scale bar indicates 1 mm). (b and c) Top: CLSM images. Bottom: fluorescent intensity profiles. (d) UV-Vis spectra of FITC-Albumin before (black line) and after (red line) loading onto PAH-CS carriers. (e) The loading wt% of FITC-Albumin calculated from the weight differences of the PAH-CS carriers before and after loading FITC-Albumin. (f) pH-dependent release characteristics of FITC-Albumin-loaded PAH-CS carriers.

1006 | J. Mater. Chem. B, 2015, 3, 1001–1009


View Article Online


Paper


the low pH (¼2.5) condition. Presumably, weak interactions between the FITC-Albumin and the PAH-CS carrier lead to a faster release rate and a higher release amount.45 The electrostatic interactions between FITC-Albumin and the PAH-CS carrier decrease as the pH is decreased from pH 7.4 to 2.5 because of repulsive forces between the positively charged PAHCS carrier and the FITC-Albumin, which has a positive charge below its isoelectric point (pH 5.0). As a result, the repulsive force between FITC-Albumin and the PAH-CS increases proportionately at low pH conditions, leading to a faster drug release rate.

3.

Conclusions

We have developed a facile method for polymer-assisted controlled synthesis of mesoporous calcium silicate hydrates (PAH-CS) with a large specic surface area and pore volume. TEOS was rapidly hydrolyzed and reacted with the amine groups of PAH to form spherical PAH-embedded SiO2 NPs (PAH-Si). Subsequently, Ca2+ ions reacted with silicate anions produced during dissolution of the PAH-Si in basic media, leading to the formation of highly porous 3D networks of PAH-CS that were only synthesized under optimized reaction conditions. Because the mesoporous 3D network structure has a large surface area and a very high cationic charge from –NH3+ and Ca2+ groups, the highest DLC reported to date (3.35 g of IBU could be loaded per gram of carrier) was achieved in this work. In addition, the PAH-CS carrier with no toxicity was entirely transformed to hydroxyapatite aer drug release in SBF, suggesting that the PAH-CS carrier has excellent bioactivity and biodegradability. Furthermore, a model protein drug, FITC-Albumin, was loaded onto the PAH-CS carrier under neutral conditions, and its release behavior was controlled by varying the pH conditions. We believe that our approach could be useful for synthesizing CS drug carriers with a large specic surface area and pore volume, enabling the development of controlled drug delivery systems and other release applications.

4. Experimental 4.1. Materials Tetraethyl orthosilicate (TEOS, 99.99%), ibuprofen (IBU, 99%), albumin (uorescein isothiocyanate-conjugated, from bovine; FITC-Albumin), poly(allylamine hydrochloride) (PAH, M.W. ¼ 25 000), phosphate buffered saline (PBS, pH ¼ 7.4) and ammonium hydroxide (NH4OH, 28–30%) were purchased from Sigma-Aldrich. Granular calcium chloride (CaCl2, 90%) and sodium hydroxide (NaOH, 93%) were purchased from Showa Chemicals Co. LTD, Japan. All chemicals were used as received without further purication. Simulated body uid (SBF, pH ¼ 7.4) was prepared according to the following reference. Deionized (DI) water was obtained from a Millipore Simplicity 185 system.


4.2. Synthesis of PAH-CS (TEOS : PAH : Ca2+ ¼ 1 : 15 : 65 (v/v)): TEOS (0.2 mL) was added to 2 mL of ethanol with NH4OH (50 mL). Then, 3 mL of aqueous PAH solution (1 mg mL1) was added to the mixture and stirred for 1 h. Subsequently, 13 mL of CaCl2 solution (0.04 M) and 1 mL of NaOH (1.0 M) were slowly added into the solution of TEOS and PAH and continuously stirred for 3 h. The white product was collected by centrifugation, washed with DI water and ethanol, and dried at 50 C. The obtained product was denoted as PAH-CS. Similarly, the PAH-CS with various compositions such as (TEOS : PAH : Ca2+ ¼ 1 : 30 : 65) and (TEOS : PAH : Ca2+ ¼ 1 : 15 : 32.5) were prepared by the same procedure. 4.3. Cytotoxicity test of PAH-CS Cytotoxicity was assessed by LDH (Lactate dehydrogenase) assay in supernatant medium, using osteoblast-like MC3T3-E1 cell lines and culture medium (i.e. MEM a + 2 mM glutamine + 10% Foetal Bovine Serum, FBS) was purchased from Sigma-Aldrich. LDH cytotoxicity detection kit was obtained from Takara Bio, Inc., Tokyo, Japan. Prior to the cytotoxicity evaluation, the mesoporous CS-PAH materials were prepared in culture medium according to the standard protocol provided by the manufacturer (manual: MK401 v.0301 (pdf), Takara Bio, Inc., Tokyo, Japan). The cells were pretreated with the indicated doses (5–20 mg mL1) of PAH-CS. Aer incubating at 37 C, 5% CO2 for 24 h, the mixtures were centrifuged and the supernatants were collected. LDH activity was determined by measuring the absorbance at 490 nm using a microplate reader (Spectra Max M2; Molecular Devices, LLC, Sunnyvale, CA, USA). 4.4. IBU loading onto PAH-CS and in vitro release Each twenty milligrams of PAH-CS was separately added to 20 mL solutions of hexane containing three different concentrations of IBU (1.0 mg mL1, 2.0 mg mL1 and 3.6 mg mL1) at room temperature. Each mixture was immediately sealed and then stirred for 24 h at 200 rpm. The IBU-loaded PAH-CS was separated by centrifugation and washing, and the supernatants ware measured by UV-Vis spectroscopy to determine the drug loading capacities. Separately, the precipitates were washed with hexane and dried at 50 C. The product was denoted as PAH-CS-IBUx (x ¼ 1.0, 2.0 and 3.6). To evaluate the IBU release proles, PAH-CS-IBUx was redispersed in 50 mL of freshly prepared SBF (pH ¼ 7.4). While constantly stirring at 200 rpm, 2 mL aliquots were collected at specied time intervals to measure their UV-Vis absorbance spectra and replaced with an equal volume of fresh SBF. 4.5. FITC-Albumin loading onto PAH-CS and in vitro release Varying concentrations of FITC-Albumin (0.5, 0.75 and 1.0 mg mL1) were prepared in PBS buffer at pH ¼ 7.4. Twenty milligrams of PAH-CS was added to 20 mL of FITC-Albumin solution, and the mixture was continuously stirred for 12 h at room temperature. Aer collecting an aliquot aer certain time intervals, the samples were washed by centrifugation at 6000

J. Mater. Chem. B, 2015, 3, 1001–1009 | 1007

View Article Online



rpm for 5 minutes twice. The loading efficiency was determined from UV measurements of the supernatants. The precipitates ware collected and denoted as PAH-CS-(FITC/Albumin)x (x ¼ 0.5, 0.75 and 1.0). Aer drying with air, the loading weight percentage (wt%) of FITC-Albumin per mg of PAH-CS was calculated. Subsequently, these samples were used to evaluate the pH triggered release prole of loaded albumin. To monitor the release behavior, 15 mg of PAH-CS-(FITC/Albumin)1.0 was re-dispersed in 15 mL of PBS at 3 different pH conditions, i.e., pH ¼ 2.5, 3.5 and 4.5, and continuously stirred at room temperature. Aliquots were collected at specied time intervals, and their absorbance values were measured by UV-Vis spectroscopy. The pH of the PBS buffer solution was adjusted using 0.1 M HNO3.

4.6. Characterization The micrographic characterization and energy dispersive X-ray spectrometry (EDX) of the fabricated CS-PAH, before and aer IBU loading, were carried out by ultra-high-resolution eldemission scanning electron microscopy (FE-SEM) using a Hitachi S-5500 instrument. The BET surface areas and BJH pore-size distributions were measured using an accelerated surface area and porosimetry system (Micromeritics ASAP2010, USA). XRD patterns were obtained using a Rigaku X-ray diffractometer equipped with a Cu Ka source. Dynamic light scattering (DLS) studies and z-potential measurements were performed using a Malvern Nano-ZS Zetasizer at room temperature. UV-Vis absorption spectra were recorded using a UV-Vis-NIR spectrophotometer (Shimadzu UV-3600). FITCAlbumin loading was also conrmed using a Zeiss LSM 510 confocal laser scanning microscope (CLSM, Thornwood, NY).

Acknowledgements This research was supported by the Korea Basic Science Institute (KBSI), grant T34740, and the Basic Science Research Program of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2014003515).

References 1 Y. Cai, H. Pan, X. Xu, Q. Hu, L. Li and R. Tang, Chem. Mater., 2007, 19, 3081–3083. 2 I. I. Slowing, B. G. Trewyn, S. Giri and V. S. Y. Lin, Adv. Funct. Mater., 2007, 17, 1225–1236. 3 X. Kang, S. Huang, P. Yang, P. Ma, D. Yang and J. Lin, Dalton Trans., 2011, 1873–1879. 4 Y. Lu, Angew. Chem., Int. Ed., 2006, 45, 7664–7667. 5 M. Mehrali, S. F. S. Shirazi, S. Baradaran, M. Mehrali, H. S. C. Metselaar, N. A. B. Kadri and N. A. A. Osman, Ultrason. Sonochem., 2014, 21, 735–742. 6 J. Zhu, L. Liao, X. Bian, J. Kong, P. Yang and B. Liu, Small, 2012, 8, 2715–2720.

1008 | J. Mater. Chem. B, 2015, 3, 1001–1009

Paper

7 R. K. Singh, K. D. Patel, J.-J. Kim, T.-H. Kim, J.-H. Kim, U. S. Shin, E.-J. Lee, J. C. Knowles and H.-W. Kim, ACS Appl. Mater. Interfaces, 2014, 6, 2201–2208. 8 K. C.-W. Wu, Y. Yamauchi, C.-Y. Honga, Y.-H. Yang, Y.-H. Liang, T. Funatsu and M. Tsunoda, Chem. Commun., 2011, 47, 5232–5234. 9 J. Yang, J. Lee, J. Kang, K. Lee, J.-S. Suh, H.-G. Yoon, Y.-M. Huh and S. Haam, Langmuir, 2008, 24, 3417–3421. 10 R. L. Grant, C. Yao, D. Gabaldon and D. Acosta, Toxicology, 1992, 76, 153–176. 11 R. Venkatesan, A. Pichaimani, K. Hari, P. K. Balasubramanian, J. Kulandaivel and K. Premkumar, J. Mater. Chem. B, 2013, 1, 1010–1018. 12 S.-H. Hu, C.-H. Tsai, C.-F. Liao, D.-M. Liu and S.-Y. Chen, Langmuir, 2008, 24, 11811–11818. 13 H. Meng, M. Xue, T. Xia, Z. Ji, D. Y. Tarn, J. I. Zink and A. E. Nel, ACS Nano, 2011, 5, 413–4144. 14 B. G. D. Geest, N. N. Sanders, G. B. Sukhorukov, J. Demeester and S. C. D. Smedt, Chem. Soc. Rev., 2007, 36, 636–649. 15 C. S. Peyratout and L. D¨ ahne, Angew. Chem., Int. Ed. Engl., 2004, 43, 3762–3783. 16 C. Vilos, F. A. Morales, P. A. Solar, N. S. Herrera, F. D. Gonzalez-Nilo, D. A. Aguayo, H. L. Mendoza, J. Comer, M. L. Bravo, P. A. Gonzalez, S. Kato, M. A. Cuello, C. Alonso, E. J. Bravo, E. I. Bustamante, G. I. Owen and L. A. Velasquez, Biomaterials, 2013, 34, 4098–4108. 17 A. V. Murueva, E. I. Shishatskaya, A. M. Kuzmina, T. G. Volova and A. J. Sinskey, J. Mater. Sci.: Mater. Med., 2013, 24, 1905–1915. 18 S. K. Jain, A. M. Awasthi, N. K. Jain and G. P. Agrawal, J. Controlled Release, 2005, 107, 300–309. 19 H. Li and J. Chang, J. Controlled Release, 2005, 107, 463–473. 20 J. L. Vivero-Escoto, I. I. Slowing, C. W. Wu and V. S. Y. Lin, J. Am. Chem. Soc., 2009, 131, 3462–3463. 21 M. Miola, C. Vitale-Brovarone, C. Mattu and E. Vern´ e, J. Biomater. Appl., 2013, 28, 308–319. 22 B. P. Bastakoti, Y.-C. Hsu, S.-H. Liao, K. C.-W. I. Wu, M. Yusa, S.-I. Kenichi, K. Nakashima and Y. Yamauchi, Chem.–Asian J., 2013, 8, 1301–1305. 23 Y.-H. Yang, C.-H. Liu, Y.-H. Liang, F.-H. Lin and K. C.-W. Wu, J. Mater. Chem. B, 2013, 1, 2447–2450. 24 Z. Li, T. Wen, Y. Su, X. Wei, C. Heb and D. Wang, CrystEngComm, 2014, 16, 4202–4209. 25 K. C.-W. Wu, Y.-H. Yang, Y.-H. Liang, H.-Y. Chen, E. Sung, Y. Yamauchi and F.-H. Lin, Curr. Nanosci., 2011, 7, 926–931. 26 J. Wu and Y. J. Zhu, Mater. Lett., 2009, 63, 761–763. 27 L. M. Rodriguez-Lorenzo, R. Garcia-Carrodeguas, M. A. Rodriguez, J. Aza, S. D. Jimenez, A. Lopez-Bravo, M. Fernandez and J. S. Roman, J. Biomed. Mater. Res., Part A, 2009, 88, 53–64. 28 S. Madieh, M. Simone, W. Wilson, D. Mehra and L. Augsburger, J. Pharm. Sci., 2007, 96, 851–863. 29 M. Vallet-Reg´ı, A. R´ amila, R. P. Del Real and J. P´ erezPariente, J. Chem. Mater., 2001, 13, 308–311. 30 L. L. Hench, J. Am. Ceram. Soc., 1998, 81, 1705–1728.


View Article Online


Paper

31 P. Saravanapavan, J. R. Jones, R. S. Pryce and L. L. Hench, J. Biomed. Mater. Res., Part A, 2003, 66, 110–119. 32 M. Vallet-Reg´ı, F. Balas and D. Arcos, Angew. Chem., Int. Ed., 2007, 46, 7548–7558. 33 J. Wu, Y.-J. Zhu, S.-W. Cao and F. Chen, Adv. Mater., 2010, 22, 749–753. 34 A. I. Petrov, A. A. Antipov and G. B. Sukhorukov, Macromolecules, 2003, 36, 10079–10086. 35 J. Choi and M. F. Rubner, Macromolecules, 2005, 38, 116–124. 36 S. L. Burkett and M. E. Davis, J. Phys. Chem., 1994, 98, 4647– 4653. 37 K. S. W. Sing, D. H. Everett, R. A. W. Haul, L. Moscou, R. A. Pierotti, J. Rouquerol and T. Siemieniewska, Pure Appl. Chem., 1985, 57, 603–619. 38 Y. F. Zhu, J. L. Shi, W. H. Shen, X. P. Dong, J. W. Feng, M. L. Ruan and Y. S. Li, Angew. Chem., Int. Ed., 2005, 44, 5083–5087.



39 X. Guo, J. Wu, Y.-M. Yiu, Y. Hu, Y.-J. Zhu and T.-K. Sham, Phys. Chem. Chem. Phys., 2013, 15, 15033–15040. 40 Y. F. Zhu, J. L. Shi, Y. S. Li, H. R. Chen, W. H. Shen and X. P. Dong, Microporous Mesoporous Mater., 2005, 85, 75–81. 41 X. Li, J. L. Shi, Y. F. Zhu, W. H. Shen, H. Li, J. Liang and J. H. Gao, J. Biomed. Mater. Res., Part B, 2007, 83, 431–439. 42 K. L. Lin, J. Chang and R. M. Cheng, Acta Biomater., 2007, 3, 271–276. 43 Q. Shi, Y. Zhou and Y. Sun, Biotechnol. Prog., 2005, 21, 516– 523. 44 N. Shamim, L. Hong, K. Hidajat and M. S. Uddin, J. Colloid Interface Sci., 2006, 304, 1–8. 45 Q. Zhao and B. Li, Nanomed. Nanotech. Biol. Med., 2008, 4, 302–310.

J. Mater. Chem. B, 2015, 3, 1001–1009 | 1009