pH-Responsive Drug Delivery System Based on Coordination ...

ARTICLE pubs.acs.org/JPCC

pH-Responsive Drug Delivery System Based on Coordination Bonding in a Mesostructured Surfactant/Silica Hybrid Haoquan Zheng, Chuanbo Gao,* Baowei Peng, Mouhai Shu, and Shunai Che* School of Chemistry and Chemical Engineering, State Key Laboratory of Metal Matrix Composites, Shanghai Jiao Tong University, Key Laboratory for Thin Film and Microfabrication Technology of the Ministry of Education, 800 Dongchuan Road, Shanghai 200240, China

bS Supporting Information ABSTRACT: pH-responsive drug delivery systems have attracted great interest because of their potential use in antitumor therapies. Herein, we report a facile one-pot fabrication of a “hostmetaldrug” coordination-bonding system in a mesostructured surfactant/silica hybrid for the pH-responsive drug delivery purpose. The mesostructure has been synthesized by self-assembly of nontoxic and biocompatible F127 Pluronic nonionic surfactant and silica source through a real liquid crystal templating route, in which F127 act as host molecules. Metal ions such as Zn, Cu, and Fe and drugs have been introduced simultaneously into the mesostructure synthesis systems, to form F127metaldrug coordination-bonding architecture. The cleavage of the coordination bonds that are sensitive to variations in external pH gives rise to the release of the drug under weakly acidic conditions (pH 5.06.0). To assist in the release of drugs without significant binding capabilities, a vector has been designed to endow coordinately inert drugs with pH-responsive properties. Furthermore, the pH responsibility has been confirmed by cell assay and in vivo tumor test, opening up new opportunities for the potential application as implants in antitumor therapies.

1. INTRODUCTION Mesoporous silica has been well recognized as a potential carrier for drug delivery systems due to the big mesopore size, high surface area, large pore volume, and nontoxic nature of the amorphous silica framework,113 which is a good candidate for hosting “smart” delivery systems that release the encapsulated drugs in response to external stimuli, such as pH,1418 temperature,1921 light irradiation,2224 and chemicals.2527 pH-sensitive drug release systems based on polymeric micelles, nanogels, and inorganic materials have recently been developed to target the acidic extracellular pH environment of solid tumors. Because the acidic tumor microenvironment is the most common in solid tumors, the pH-targeting approach is regarded as a more general strategy than many other targeting approaches.2841 On the basis of the mesoporous silica carrier, we have developed a novel pH-responsive mesoporous silica drug delivery system by constructing the coordination bonding in mesopores, in which guest drug molecules were incorporated into the extracted mesopores by coordination bonding of a “hostmetalguest” architecture, whereas host, metal, and guest represent functional groups on the mesopore surface, metal ions, and drug molecules, respectively, which can be potentially utilized in antitumor applications.42 The formation and cleavage of the coordination bonds are sensitive to variations in external pH, r 2011 American Chemical Society

which gives rise to the effective loading and release of the drug at a designated pH. However, in this system, the installation of a drug delivery system in the mesopores of a silica carrier commonly involves complicated stepwise procedures. Typically, the construction of a coordination-bonding system in mesopores for pH-responsive drug delivery was achieved through synthesis of functional mesoporous silica, removal of surfactant, and adsorption of metal ions followed by adsorption of drugs. The complexity in preparation often leads to a great deal of cost in large-scale productions. Herein, we report a facile one-pot fabrication of a hostmetal guest coordination-bonding system in a mesostructured surfactant/silica hybrid for the pH-responsive drug delivery system, which has been considered a solution to this problem.43,44 The mesostructured surfactant/silica hybrids were synthesized by a real liquid crystal templating route. The methanol formed by the hydrolysis of tetramethyl orthosilicate (TMOS) was rapidly removed under gentle vacuum. It was possible to regenerate the liquid crystalline properties associated with Pluronic F127 (EO106PO70EO106)/water systems. A material that had a bodycentered cubic Im3m structure was obtained by the condensation Received: November 12, 2010 Revised: March 6, 2011 Published: March 28, 2011 7230

dx.doi.org/10.1021/jp110808f | J. Phys. Chem. C 2011, 115, 7230–7237

The Journal of Physical Chemistry C

Figure 1. One-pot construction of a pH-responsive system in a mesostructured surfactant/silica hybrid, which releases the encapsulated drugs in response to a decrease in external pH. Metal ions represent Cu2þ, Zn2þ, and Fe3þ. Drug molecules represent MX, DOX, and R8FITC.

of the silicate species. F127 was employed as the template for mesochannels because of its lack of significant cytotoxity;45 on the other hand, the EO (or PO) moiety of the F127 is coordinately active and can serve as the binding site (host) for the metal ions,46,47 while drugs with coordination-bonding capabilities act as guest binding to the unsaturated metal ions. Therefore, a “F127metaldrug” coordination-bonding system could be readily established, accompanied by the formation of the mesochannel system, by a one-pot incorporation of a metal salt and the drugs into the F127-templating system of the mesostructured silica (Figure 1). This hostmetalguest coordination-bonding system is stable under physiological pH conditions and readily releases the encapsulated drugs in response to a reduction in pH due to a breakdown of both or either of the F127metal and metaldrug coordination bonds. The nontoxic, biocompatible, and metabolizable microelements/metal cations in organisms, such as copper, iron, and zinc, have been utilized for coordination combining host and guest molecules. Anticancer drugs mitoxantrone (MX) and doxorubicin (DOX) with coordination-bonding capable NH and OH groups have been chosen as a model guest molecule. As another guest molecule, a vector that possesses abundant binding sites could be employed to conjugate to the guest molecule that is weak in binding affinity and endow it with the capability to be charged into the carrier and then released in response to external pH stimulus. A model drug, fluorescein isothiocyanate (FITC), was conjugated to octaarginine (R8), which is capable of coordinating to metal ions for the construction of the coordinationbonding system, in addition can easily penetrate biomembranes and translocate into living cells.48 Through these routes, the onepot-fabricated pH-responsive system becomes versatile and designable, as the pH-responsive property of any guest molecule could be theoretically realized.

2. MATERIALS AND METHODS 2.1. Materials. Chemicals, including tetramethyl orthosilicate (TMOS, J&K chemical), Cu(ClO4)2 3 6H2O (Sinopharm), ZnCl2 (Sinopharm), CuCl2 (Sinopharm), FeCl3 (Sinopharm), HCl (Sinopharm), Pluronic F127 (EO106PO70EO106, BASF), MX (FINC), DOX (Yinghuan Chempharm), fluorouracil (FU, TCI), and SPCA-1 cancer cells (Sigma-Aldrich), were used as purchased.

ARTICLE

R8FITC was prepared by fmoc (9-fluorenylmethyloxycarbonyl)-solid-phase peptide synthesis on a Rink amide resin, where γ-aminobutyric acid was employed as a spacer to connect FITC with the R8 segment. 2.2. Synthesis of the Mesostructured Surfactant/Silica Hybrid Incorporated with the ph-Responsive System. F127ZnR8FITC Architecture. In a typical synthesis, 0.12 g of F127 was dissolved in 0.3 g of TMOS followed by addition of 0.15 g of a water solution containing 0.05 M HCl and 0.33 M ZnCl2. The reaction was kept at 40 °C for 1 min, and then 10 mg of FITCR8 was added, giving rise to a clear solution. Methanol produced by hydrolysis of TMOS was removed by vacuum, and a solid monolith was formed after aging at 40 °C for 5 days. The monolith was ground into powders, which were afterward sieved to get mesoporous silica particles with a particular range of size. For example, to get the particles having a diameter of 5076 μm, the powders were sieved using #200 and #300 sieves. The typical loading amount of FITCR8 is 33.3 mg/g, which could be tuned by the additional amount of the FITCR8 in the synthesis. All F127metaldrug architectures were synthesized through the synthesis procedure above. MX, DOX, and FU were dissolved in methanol and added after the dissolution of surfactant. The loading amount of MX, DOX, and FU can be easily controlled by changing the concentration of their methanol solution. Compared with other methods, we used a one-pot fabrication of F127metaldrug to prevent the complicated stepwise procedures and great deal of cost in large-scale productions. The synthesis procedure of the mesostructured surfactant/silica hybrid was mild without causing decomposition of the drugs, in which the synthesis and aging temperature was 40 °C, similar to normal body temperature. The low acidic condition caused by the reduction of the acid during the condensation process and the removal of HCl in the synthesis solution by the rapid vacuum condition resulted in the formation of the F127metaldrug coordination-bonding architecture. 2.3. UVVis Spectrophotometric Analysis of Formation/ Cleavage of Coordination Bonds in Solutions. UVvis spectrophotometric analysis of formation/cleavage of coordination bonds in solutions: The formation/cleavage of the coordination bond between metal ions and guest molecules can be characterized by UVvis spectra. In a typical example, a mixture solution of 0.1 mM of Cu(ClO4)2 3 6H2O and 0.2 mM of R8 was prepared, and the pH was controlled by NaOH or HClO4 before being diluted to volume. The final solution was spectrophotometrically analyzed in the UVvis region after reaching a stable state. UV absorbance was measured at wavelengths of 250, 252, and 715 nm, respectively, when F127, R8, and MX were used as the ligand. 2.4. Release of Drugs from the Mesostructured Surfactant/Silica Hybrid in Solution. In a typical release experiment, 6.0 mg of the mesostructured surfactant/silica hybrid (diameter: 5076 μm) was suspended by vibration in 20 mL of phosphate buffer solution (PBS) with various pH values, at 37 °C. When sampling, the release system was centrifuged, after which 1.0 mL of supernatant solution was withdrawn and replaced by the same amount of fresh medium. The released amount of guest molecules was measured by UVvis spectrophotometer or fluorescence spectrophotometer. 2.5. In Vitro Cell Assay and in Vivo Release in Tumors. For in vitro release study, SPCA-1 cancer cells were seeded into a 35 mm 10 mm Petri dish (Corning) at the concentration of 2 105 cells/mL in RPMI-1640 plus 10% FBS culture medium in 7231

dx.doi.org/10.1021/jp110808f |J. Phys. Chem. C 2011, 115, 7230–7237

The Journal of Physical Chemistry C 2 mL. Cells were allowed to adhere to the Petri dish for 24 h. Before the particles were added, the culture media were replaced with pH 7.4 or pH 6.2 RPMI-1640 plus 10% FBS with 1% penicillin and streptomycin. To each Petri dish, 35 μL of the mesostructured hybrid particles (0.7 mg, loading of FITCR8: 71.1 mg/g) was added. Nanoparticles were incubated with cells for 16 h. After incubation, cells were washed with PBS five times and examined under a confocal microscope (Zeiss, LSM-510). For in vivo release study, 100 μL of the mesostructured particles (200 μg, loading of FITCR8: 33.3 mg/g) were injected into xenograft tumors no larger than 2 cm3 in size, which were grafted subcutaneously in nude mice. At the same time, 100 μL of the particles was injected into the tibialis anterior (TA) muscles. After 16 h, mice were terminated and the tumors and TAs were extracted. Frozen sections were made out of the tumors and TAs and viewed under fluorescent microscopy. 2.6. Instrumentation. Powder Xray diffraction (XRD) patterns were recorded on a Rigaku X-ray diffractometer D/ MAX-2200/PC equipped with Cu KR radiation (40 kV, 20 mA) at a rate of 0.5°/min over the range of 0.84° (2θ). Highresolution transmission electron microscopy (HRTEM) images were taken with a JEOL JEM-2100 microscope operating at 200 kV. The nitrogen adsorption/desorption isotherms were measured at 196 °C with a Quantachrome Nova 4200E porosimeter. The surface area and pore size were calculated by the BrunauerEmmettTeller (BET) method and the BarrettJoynerHalenda (BJH) method, respectively. The pore volume was calculated by the adsorbed amount at relative pressure of ∼0.99. Thermogravimetric analyses (TGA) of the materials were recorded on a Perkin-Elmer thermal gravimetric analyzer (TGA-7) from ambient temperature to 800 °C at the heating rate of 10 °C/min. The Zn element in the monoliths was measured by using an inductively coupled plasma (ICP) emission analysis apparatus (IRIS advantage 1000). Solid-state UV spectra were taken on a Shimadzu UV-2450 spectropolarimeter. The concentration of FITC-R8 in PBS of different pH values was measured by a UNICO UV-4802 UVvis double-beam spectrophotometer. The fluorescent microscopy images of cells and tissue sections were taken by a Zeiss LSM-510 confocal microscope.

3. RESULTS AND DISCUSSION 3.1. UVVis Spectrophotometric Analysis of Formation/ Cleavage of Coordination Bonds in Solutions. The formation

and cleavage of the coordination bond between metal ions and surfactant or metal ions and guest molecules can be tested by UVvis spectra. In a typical example, Cu(ClO4)2 3 6H2O, F127, and R8 or MX were chosen as metal source, host, and guest molecules, respectively, to detect the pH sensitivity of the coordination bonds formed by Cu2þ and the ligands. The sharply enhanced UV absorbance of the combined solution with increasing pH value indicated the formation of coordination bonds between the ligands and metal in response to pH elevation due to competitive bonding of ligand molecules with metal ions and protons (Figure 2 and Supplementary Figure S1). The onsets of the coordination bonds of CuMX, CuR8, and CuF127 are in increasing sequence of pH 4.55.5, 5.57.0, and 6.57.5, respectively, indicating their bonding strength decreasing order. Therefore, the responsive pH onset could be well tuned by controlling the strength of the coordination bonding via proper combination of guest molecules. Nevertheless, for the Znguest

ARTICLE

Figure 2. UV absorbance of water solutions comprised from Cu(ClO4)2 3 6H2O with F127, R8, and MX under different pH conditions, respectively.

molecule system, the formation of the coordination bond cannot be detected by UV absorbance, because the absorbance of Zn ion is out of the range of UVvis spectra. The absorbance of the solution of Fe3þ was complex and significantly changed by the variation of pH value. Therefore, it was difficult to detect the formation of the coordination bond. 3.2. Characterization and pHResponsive Release of Anticancer Drugs of Different Coordination-Bonding Systems. The symmetry of the F127metalMX coordinationbonding mesostructured hybrid monolith has been clearly determined as body-centered cubic Im3m from TEM images in combination with its XRD patterns (Figure 3). The nitrogen adsorption/desorption isotherm of the calcined sample is type IV with an obvious hysteresis loop of H2 type, indicating the inkbottle geometry of the mesopores. Mitoxantrone (MX) as a model anticancer drug was used in this pH-responsive drug delivery system via a F127metalMX architecture in mesostructured hybrid. The Cu (2p3/2), Zn (2p3/2), and Fe (2p3/2) XPS spectra of F127CuMX, F127ZnMX, and F127FeMX, respectively, have been shown in Figure S2. The XPS peak of Cu (2p3/2), Zn (2p3/2), and Fe (2p3/2) at low binding energies of 933.8, 1023.3, and 712.6 eV, respectively, reveals the formation of the coordinationbonding architectures and a large coordination number of metal ions (Figure S2). Figure 4 shows the release of MX via different hostmetalguest architectures by utilizing different metal ions in the mesostructured hybrid. It demonstrates that the system is pH stimulus responsive and the responsiveness offers great potential application of this system to pH variations and contributes to its designability. As shown in Figure 4 II, the responsive pH onset was 6.0 when Zn2þ was chosen as the coordinating center, while the release percentages were about 33% and 62% at PBS pH 6.0 and 5.0, respectively, and only about 8% at PBS pH 7.4. The solid-state UVvis spectra of F127 metalMX mesostructured hybrids at different stages with Zn2þ as the coordinating center, which are as-synthesized sample and the sample after treatment at PBS pH 5.07.0 for 12 h, are shown in Figure 5II. The band at 666 nm belonging to the complex of Zn2þMX disappeared with a concurrent emergence of a new band at 646 nm due to a free MX moiety at pH PBS 6.0 and pH PBS 5.0, indicating the cleavage of the coordination bond between Zn2þ and MX (Figure 5II, sample c and d). 7232



ARTICLE

Figure 3. XRD patterns (I, left), N2 adsorption/desorption isotherms (II), and TEM images (III) of the F127metalMX coordination-bonding mesostructured hybrid. The metal ions are (a) Cu2þ, (b) Zn2þ, and (c) Fe3þ, respectively.

Figure 4. pH-responsive release of MX from different F127metalMX coordination-bonding systems. The metal ions are (I) Cu2þ, (II) Zn2þ, and (III) Fe3þ, respectively.

Figure 5. Solid-state UVvis adsorption spectra of different F127metalMX coordination-bonding mesostructured hybrids at different stages: (a) as-synthesized sample; (b) the sample in PBS pH 7.0 for 12 h; (c) the sample in PBS pH 6.0 for 12 h; (d) the sample in PBS pH 5.0 for 12 h. The metal ions are (I, left) Cu2þ, (II, center) Zn2þ, and (III, right) Fe3þ, respectively.

The UVvis spectra of the supernatant solution after release at PBS pH 5.0 is shown in Supplementary Figure S3. Meanwhile, there is no red shift observed in the release at pH 5.0 compared with the free MX, indicating that free MX instead of Zn2þMX

exists in the releasing solution (Figure S3). The Zn contents in mesostructured hybrid at various pH values have been calculated by ICP analysis results. The metal ions would flow out from the mesostructured hybrid as the free metal ions or metalMX, 7233


The Journal of Physical Chemistry C when the cleavage of coordination bonds between F127 and metal ions happens. The Zn amount remains relatively stable after treatment in PBS pH 7.0, while significant losses of the Zn amounts could be observed under acidic conditions (from 181 μmol/g at pH 7.0 to 72 and 39 μmol/g at pH 6.0 and pH 5.0, respectively). Therefore, the releases of MX from the F127ZnMX architecture at pH 6.0 and 5.0 are caused by the cleavage of both sides of the F127ZnMX complex. The responsive pH onset was 5.0 when Cu2þ was chosen as the coordinating center, while the release percentages were about 35% at PBS pH 5.0 and below 10% at PBS pH 6.0 and 7.0 (Figure 4 I). The solid-state UVvis spectra of mesostructured hybrids at different stages with Cu2þ as the coordinating center are shown in Figure 5 I. The band at 635 nm belonging to the complex of Cu2þMX shows no significant blue shift, indicating the stability of the complex of Cu2þMX even at PBS pH 5.0. Thus, the release of MX at PBS pH 5.0 is mostly caused by the cleavage of the coordination bond between F127 and Cu2þ. The Cu amounts in mesostructured hybrid are 203, 187, and 78 μmol/g after treatment in PBS pH 7.0, 6.0, and 5.0, respectively. The decrease of Cu amounts after treatment in PBS pH 5.0 could be attributed to cleavage of the coordination bonds between F127 and Cu ions (Table S1). The UVvis spectra of the supernatant solution of F127Cu2þMX architecture after release at pH 5.0 show a clear red shift of the absorbance band from 608 to 615 nm, indicating that the state of MX in the released supernatant solution was Cu2þMX complex (Supplementary Figure S3). As shown in Figure 4 III, when Fe3þ was chosen as the coordinating center, the responsive pH onset was dramatically reduced. No significant release of MX was observed even at PBS pH 5.0, due to the high stability of the F127Fe3þMX architecture. The solid-state UVvis spectra of mesostructured hybrids at different stages with Fe3þ as the coordinating center are shown in Figure 5 III. The band belonging to the complex of Fe3þMX shows no significant change, indicating the stability of the complex of Fe3þMX even at PBS pH 5.0 (Figure 5 III). Furthermore, the Fe amounts in mesostructured hybrid remain constant after treatment in PBS pH 7.0, 6.0, and 5.0. Taken together, both sides of the F127Fe3þMX architecture are stable at PBS pH 7.0, 6.0, and 5.0. This responsive pH onset of a pH-responsive system could be well tuned by choosing the type of the metal ions in the construction of the pH-responsive system. 3.3. Vector-Aided pH-Responsive Release of Drugs. Zn2þ is found in nearly 100 specific enzymes, serves as structural ions in transcription factors, and is stored and transferred in metallothioneins. So Zn2þ, also an abundant trace element and essential for organisms, was employed as the typical coordination binder.49 The in vitro release of the FITCR8 from the F127 ZnFITCR8 coordination-bonding system in PBS of various pH values has been investigated (Figure 6). Under the physiological condition (pH 7.4), the release is slow, and only 7.9% of FITCR8 has been detected in PBS within 12 h. However, significant release of FITCR8 has been observed when the pH of the medium was decreased (pH 6.8, 6.5, and 6.2), indicating the high sensitivity of the drug delivery system to pH stimulus. The release percentages of the FITCR8 were 7.9, 48, 67, and 81% within 12 h and 20, 67, 81, and 83% within 22 h, at pH 7.4, 6.8, 6.5, and 6.2, respectively. The release of FITCR8 was in a sustained manner as a result of the diffusion-limiting function of

ARTICLE

Figure 6. In vitro release of the FITCR8 from the F127Zn FITCR8 coordination-bonding mesostructured hybrid in PBS of pH 7.4, 6.8, 6.5, and 6.2, respectively.

the pore-filling surfactant F127 in the surfactant/silica hybrid. The lower the pH the mesostructured hybrid encountered, the faster the drug was released, due to the weak affinity of the Zn2þ ions to both FITCR8 and F127 under low pH conditions. Control release experiments have been further conducted to understand the mechanism of the pH-responsive drug release from the mesostructured surfactant/silica hybrid. In our strategy, the Zn2þ ion plays the central role in the construction of the pHresponsive system. To confirm this, a control experiment was conducted without using any Zn2þ ion in the synthesis of the mesostructured hybrid. The in vitro drug release under the physiological condition (pH 7.4) indicates that the FITCR8 was sustainedly released within 12 h, which is much different from the release shown in Figure 6 (Supplementary Figure S4). When coordinately inactive acyltrimethylammonium surfactant was used as the template for the mesopores in the one-pot fabrication of the drug delivery system, no pH responsibility was detected in vitro release, and the encapsulated FITCR8 was released in the first few hours in PBS at pH 6.27.4 (Supplementary Figure S5), showing typical “burst effect” of drug release. Moreover, solely FITC without R8 conjugating could not be released in a stimulus-responsive manner from the mesostructured hybrid by varying the pH of the media, due to a lack of the coordination binding sites in the drug molecules (Supplementary Figure S6). These results emphasize the role of the Zn2þ, the coordinately active EO (or PO) moieties of the Pluronic surfactant, and the R8 moiety of the drug in the construction of the hostmetalguest pH-responsive system. 3.4. The Stability of Mesostructured Hybrid. XRD patterns of the untreated F127Zn2þFITCR8 coordination-bonding mesostructured hybrid and those after being treated in PBS are shown in Figure 7 I. Although only one reflection peak is detected for either as-synthesized or calcined hybrids, posing difficulty in determining the structure, the sharp X-ray reflection suggests that the material possesses highly ordered mesostructure. After being treated in PBS for 16 h, the mesostructure remained undamaged, as suggested by the intact XRD patterns, indicating high stability of the mesostructured hybrid. High crystallinity with some disordered mesophase and amorphous impurity were coexistent in the mesostructures (Figure 7 III). The single peak with highest intensity can be indexed as a 110 reflection of the Im3m phase. The highly ordered 7234



ARTICLE

Figure 7. XRD patterns (I), N2 adsorption/desorption isotherms (II), and TEM images (III) of the F127Zn2þFITCR8 coordination-bonding mesostructured hybrid. The samples are (a) untreated; (b) treated in PBS of pH 7.4 for 16 h; (c) treated in PBS of pH 6.2 for 16 h.

Table 1. Porous and Compositional Properties of the F127Zn2þFITCR8 Coordination-Bonding Mesostructured Hybrid during the pH-Responsive Release of FITCR8 surface areaa (m2/g) 0h

pore volumea (mm3/g)

Zn amountb (μmol/g)

F127 amountc (wt %)

0 (429)

0 (330)

176

52.3

pH 7.4, 1 h

56

52

111

50.9

pH 7.4, 6 h

93

58

57

55.7

107 (406)

97 (346)

43

47.6

30

37

32

48.7

46 62 (395)

65 78 (343)

14 6

51.0 45.5

pH 7.4, 16 h pH 6.2, 1 h pH 6.2, 6 h pH 6.2, 16 h a

Calculated from N2 adsorption/desorption data; in parentheses, data for calcined samples. b Calculated from ICP. c Calculated from TGA.

mesostructure could still be observed by HRTEM without significant deterioration after treatment in PBS for 16 h, once again confirming the high stability of the mesostructured silica materials. The nitrogen adsorption/desorption isotherms of the calcined samples are type IV with an obvious hysteresis loop of H2 type (Figure 7 II). The capillary condensations within the relative pressure of 0.40.8 suggest a uniform pore size distribution. For the as-synthesized sample, the diameters of the cavity and the window connecting the cavities are 4.3 and 3.4 nm, respectively, derived from the pore size distributions calculated by the adsorption and desorption branches of the isotherm, respectively

(Supplementary Figure S7). The surface area and the pore volume are 429 m2/g and 0.33 cm3/g, respectively, confirming the high mesoporosity of the material. Table 1 summarizes the porous and compositional properties of the mesostructured hybrid during drug release. The BET surface area of the mesoporous materials after calcination decreased from 429 to 406 and 395 m2/g after being treated in PBS pH 7.4 and 6.2 for 16 h, respectively, indicating a slight damage of the mesostructure, with the most of the porosity remaining intact. The amount of Zn2þ and F127 in the as-synthesized mesostructured hybrid has been calculated as 0.18 mmol/g and 52.3% from the ICP and the TGA results, which are close to their theoretical values, 0.21 mmol/g and 50%, respectively. After being treated in PBS in pH 7.4 and 6.2, the mesostructured hybrid has a decreased loading amount of the Zn2þ ions to different degrees, and the lower pH caused much dramatic leaching of the metal ions out of the hybrid. In this process, the amount of F127 in the hybrid kept almost constant, consistent with the nonporous properties of the mesostructured hybrids indicated by the nitrogen adsorption/ desorption isotherms before calcination (Figure 7 II). During the release of the FITCR8 from the F127Zn2þ FITCR8 coordination-bonding mesostructured hybrid, the structure of the hybrid stayed stable, which has been proven by XRD, TEM, and N2 adsorption/desorption results (Figure 7, Table 1, and Supplementary Figure S8, S9). Moreover, the surfactant F127 has been mostly retained in the mesopores during the pH-responsive drug release, indicated by the similar weight loss from the TGA results during the whole procedure 7235



Figure 8. Cell assay (ab) and in vivo tumor test (cd) of the pHresponsive drug delivery system in the F127ZnFITCR8 coordination-bonding mesostructured hybrid. (ab) Merged bright-field and fluorescent confocal micrographs of the SPCA-1 cells after incubation in RPMI-1640 under different pH values, (a) pH 7.4 and (b) pH 6.2, with the hybrid for 16 h. (cd) Fluorescent micrographs of the typical skeletal muscle (c) and solid tumor (d) sections of mice, sampled 16 h after the tumor and skeletal muscle were injected with the hybrid.

(Table 1 and Supplementary Figure S10). The presence of the pore-filling surfactant in the mesostructured hybrid was further proven by the small pore volume of the mesostructured hybrid after PBS treatment. The structural and compositional stability of the mesostructured surfactant/silica hybrid ensures the safe storage of the drug delivery system and excludes the possibilities of the drug release caused by the disintegration of the mesostructured hybrid. When the hybrid was immersed in PBS of pH 6.2, the amount of Zn2þ ions in it decreased dramatically (Table 1), confirming that the cleavage of the F127Zn coordination bonds accounts for the significant release of model drug in the form of FITCR8 or ZnFITCR8 composites from the hybrid under low pH conditions. On the other hand, as can be inferred from Table 1, at pH 7.4, the diffusion kinetics of the Zn2þ ions was much slower and the hybrid maintained a relatively high loading amount of Zn2þ, which accounts for the stability of the F127ZnR8 coordination-bonding system in the hybrid and avoids significant release of the drug under physiological conditions. 3.5. Cell Assay and in Vivo Tumor Test. To further verify the pH responsiveness of the system and the feasibility of its use in antitumor application, cell assay and tumor test have been conducted, and the results are demonstrated in Figure 8. As shown in Figure 8a, after incubating with the F127ZnFITC R8 pH-responsive system, the SPCA-1 cells were incubated in PBS RPMI-1640 media with 10% FBS under pH 7.4 for 16 h in the presence of the pH-responsive system at pH 7.4, and no/very low faint fluorescence signals were observed in the cultured tumor cells SPCA-1, indicating no significant release of the FITCR8 from the mesostructured hybrid. On the other hand, distinct fluorescence signals were observed in the SPCA-1 cells after incubating with the pH-responsive system in the same conditions under pH 6.2 for 16 h under the same conditions

ARTICLE

except that the pH of the incubation medium was reduced to 6.2 (Figure 8b). It suggests that the F127ZnFITCR8 coordination-bonding system in the mesostructured hybrid has broken down under the low pH condition and released the encapsulated FITCR8, which further penetrated into the living cells. This result is consistent with the in vitro solution release features shown in Figure 6 and confirms the pH responsibility of the system. The pH-responsive system was further implanted into tumor and skeletal muscle of mice to check its effectiveness in different tissues. As shown in Figure 8c, after the drug delivery system was buried in skeletal muscle for 16 h, no significant fluorescence signals were observed in tissue cells which are close to the background, confirming the stability of the system in normal tissues. The low fluorescence signals from the left area of the picture were attributed to a few particles of FITCR8 loaded mesoporous hybrid, which still remained in normal tissues. On the contrary, after the drug delivery system was buried in a solid tumor for 16 h, due to the lower pH in the tumor than physiological tissues, the FITCR8 was released and penetrated into the tumor cells, giving significant fluorescence signals in these tumor cells (Figure 8d). Therefore, the drug delivery system discriminates the tumor and the normal tissues and releases the encapsulated drug specifically in a tumor, strongly indicating its promising use in antitumor applications. Furthermore, this one-pot-fabricated pH-responsive system is extendable to many other types of drugs, including doxorubicin and fluorouracil (Supplementary Figure S11), and the release kinetics and the pH-responsiveness could be readily tuned by properly choosing different types of surfactants and metal ions.

4. CONCLUSIONS In conclusion, a coordination-bonding-based pH-responsive system was constructed into a mesostructured surfactant/silica hybrid by a facile one-pot synthesis method. Surfactant F127 serves dually as the template for mesopores and as the host for metal ions forming the F127metaldrug coordination architecture. Successful pH-responsive release of anticancer drugs has been released under weakly acidic conditions (pH 56). A vector has been designed to endow coordinately inert drugs with pH-responsive properties. The drug delivery system was stable under the physiological conditions while releasing the encapsulated drugs under low pH conditions in a sustained manner (pH 6.26.8). The pH responsibility has been further confirmed by cell assay and in vivo tumor test, suggesting the potential application in antitumor therapies. ’ ASSOCIATED CONTENT

bS

Supporting Information. UVvis adsorption spectra of water solutions under different pH values and the released solution of different systems; XPS spectra and ICP results of the coordination-bonding architecture; control release experiment; characterizations (XRD, TEM, N2 adsorptiondesorption, TGA) of the carrier material during the drug release; in vitro release of other drugs. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Fax þ86-21-5474-1297; tel. þ86-21-5474-2852; e-mail chesa@ sjtu.edu.cn (S.C.), [email protected] (C.G.). 7236



’ ACKNOWLEDGMENT The authors acknowledge the support of the 973 project (2009CB930403) and Grand New Drug Development Program (No.2009ZX09310-007) of China. ’ REFERENCES (1) Vallet-Regí, M.; Ramila, A.; del Real, R. P.; Perez-Pariente, J. Chem. Mater. 2001, 13, 308. (2) Vallet-Regí, M.; Balas, F.; Arcos, D. Angew. Chem., Int. Ed. 2007, 46, 7548. (3) Slowing, I. I.; Trewyn, B. G.; Giri, S.; Lin, V. S.-Y. Adv. Funct. Mater. 2007, 17, 1225. (4) Tsai, C. P.; Chen, C. Y.; Hung, Y.; Chang, F. H.; Mou, C. Y. J. Mater. Chem. 2009, 19, 5737. (5) Xia, W.; Chang, J. J. Controlled Release 2006, 110, 522. (6) Vallet-Regi, M. Chem.—Eur. J. 2006, 12, 5934. (7) Qu, F. Y.; Zhu, G. S.; Huang, S. Y.; Li, S. G.; Sun, J. Y.; Zhang, D. L.; Qiu, S. L. Microporous Mesoporous Mater. 2006, 92, 1. (8) Yiu, H. H. P.; Wright, P. A. J. Mater. Chem. 2005, 15, 3690. (9) Charnay, C.; Begu, S.; Tourne-Peteilh, C.; Nicole, L.; Lerner, D. A.; Devoisselle, J. M. Eur. J. Pharm. Biopharm. 2004, 57, 533. (10) Andersson, J.; Rosenholm, J.; Areva, S.; Linden, M. Chem. Mater. 2004, 16, 4160. (11) Cauda, V.; Engelke, H.; Sauer, A.; Arcizet, D.; Braeuchle, C.; Raedler, J.; Bein, T. Nano Lett. 2010, 10, 2484. (12) Cauda, V.; Muehlstein, L.; Onida, B.; Bein, T. Microporous Mesoporous Mater. 2009, 118, 435. (13) Sauer, A. M.; Schlossbauer, A.; Ruthardt, N.; Cauda, V.; Bein, T.; Br€auchle, C. Nano Lett. 2010, 10, 3684. (14) Zhu, Y.; Shi, J.; Shen, W.; Dong, X.; Feng, J.; Ruan, M.; Li, Y. Angew. Chem., Int. Ed. 2005, 44, 5083. (15) Park, C.; Oh, K.; Lee, S. C.; Kim, C. Angew. Chem., Int. Ed. 2007, 46, 1455. (16) Song, S. W.; Hidajat, K.; Kawi, S. Chem. Commun. 2007, 4396. (17) Angelos, S.; Yang, Y. W.; Patel, K.; Stoddart, J. F.; Zink, J. I. Angew. Chem., Int. Ed. 2008, 47, 2222. (18) Casasus, R.; Climent, E.; Marcos, M. D.; Martínez-Ma~ nez, R.; Sancenon, F.; Soto, J.; Amoros, P.; Cano, J.; Ruiz, E. J. Am. Chem. Soc. 2008, 130, 1903. (19) Zhou, Z.; Zhu, S.; Zhang, D. J. Mater. Chem. 2007, 17, 2428. (20) Tian, B. S.; Yang, C. J. Phys. Chem. C 2009, 113, 4925. (21) You, Y. Z.; Kalebaila, K. K.; Brock, S. L.; Oupicky, D. Chem. Mater. 2008, 20, 3354. (22) Mal, N. K.; Fujiwara, M.; Tanaka, Y.; Taguchi, T.; Matsukata, M. Chem. Mater. 2003, 15, 3385. (23) Nguyen, T. D.; Leung, K. C. F.; Liong, M.; Liu, Y.; Stoddart, J. F.; Zink, J. I. Adv. Funct. Mater. 2007, 17, 2101. (24) Park, C.; Lee, K.; Kim, C. Angew. Chem., Int. Ed. 2009, 48, 1275. (25) Lai, C. Y.; Trewyn, B. G.; Jeftinija, D. M.; Jeftinija, K.; Xu, S.; Jeftinija, S.; Lin, V. S.-Y. J. Am. Chem. Soc. 2003, 125, 4451. (26) Schlossbauer, A.; Kecht, J.; Bein, T. Angew. Chem., Int. Ed. 2009, 48, 3092. (27) Zhao, Y.; Trewyn, B. G.; Slowing, II; Lin, V. S.-Y. J. Am. Chem. Soc. 2009, 131, 8398. (28) Thomas, C. R.; Ferris, D. P.; Lee, J.-H.; Choi, E.; Cho, M. H.; Kim, E. S.; Stoddart, J. F.; Shin, J.-S.; Cheon, J.; Zink, J. I. J. Am. Chem. Soc. 2010, 132, 10623. (29) Gerweck, L. E.; Vijayappa, S.; Kozin, S. Mol. Cancer Ther. 2006, 5, 1275. (30) Yang, X. Q.; Grailer, J. J.; Pilla, S.; Steeber, D. A.; Gong, S. Q. Bioconjugate Chem. 2010, 21, 496. (31) Li, F.; Wu, H.; Zhang, H.; Li, F.; Gu, C. H.; Yang, Q. Carbohydr. Polym. 2009, 77, 773. (32) Engin, K.; Leeper, D. B.; Cater, J. R.; Thistlethwaite, A. J.; Tupchong, L.; McFarlane, J. D. Int. J. Hyperthermia 1995, 11, 211.

ARTICLE

(33) Cauda, V.; Argyo, C.; Schlossbauer, A.; Bein, T. J. Mater. Chem. 2010, 20, 4305. (34) Lee, E. S.; Oh, K. T.; Kim, D.; Youn, Y. S.; Bae, Y. H. J. Controlled Release 2007, 123, 19. (35) Casasus, R.; Marcos, M. D.; Martinez-Manez, R.; Ros-Lis, J. V.; Soto, J.; Villaescusa, L. A.; Amoros, P.; Beltran, D.; Guillem, C.; Latorre, J. J. Am. Chem. Soc. 2004, 126, 8612. (36) Yang, Q.; Wang, S. H.; Fan, P. W.; Wang, L. F.; Di, Y.; Lin, K. F.; Xiao, F. S. Chem. Mater. 2005, 17, 5999. (37) Song, S. W.; Hidajat, K.; Kawi, S. Langmuir 2005, 21, 9568. (38) Nguyen, T. D.; Leung, K. C. F.; Liong, M.; Pentecost, C. D.; Stoddart, J. F.; Zink, J. I. Org. Lett. 2006, 8, 3363. (39) Leung, K. C. F.; Nguyen, T. D.; Stoddart, J. F.; Zink, J. I. Chem. Mater. 2006, 18, 5919. (40) Hoekstra, R. M.; Telo, J. o. P.; Wu, Q.; Stephenson, R. M.; Nelsen, S. F.; Zink, J. I. J. Am. Chem. Soc. 2010, 132, 8825. (41) Zhao, Y.-L.; Li, Z.; Kabehie, S.; Botros, Y. Y.; Stoddart, J. F.; Zink, J. I. J. Am. Chem. Soc. 2010, 132, 13016. (42) Gao, C.; Zheng, H.; Xing, L.; Shu, M.; Che, S. Chem. Mater. 2010, 22, 5437. (43) Son, Y. H.; Park, M.; Bin Choy, Y.; Choi, H. R.; Kim, D. S.; Park, K. C.; Choy, J. H. Chem. Commun. 2007, 2799. (44) Gao, C.; Izquierdo-Barba, I.; Nakase, I.; Futaki, S.; Ruan, J.; Sakamoto, K.; Sakamoto, Y.; Kuroda, K.; Terasaki, O.; Che, S. Microporous Mesoporous Mater. 2009, 122, 201. (45) Khattak, S. F.; Bhatia, S. R.; Roberts, S. C. Tissue Eng. 2005, 11, 974. (46) Chintapalli, S.; Frech, R.; Grady, B. Polymer 1997, 38, 6189. (47) Bernson, A.; Lindgren, J.; Huang, W.; Frech, R. Polymer 1995, 36, 4471. (48) Rothbard, J. B.; Garlington, S.; Lin, Q.; Kirschberg, T.; Kreider, E.; McGrane, P. L.; Wender, P. A.; Khavari, P. A. Nat. Med. 2000, 6, 1253. (49) Prasad, A. S. Mol. Med. 2008, 14, 353.

7237
