Biomaterials 35 (2014) 1676e1685

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Feesalphen complexes from intracellular pH-triggered degradation of Fe3O4@Salphen-InIII CPPs for selectively killing cancer cells Shuai Xu a,1, Jing Liu b,1, Dian Li a, Liming Wang c, Jia Guo a, **, Changchun Wang a, Chunying Chen b, * a

State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, No. 220, Handan Road, Yangpu District, Shanghai 200433, PR China CAS Key Laboratory for Biomedical Effects of Nanomaterials and Nanosafety, National Center for Nanoscience and Technology, Chinese Academy of Science, No. 11, Beiyitiao Zhongguancun, Beijing 100190, PR China c Institute of High Energy Physics Chinese Academy of Sciences, Beijing 100049, PR China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 11 September 2013 Accepted 31 October 2013 Available online 16 November 2013

We propose a modular synthetic strategy to constitute metallosalphen prodrugs in the form of coordination polymer nanoparticles, comprising magnetite nanocrystal colloidal cluster as core and salphenInIII coordination polymer as shell. These composite nanoparticles are not only equipped with intense photoluminescence, sensitive magnetic responsiveness and pH-dependent degradability, but also serve as prodrugs to accomplish intercellular conversion from non-toxic nanoparticles (Fe3O4@Salphen-InIII) to pharmacologically active complexes (Feesalphen), allowing to specifically inhibit the proliferation of A549 cancer cells via caspase activation. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Coordination polymer Magnetic nanoparticle Metal complex Prodrug Target drug delivery

1. Introduction Medical inorganic chemistry is a thriving area of research [1e7], which was initially fueled by the serendipitous discovery of the cisplatin anti-proliferative activity [8].Tremendous efforts have been dedicated to exploring a variety of metal complexes in pharmaceutical use [1e11]. From the point of view of synthesis, metal complexes offer a versatile platform for design of anticancer agents due to the distinct characteristic of central metal ions, such as multiple coordination numbers, accessible redox states, flexible ligand substitutions, and diverse geometries. Of these members, cisplatin is one of the leading metal-based chemotherapeutics, being pronouncedly potent against a wide spectrum of cancer cell lines while used alone or in combination with other drugs [12]. Significant toxicity side effects and drug resistances, however, have limited its clinical applications [13,14]. There is a need, thereby, for new metallodrugs that are aimed at the lower side effects as well as specificity and efficacy in cancer therapies. Linppard et al. prepared

* Corresponding author. Tel.: þ86 10 82545560. ** Corresponding author. Tel.: þ86 21 51630304. E-mail addresses: [email protected] (J. Guo), [email protected] (C. Chen). 1 Theses authors contributed equally to this work. 0142-9612/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biomaterials.2013.10.081

a polymeric nanoparticle to encapsulate the water-soluble Pt(IV) prodrug by double emulsion for the purpose of reduced toxic side effect and controlled release [15]. Metallo-salen/salphen complexes are widely used as catalysts in selective oxidation, organic epoxidation, CO2 fixation, and so on [16]. More intriguingly, their notable apoptotic and antitumor activities have been specifically investigated in the primary studies, which elucidated that functions of central metal ions (e.g. MnIII, FeII, and FeIII) and substituents of salen/salphen ligands were both responsible for tumor-selective apoptosis and cytotoxicity toward cisplatin-resistant cancer cells [17e21]. Albeit with the achievements of metallo-salen/salphen complexes, there still exist some major drawbacks, i.e., large dose of administration, poor water solubility, short circulating time and low bioavailability, all of which would elicit pharmacological deficiencies and deleterious effects [22]. To circumvent these issues, rationally designed drug delivery systems rather than new metal complexes should be greatly developed with the aim of enhancing the performance profile of current metallodrugs. Meanwhile, “safe” delivery of metal complexes to their targets also poses one crucial challenge in cancer chemotherapy, owing to their strong coordination interaction with specific biomolecules [23]. Since Mirkin et al. pioneered the study in synthesis of coordination polymer particles (CPPs), CPPs arouse mounting interests in a wide range of fields [24,25]; they promise great potentials for gas

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

1677

adsorption [26e28], drug delivery [29e34], biological detection [35], optical imaging [36e40], supercapacitive storage [41] and heterogeneous catalysis [42,43]. In light of flexibility and manipulability over coordination polymerization, organic anticancer drugs could either be in situ encapsulated within CPPs as secondary ligands [30,33], or yield nanoparticles self-supported via the interaction of metal-ion-drug coordination [29,32,34]. Thus the innovative model of nanocarriers for drug delivery is constructed in the form of CPPs, and, essentially, acid cleavability of coordination bonds allows for the sustained, pH-responsive drug release. For the aforementioned metallodrugs, however, integration of them into CPPs has been rarely concerned thus so far. Lin et al. realized the first metallodrug-incorporated CPPs by catenating Ir ions with the decorated cisplatin, whereas the activity of chemically modified metallodrugs is compromised to some extent [31]. Therefore, we considered not to comply with the conventional formulation that usually engineers “small” metal complexes into “large” CPPs, but to envision an alternative solution to originate a modular synthesis of CPPs using constituent components of metallodrugs. Hence, we designed a well-defined coreeshell Fe3O4@Salphen-InIII CPP comprising Fe3O4 colloidal nanocrystal cluster as core and salphenInIII coordination polymer as shell with the aim of tackling systematic toxicity and side effects as well as enhancing site-specific delivery of metallodrugs.

flowing air atmosphere at a heating rate of 20  C min1 from 100 to 800  C. Fluorescence spectra were obtained at room temperature using an FLS920 spectrofluorimeter. MALDI-TOF mass spectrometry analysis was performed in positive reflection mode on a 5800 Proteomic Analyzer (Applied Biosystems, Framingham, MA, USA) with a Nd:YAG laser at 355 nm, a repetition rate of 200 Hz, and an acceleration voltage of 20 kV. Nitrogen sorption isotherms were obtained on an ASAP2020 (Micromeritics, USA) accelerated surface area analyzer at 77 K. Before measurements, the samples were degassed under vacuum at 200  C for at least 6 h. The BrunauereEmmetteTeller (BET) method was utilized to calculate the specific surface areas. By using the Barrett-Joyner-Halenda (BJH) model, the pore size distributions were derived from the desorption branches of isotherms, and the total pore volumes were estimated from the adsorbed volume at a relative pressure of 0.971.

2. Materials and methods

2.6. Study of the change in cell morphology

2.1. Materials Iron(III) chloride hexahydrate (FeCl3$6H2O), sodium acetate (NaOAc), ethylene glycol (EG), trisodium citrate dihydrate (Na3Cit$2H2O) were purchased from Sinopharm Chemical Reagents Co. Ltd. Indium nitrate hydrate (In(NO3)3$xH2O) was purchased from Aladdin Reagent Company. Trifluoroacetic acid (99.8%) was purchased from Merck (Darmstadt, Germany). Cell Counting Kit-8 (CCK-8) was purchased from Dojindo Laboratories (Japan). Lung carcinoma cell (A549) was obtained from the American Type Culture Collection (ATCC). All the reagents were analytical grade and used as received. Deionized water (Millipore) of resistivity greater than 18.0 MU cm were used all through the experiments. The modified salphen ligands, N,N0 -phenylenebis(salicylideneimine) dicarboxylic acid, were synthesized according to the literature [44]. Salphen-InIII CPPs were prepared by following the known method [45]. 2.2. Synthesis of citrate-stabilized MCNCs MCNCs were prepared through a solvothermal process [46]. Typically, 2.2 g FeCl3$6H2O, 2.6 g NaOAc, and 0.48 g Na3Cit were dissolved in 40 mL of EG to form a homogeneous yellow dispersion with the aid of ultrasonication. The mixture was then stirred vigorously at 160  C for 1 h. After that, it was instantly transferred into a Teflon-lined stainless-steel autoclave with 50 mL capacity. The reaction was allowed to proceed at 200  C in an oven for 15 h. After the solvothermal process, the precipitates were collected by a magnet and rinsed several times with ethanol and deionized water to remove residues. The products were dried under vacuum for further use.

2.5. Cell viability assay All cells were cultured in complete 1640 medium, with 10% (v/v) fetal bovine serum (FBS), at 37  C, 5% CO2 and 10% humidity. Cell viability was measured using a Cell Counting Kit-8. Firstly, cells were seeded into 96-well plates (Costar, Corning, NY). After incubating (37  C, 5% CO2 and 10% humidity) for 24 h, the culture medium was removed and replaced with the complete medium containing 0, 1, 5, 10, 20 and 50 mM salphen-InIII CPPs, Fe3O4 and Fe3O4@Salphen-InIII CPPs, respectively. Moreover, cell viability was assessed after incubation times of 12, 24, and 48 h. Meanwhile, the wells unexposed to any samples were regarded as control. Then a mixture of the tetrazolium reagent (from the Cell Counting Kit-8) and the complete medium (1:10) was added into each well. Finally, cell viability was calculated as the absorbance ratio between test and control wells. The absorbance at 450 nm was measured and referenced with that at 650 nm by Infinite M200 microplate reader (Tecan, Durham, USA).

Firstly, cells were seeded into 6-well plates. After 24-h incubating (37  C, 5% CO2 and 10% humidity), the culture medium was replaced with complete medium containing 50 mM Fe3O4@Salphen-InIII CPPs. After another 24 h, adherent cells were washed with PBS for three times. The nucleus was labeled with 4 mg mL1 Hoechst 33,342 for 5 min, and then the cells were washed and observed using fluorescence microscope at 10 magnification. 2.7. Quantitative measurement for Fe3O4@Salphen-InIII CPPs in cellular uptake A549 and 16HBE cells were cultured with 2 mL of complete medium in 6-well plates at a density of 2  105 cells mL1. After cultured for 24 h at 37  C, 5% CO2 and 10% humidity, the medium was removed and changed to the complete medium, respectively, containing 10 mM and 20 mM Fe3O4@Salphen-InIII CPPs, for internalization experiment. The cells were incubated for different time slice including 1, 3, 6, 12, 24 and 48 h, four wells of cells at each time point. Cells in each well were gently washed three times with PBS, digested with 0.25% trypsin containing 0.02% EDTA, centrifuged for 10 min at 1500 rpm, collected and counted. Afterwards, 3 mL HNO3 was added to each cell sample, and then transferred to flasks for pre-digestion overnight. In the next day, 2 mL 30% H2O2 was added to each flask. The flasks were placed onto a hot plate and maintained at 150  C for 3 h until digestion was complete, and then they were cooled to room temperature. The solution in each flask was diluted to 5 mL with 2% HNO3. A series of Fe standard solutions (0, 0.1, 0.5, 1, 5, 10, 50 and 100 ppm) were prepared with the above solution. Both standard and test solutions were measured by inductively coupled plasma mass spectrometry (ICP-MS, Thermal Elemental X7, Thermal Fisher Scientific Inc, USA).

2.3. Synthesis of Fe3O4@Salphen-InIII CPPs

2.8. Study of pH changes in lysosome

The carboxylate-modified salphen ligands (6 mg, 15 mmol) and In(NO3)3$xH2O (4.47 mg, 15 mmol) were dissolved in 4 mL DMF to form a yellow solution, to which 8 mg MCNC powder was added. The above precursor mixture was ultrasonicated, transferred to a 25 mL three-necked flask, and stirred vigorously at 120  C. The reaction was allowed to proceed for 15 min, during which the brownish precipitates were gradually formed in the mixture. After cooling to ambient temperature, the products were collected by an applied magnet and washed with DMF and deionized water, respectively.

Cells were seeded into 6-well plates for 24 h at 37  C, 5% CO2 and 10% humidity. After that, cells were incubated with 2 mM LysoSensorÔ Green DND-189 for 40 min prior to rinsing three times with PBS. Then pH of lysosome was analyzed when 20 mM Fe3O4@Salphen-InIII CPPs were incubated in complete medium for different time slice (1, 2, 3, 4, 5, 6, 8, 10 and 12 h). Finally, cells were digested, collected, acquired on flow cytometry (BD FACS Calibur) and analyzed with FCS Express software.

2.4. Characterization High resolution transmission electron microscopy (HR TEM) images were taken on a JEM-2010 (JEOL, Japan) transmission electron microscope at an accelerating voltage of 200 kV. Samples dispersed at an appropriate concentration were cast onto a carbon-coated copper grid. Magnetic characterization was carried out with a vibrating sample magnetometer on a Model 6000 physical property measurement system (Quantum Design, USA) at 300 K. Fourier transform infrared (FT IR) spectra were recorder on a Magna-550 (Nicolet) spectrometer. The dried samples were mixed with KBr, and they were compressed to a plate for measurement. TG analysis was obtained with a Pyrisis-1 (PerkineElmer, USA) thermal analysis system under a

2.9. Assessment of the integrity of lysosomal membrane Cells were seeded into 35 mm petri-dishes for 24 h at 37  C, 5% CO2 and 10% humidity. Cells were then incubated for 15 min in the complete medium containing 5 mg mL1 AO and 10% FBS before rinsed three times with PBS. Lysosomal membrane permeation of the two cells was analyzed after 24-h treatment with a complete medium containing 50 mM Fe3O4@Salphen-InIII CPPs and 2 mg mL1 PEI (as positive control), respectively. Finally, the dead cells were removed by wash and the remaining cells were observed using confocal microscope (Perkin Elmer Ultra View Vox system, USA). The emission of samples was detected at 537 nm (green) and 615 nm (red) with excitation wavelength of 488 nm. Identically, the cells were stained with AO, digested, collected and measured by flow cytometry using an excitation wavelength of 488 nm and an emission wavelength of 670 nm.

1678

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

Fig. 1. (a) Schematic illustration of synthesis of Fe3O4@Salphen-InIII CPPs and sequential degradation of a salphen-InIII CPP shell and a Fe3O4 nanocluster core at pH of 5.0; TEM images of Fe3O4 nanoclusters (b), Fe3O4@Salphen-InIII CPPs (c) and the magnified view of a representative composite CPPs (d).

2.10. Evaluation of mitochondrial membrane potential Cells were seeded into 35 mm petri-dishes for 24 h and washed three times with PBS. Then, they were incubated with 5 mg mL1 Rh123 for 1 h. Mitochondrial membrane potential of the two cells was analyzed after 24-h treatment by complete medium that respectively mixed with 20 mM Fe3O4@Salphen-InIII CPPs and 2.5 mg mL1 carbonyl cyanide m-chlorophenylhydrazone (CCCP) as a positive control. Finally, the treated cells were observed using fluorescence microscopy at 40 magnification. The similar treatment for two cells was conducted for measurement of flow cytometry.

fluorescent emission are then promptly released from shell and could serve as optical probe, but without showing any pharmacological activity; (3) the sustained degradation of Fe3O4 core will produce abundant Fe2þ and Fe3þ ions, and the free salphen ligands are coordinated with them to form Feesalphen complexes, which

2.11. Western-blotting A549 cells and 16HBE cells, grown to 80% confluence in a 6-well plate at 37  C, 5% CO2 and 10% humidity, were cultured with a complete medium containing 50 mM Fe3O4@Salphen-InIII CPPs for 12 h, 24 h and 48 h, respectively. Then, cells were washed with PBS three times, harvested with a scraper and centrifuged at 4  C. After discarding the supernatant, the cells were lysed to collect the cellular proteins. Equal amounts of proteins were separated on 12% SDS-PAGE gels and transferred to nitrocellulose membranes. Then the membranes were blocked with 5% non-fat milk in TBST at room temperature. The primary antibody was used at 1:1000 of dilution, and the secondary antibody was used at 1:5000 of dilution. Blots were developed using enhanced chemiluminescence. The tested proteins were procaspase-3 and actin.

3. Results and discussion 3.1. Synthesis and characterization of Fe3O4@Salphen-InIII CPPs As shown in Fig. 1a, we proposed a tentative metallodrug delivery and release by a pH-sensitive degradation and conversion process that includes: (1) a sequential degradation occurs from shell to core moieties at pH of 5.0; (2) salphen ligands with intense

Fig. 2. FT IR spectra of Fe3O4@Salphen-InIII CPPs, salphen-InIII CPPs, and MCNCs.

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

Fig. 3. Thermogravimetric analysis of Fe3O4@Salphen-InIII CPPs, salphnen-InIII CPPs, and MCNCs measured in air from 100 to 800  C with an increment of 20  C min1.

1679

will trigger urgent cytotoxicity. Since the CPP-based magnetic nanoparticles are non-toxic in delivery prior to the pH-stimulated “activation”, they thus function not just as nanocarriers, but prodrugs as well, which are commonly thought as inert drug derivatives that undergo in vivo metabolism to release the active species [47]. Besides, Fe3O4@Salphen-InIII CPPs could achieve a longer circulation time due to improved solution dispersibility, and are conducive to magnetic manipulation in targeted delivery, thereby superior to the free prodrug solution as well. Synthesis of Fe3O4@Salphen-InIII CPPs involves two steps. Magnetite colloidal nanocrystal clusters (MCNCs) were solvothermally prepared in ethylene glycol with stabilizers of sodium citrate [46,48e50]. As displayed in TEM image (Fig. 1b), they could afford relatively narrow size distribution with a diameter of approximately 200 nm. Numerous carboxylate groups attached on the surface of MCNCs render them negatively charged and improvingly solution dispersible, and are available to chelate metal ions for the following coordination polymerization. The carboxylate-functionalized salphen ligands, N,N0 -phenylenebis(salicylideneimine) dicarboxylic acid, were then added with In3þ ions in DMF solution of MCNCs, and the complexation reaction between eCOOH of salphen and In3þ ions was allowed to proceed at 120  C for 30 min. The formed coordination polymers deposited onto the MCNCs, leading to the continuous

Fig. 4. (a) Magnetic hysteresis curve of Fe3O4@Salphen-InIII CPPs and MCNCs. The inset enlarges the view of magnetic hysteresis curve for Fe3O4@Salphen-InIII CPPs at the low magnetic field, wherein the two markers present the values of saturation remanence (Mrs ¼ 2.5 emu g1) and coercivity (hc ¼ 32 Oe), respectively. (b) Photographs of the aqueous dispersions of MCNCs and Fe3O4@Salphen-InIII CPPs without (left) and with (right) the applied magnet.

1680

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

evolution of salphen-InIII shell; ultimately, the brownish products precipitated out of DMF solution. A close look at the TEM images (Fig. 1c,d) reveals that one MCNC was wholly wrapped by a homogenous polymer shell with a thickness of roughly 40 nm, without any free polymer residues observed. Controllability over size of salphenInIII shell could be achieved by varying the feeding concentrations of salphen ligands and In3þ ions. TEM images (Fig. S1 in SI) display that the polymer shells in dimension were tuned from 5 to 50 nm, and appeared an increment trend of uniformity in morphology and density in texture. FT IR spectra show the characteristic absorption bands of eC]O (1608 cm1), FeeO (570 cm1) for the resulting CPP composites, both of which could be well ascribed to the coordinated carboxylate groups of salphen-InIII polymers [45] and iron oxide of MCNCs [48], respectively (Fig. 2). TGA measurement was performed in air from 100 to 800  C for quantifying organic components. It was found that the decomposition temperature of the coordination polymer shell reached up to 392  C, and 29.3 wt.% of salphen moiety was involved in it. The results unequivocally prove an exceptional thermal stability and a high coverage of coordination polymers (Fig. 3). The superparamagnetism of Fe3O4@Salphen-InIII CPPs was investigated by utilizing a vibrating sample magnetometer in an external magnetic field at 300 K. As exhibited in Fig. 4a, the typical magnetic hysteresis curves were both obtained for MCNCs and Fe3O4@Salphen-InIII CPPs [51]. The saturation magnetization of the composite CPPs was up to 25.5 emu g1, and when compared with 63.8 emu g1 for MCNCs, they could give a roughly 60.0% of magnetic content. The inset of Fig. 4a depicts an enlarged section of the magnetization plot near the zero magnetic field. Saturation remanence (Mrs) and coercivity (hc) were determined from the intersection of the hysteresis loop with the two axes, giving 2.5 emu g1 and 32 Oe, respectively. These two values mean that a rather low residual magnetization is present on removal of the external magnetic field, and only a low intensity of applied magnetic field is required to reduce the magnetization to zero. This is substantial evidence that the MCNC moiety in core portion behaves as a singledomain superparamagnetic Fe3O4 nanocrystal, which allows for the reversible enrichment/dispersion of Fe3O4@Salphen-InIII CPPs with control of an external magnetic field in aqueous solution (Fig. 4b). Acidic degradation of Fe3O4@Salphen-InIII CPPs was meticulously studied in vitro using the inductively coupled plasma atomic emission spectroscopy (ICP-AES) and fluorescence spectrometry together. Prior to the degradation experiments, the analysis of N2 adsorption at 77 K was conducted for the MCNCs (Fig. S2 in SI), showing a high surface area of 207 m2 g1 and a mesopore of around 5.0 nm. The porosity of MCNCs is beneficial of guest molecules infiltrated from outside to inside of particles. As shown in Fig. 5a, MCNCs were decomposed gradually at pH of 5.0 over 50 h, and were completely converted into iron ions (FeII and FeIII) confirmed by ICP-APE measurement, as previously reported [52,53]. After a coating of salphen-InIII coordination polymers, the collapse of coreeshell structure at pH of 5.0 was monitored in real time by fluorescence spectrometry (Fig. S3 in SI). The results were compiled as a function of incubation time in Fig. 5b. It can be seen that the fluorescence intensity at the wavelength of 440 nm was culminated in the dialyzate in 30 min, indicating that the salphenInIII shell depolymerized to release the majority of free salphen ligands. As the incubation time prolonged, the fluorescence was then declined in 30 h and quenched eventually. The photographs for the dialyzate recorded at 1 h and 5 h can visualize the blue emission under irradiation of UV lamp (365 nm), but after 10 h, the fluorescence emission turned rather weak (inset of Fig. 5b). The big difference in fluorescence (Fig. S4 in SI) claimed the possibility that Feesalphen complexes were yielded by complexation of the iron ion with the tetradentate N2O2 donor rather than the carboxylate groups of salphen, since the photo induced-electron-transfer

Fig. 5. Degradation of Fe3O4 nanoclusters (a) and Fe3O4@Salphen-InIII CPPs (b) as a function of incubation time in buffer solution (pH 5.0), which are plotted by measuring the variation on the concentrations of released iron ions and the photoluminescence (PL) intensity of released salphen, respectively. Inset in (b) is a photograph of dialyzates obtained from degradation of Fe3O4@Salphen-InIII CPPs at different time intervals, which is made under irradiation of UV lamp (l ¼ 365 nm).

quenching was observed, which only happens while iron ions are coordinated with the nitrogen atoms of salphen ligands [54]. Meanwhile, the product in dialyzate was examined by MALDI-TOF MS, which ascertained the presence of Feesalphen complexes during the degradation of composite CPPs (Fig. S5 in SI). The sustained degradation of MCNC cores would extend over 30 h, and thus the dose of active Feesalphen complexes was cumulated gradually; thereby the effectiveness seems to be similar to that of free metallodrugs administrated in a controlled manner. 3.2. The cytotoxicity of Fe3O4@Salphen-InIII CPPs To assess the antitumor activities of Fe3O4@Salphen-InIII CPPs, we investigated the viability of carcinoma cells (A549 cells) and normal bronchial epithelial cells (16HBE cells) upon treatment in the presence of Fe3O4@Salphen-InIII CPPs; meanwhile, the references, MCNCs and salphen-InIII CPPs, were used as well. Dose- and time-dependent cytotoxicity profiles for salphen-InIII CPPs, MCNCs and Fe3O4@Salphen-InIII CPPs are shown in Fig. 6a,b. One can see that when the concentration was up to 10 mM, Fe3O4@Salphen-InIII CPPs became inhibitory activity on A549 cells but caused little damage to 16HBE cells at 24 h. With the following increase in administration dose and incubation time, the viability of A549 cells declined correspondingly, and approximately 80% of A549 cells were dead after 48 h as 50 mM of Fe3O4@Salphen-InIII CPPs was used. In a sharp contrast, 16HBE cells were nearly not suffered from the Fe3O4@Salphen-InIII CPPs, regardless of the dose- and timedependent trend implicated in the former case. Also, this indicates

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

1681

Fig. 6. Time- and dose-dependent cytotoxicity to A549 (a) and 16HBE (b) cells for salphen-InIII CPPs, MCNCs, and Fe3O4@Salphen-InIII CPPs, respectively, and quantitative analysis of internalized Fe3O4@Salphen-InIII CPPs in A549 (c) and 16HBE (d) cells by ICP-MS.

the coordinated indium ions are of biological compatibility in utility [55]. As a control, it was verified that salphen-InIII CPPs and MCNCs were both inactive to A549 cancer cells, showing a rather limited inhibitory. Microscopic observation (Fig. S6 in SI) revealed that the marked morphological change occurred to A549 cells upon treatment using Fe3O4@Salphen-InIII CPPs, and the change in shape was not observed on 16HBE cells. It is thus confirmed that Fe3O4@Salphen-InIII CPPs possess dose-dependent cytotoxic effect on A549 cancer cells, rather than 16HBE normal cells. The half maximal inhibitory concentration (IC50) values against A549 cells were estimated as 50 and 10 mmol L1 at 48 and 72 h (Table S1 in SI), comparable to those of pharmaceutically active Mn(III)-salen complexes (12e55 mmol L1) [17,21]. 3.3. Cellular uptake of Fe3O4@Salphen-InIII CPPs Due to the prominent activity of Fe3O4@Salphen-InIII CPPs against cancer cells, the investigation of their cellular uptake and intracellular localization appealed our lasting interest. In order to understand the

effect of the internalized quantities of Fe3O4@Salphen-InIII CPPs on cytotoxicity toward A549 and 16HBE cells, we measured the cellular uptake using ICP-MS spectrometry. As is evident from the results compiled in Fig. 6c,d, the internalization of Fe3O4@Salphen-InIII CPPs is time-dependent and cell-type specific. A549 cells have a more rapid rate of uptake as well as improved internalization for Fe3O4@Salphen-InIII CPPs; both are pronouncedly superior to 16HBE cells. Each A549 cell could greatly internalize the Fe3O4@Salphen-InIII CPPs, which was three times higher than 16HBE cells did. 3.4. Lysosome changes caused by Fe3O4@Salphen-InIII CPPs To our knowledge, lysosomes have intra-compartmental pH of 4.0e5.0, substantially less than that of the cytosol (wpH 7.2) [56]. All the enzymes in the lysosome work best at the low pH. This facilitates reducing risk of digesting their own cells if the enzymes are excluded out of the lysosome. As revealed in cell-free system, Fe3O4@Salphen-InIII CPPs are amenable to acid degradation and regeneration of new metal complexes. Therefore, it is needed to

1682

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

Fig. 7. Changes of pH in lysosome determined by flow cytometry using LysoSensorÔ Green DND-189 staining after cells treated by 20 mM Fe3O4@Salphen-InIII CPPs. Comparison of integrity of lysosomal membranes (AO staining) in A549 (cee) and 16HBE (feh) cells between control (c,f) and test groups including 50 mM Fe3O4@Salphen-InIII CPPs for 24 h (d,g) and 2 mg mL1 PEI for 6 h (e,h) by using confocal microscopy, respectively. Their quantitative data for A549 cells (i) and 16HBE cells (j) was determined by flow cytometry when applying 6 mg mL1 PEI, and the same concentration of MCNCs, salphen-InIII CPPs, and Fe3O4@Salphen-InIII CPPs for 24 h, respectively. *Represents P value less than 0.05. Scale bars in (ceh) are 50 mm.

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

confirm whether the Fe3O4@Salphen-InIII CPPs enter the lysosome and also work there. Following this line, we applied the Lysoe Tracker fluorescent probes for labeling and tracking acidic organelles in living cells; the intercellular fluorescence observed will be varied upon pH fluctuation. As displayed in Fig. 7a,b, the fluorescence intensity decreased promptly in the first 10 h and then was recovered in both A549 and 16HBE cells after treatment for 24 h. This indicates that pH value in the lysosomal environment has a dynamic change, namely, decreasing firstly and restoring subsequently. The lysosome can maintain intrinsic pH balance by virtue of proton pumps and chloride ion channels on the membrane [57]. As a result, we reason that the acid degradation of Fe3O4@SalphenInIII CPPs may lead to depletion of a large number of protons and thus allow for the localized pH change in the lysosome. The finding also elucidates the occurrence of intracellular pH-triggered degradation of Fe3O4@Salphen-InIII CPPs. Apart from the change of pH values inside lysosomes, the lysosomal membrane permeation (LMP) was also analyzed using acridine orange (AO) staining with the aim to determine whether LMP can be altered due to internalization of Fe3O4@Salphen-InIII CPPs. Compared with the control group (Fig. 7c), we found the great disruption to the lysosomal membrane while 50 mM of Fe3O4@Salphen-InIII CPPs was exposed to A549 cells for 24 h (Fig. 7d). On the contrary, 16HBE cells did not elicit any remarkable effect on LMP under the same conditions; also, this ignorable change was well consistent with their control group behaving (Fig. 7f,g). When PEI was used as a positive control, the equal effectiveness to A549 rather than 16HBE cells was achieved (Fig. 7e,h), corroborating that Fe3O4@Salphen-InIII CPPs acted well like PEI did. The quantitative data of integrity of the lysosomal membrane in A549 and 16HBE cells were determined by flow cytometry (Fig. 7i,j). Compared to MCNCs and salphen-InIII CPPs, Fe3O4@Salphen-InIII CPPs and PEI both resulted in a significant reduction in the integrity of lysosomal membrane on A549 cells and a negligible effect on that of 16HBE cells, respectively. As such, it is naturally envisaged that the freshly formed Feesalphen complexes are released into cytoplasm and further inhibit the tumor cell growth and proliferation.

1683

3.5. The alteration of mitochondrial membrane potential As far as known, the mitochondrion is the energy factory of the cell, and, therefore, it plays a pivotal role in cellular survival and apoptotic death [4,58]. Thus it prompted us to study the locally activated pharmacological behavior of Fe3O4@Salphen-InIII CPPs to mitochondria portion. To gain insight into the change of mitochondrial potential, we applied the specific dye rhodamine 123 (Rh-123) to stain and visualize the cells in the presence of Fe3O4@Salphen-InIII CPPs. For apoptotic cells, transmembrane potentials will reduce, leading to the release of Rh-123 out of mitochondrion, and, in turn, a rising of fluorescence intensity can be observed. Also, the total content of the released Rh-123 could be quantified by flow cytometry. Accompanied with fluorescence monitor, carbonyl cyanide m-chlorophenylhydrazone (CCCP) was used as a positive control to estimate the efficacy of Fe3O4@Salphen-InIII CPPs. After a 24-h exposure, the results validated that Fe3O4@Salphen-InIII CPPs induced mitochondrial damage in A549 cells but not in 16HBE cells (Fig. 8a,b). This adequately supports the occurrence of prodrug-like conversion, wherein the salphen ligands and iron ions released from the degraded Fe3O4@SalphenInIII CPPs in lysosomes are interacted with each other to form Fee salphen complexes, giving rise to the apoptosis activation as well as the mitochondrion damage. 3.6. Activation of caspase signal pathways Fe-salen/salphen complexes have been subjected to intense studies for demonstrating their functions in activation of caspase enzymes, which are essential to all apoptotic cell death and apoptosis pathways in human cancer cells [18e20]. Thus, the apoptosis activation generated by Fe3O4@Salphen-InIII CPPs could be studied through the assessment of caspase enzyme activity. As reported previously, caspase activation is initiated by the formation of a multimeric apoptosis-activating factor 1 (Apaf-1)/cytochrome c complex that makes procaspases activated; then the procaspases cleave and activate downstream caspases such as caspase-3, -6, and

Fig. 8. Changes of mitochondrial membrane potentials (Rh123 staining) by measuring photoluminescence (PL) intensity using flow cytometry for A549 (a) and 16HBE (b) cells after treatment of 20 mM Fe3O4@Salphen-InIII CPPs for 24 h and 5 mg mL1 CCCP (carbonyl cyanide m-chlorophenylhydrazone) for 1 h, respectively; (c) analysis of caspase-3 by immunoblotting when using 50 mM Fe3O4@Salphen-InIII CPPs for 12, 24 and 48 h, respectively.

1684

S. Xu et al. / Biomaterials 35 (2014) 1676e1685

-7 that ultimately activates endonucleases, resulting in nuclear fragmentation and apoptosis [19]. Hence, caspase-3 is generally regarded as one of the most important executional caspase [59]. We therefore measured the expression of procaspase-3 in the sampletreated A549 and 16HBE cells via the western-blotting method. The result exhibited a time-dependent decrease in A549 cells (Fig. 8c). It can be inferred that Fe3O4@Salphen-InIII CPPs evoke apoptosis of A549 cells via caspase-3 activation. In intrinsic activation, cytochrome c from the mitochondria works in combination with Apaf-1 and ATP to process procaspase-3. These molecules are sufficient to activate caspase-3 in vitro [58,59]. 4. Conclusion To summarize, Fe3O4@Salphen-InIII coordination polymer particles with an exquisite coreeshell structure were modularly synthesized for the sake of a rational design of metal complex prodrugs. Two building modules within particles, i.e., a Fe3O4 colloidal nanocrystal cluster in core and a salphen-InIII coordination polymer in shell, were susceptible to degradation in acid medium (wpH 5.0) sequentially. The released salphen ligands and iron ions were both non-toxic, but after Feesalphen complexes were formed, the cytotoxicity against cancer cells was pronouncedly improved, accompanied with the fluorescence quenching. Cell experiments well proved that the magnetically targeted nanoprodrugs could selectively kill cancer cells; this process is unprecedented in that: (1) pHtriggered degradation of Fe3O4@Salphen-InIII CPPs in the lysosome, (2) formation of the active Feesalphen complexes by the reaction of released iron ions and salphen ligands in the cytoplasm, and (3) mitochondrion-mediated activation of caspase-3 promoting apoptosis via the freshly formed Feesalphen complexes. Moreover, Fe3O4@Salphen-InIII CPPs showed significant selectivity toward cancer cells (such as A549 cells) over normal non-malignant cells (16HBE cells), indicating their potential application towards antitumor therapy. Therefore, it is reasonably anticipated that pHdependent intercellular conversion from the inactive and nontoxic CPPs to therapeutically active metal complexes will initiate a potentially promising line for bio-applications of coordination polymer nanomaterials. Acknowledgment We acknowledge the financial support from the National Natural Science Foundation of China (21004012 and 21128001), National Basic Research Programs (2011CB933401, 2012CB934003 and 2010CB934004), and International Science & Technology Cooperation Program of China, Ministry of Science and Technology of China (2013DFG32340). Appendix A. Supplementary data Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.biomaterials.2013.10.081. References [1] Hambley TW. Metal-based therapeutics. Science 2007;318:1392e3. [2] Kang S-G, Zhou G, Yang P, Liu Y, Sun B, Huynh T, et al. Molecular mechanism of pancreatic tumor metastasis inhibition by Gd@C82(OH)22 and its implication for de novo design of nanomedicine. Proc Natl Acad Sci U S A 2012;109: 15431e6. [3] van Rijt SH, Sadler PJ. Current applications and future potential for bioinorganic chemistry in the development of anticancer drugs. Drug Discov Today 2009;14:1089e97. [4] Wang L, Liu Y, Li W, Jiang X, Ji Y, Wu X, et al. Selective targeting of gold nanorods at the mitochondria of cancer cells: implications for cancer therapy. Nano Lett 2010;11:772e80.

[5] Zhang Z, Wang J, Chen C. Near-infrared light-mediated nanoplatforms for cancer thermo-chemotherapy and optical imaging. Adv Mater 2013;25:3869e80. [6] Li YF, Chen C. Fate and toxicity of metallic and metal-containing nanoparticles for biomedical applications. Small 2011;7:2965e80. [7] Zhang Z, Wang L, Wang J, Jiang X, Li X, Hu Z, et al. Mesoporous silica-coated gold nanorods as a light-mediated multifunctional theranostic platform for cancer treatment. Adv Mater 2012;24:1418e23. [8] Rosenberg B, Vancamp L. Platinum compounds: a new class of potent antitumour agents. Nature 1969;222:385e6. [9] Fricker SP. Metal based drugs: from serendipity to design. Dalton Trans 2007;43:4903e17. [10] Gasser G, Ott I, Metzler-Nolte N. Organometallic anticancer compounds. J Med Chem 2010;54:3e25. [11] Zhang CX, Lippard SJ. New metal complexes as potential therapeutics. Curr Opin Chem Biol 2003;7:481e9. [12] Kelland L. The resurgence of platinum-based cancer chemotherapy. Nat Rev Cancer 2007;7:573e84. [13] Dempke W, Voigt W, Grothey A, Hill BT, Schmoll H-J. Cisplatin resistance and oncogenes-a review. Anticancer Drugs 2000;11:225e36. [14] Jung Y, Lippard SJ. Direct cellular responses to platinum-induced DNA damage. Chem Rev 2007;107:1387e407. [15] Johnstone TC, Kulak N, Pridgen EM, Farokhzad OC, Langer R, Lippard SJ. Nanoparticle encapsulation of mitaplatin and the effect thereof on in vivo properties. ACS Nano 2013;7:5675e83. [16] Gupta K, Sutar AK. Catalytic activities of Schiff base transition metal complexes. Coord Chem Rev 2008;252:1420e50. [17] Ansari KI, Grant JD, Kasiri S, Woldemariam G, Shrestha B, Mandal SS. Manganese (III)-salens induce tumor selective apoptosis in human cells. J Inorg Biochem 2009;103:818e26. [18] Ansari KI, Grant JD, Woldemariam GA, Kasiri S, Mandal SS. Iron (III)-salen complexes with less DNA cleavage activity exhibit more efficient apoptosis in MCF7 cells. Org Biomol Chem 2009;7:926e32. [19] Ansari KI, Kasiri S, Grant JD, Mandal SS. Fe(III)-salen and -salphen complexes induce caspase activation and apoptosis in human cells. J Biomol Screen 2011;16:26e35. [20] Woldemariam GA, Mandal SS. Iron(III)-salen damages DNA and induces apoptosis in human cell via mitochondrial pathway. J Inorg Biochem 2008;102:740e7. [21] Ansari K, Kasiri S, Grant JD, Mandal SS. Apoptosis and anti-tumour activities of manganese(III)-salen and -salphen complexes. Dalton Trans 2009;40:8525e31. [22] Lainé A-L, Passirani C. Novel metal-based anticancer drugs: a new challenge in drug delivery. Curr Opin Pharmacol 2012;12:420e6. [23] Rodgers M, Armentrout P. A thermodynamic “vocabulary” for metal ion interactions in biological systems. Acc Chem Res 2004;37:989e98. [24] Oh M, Mirkin CA. Chemically tailorable colloidal particles from infinite coordination polymers. Nature 2005;438:651e4. [25] Spokoyny AM, Kim D, Sumrein A, Mirkin CA. Infinite coordination polymer nano-and microparticle structures. Chem Soc Rev 2009;38:1218e27. [26] Farha OK, Spokoyny AM, Mulfort KL, Galli S, Hupp JT, Mirkin CA. Gas-sorption properties of cobalt(II)-carborane-based coordination polymers as a function of morphology. Small 2009;5:1727e31. [27] Jeon YM, Armatas GS, Heo J, Kanatzidis MG, Mirkin CA. Amorphous infinite coordination polymer microparticles: a new class of selective hydrogen storage materials. Adv Mater 2008;20:2105e10. [28] Lee HJ, Cho W, Jung S, Oh M. Morphology-selective formation and morphology-dependent gas-adsorption properties of coordination polymer particles. Adv Mater 2009;21:674e7. [29] Huxford RC, Boyle WS, Liu D, Lin W. Lipid-coated nanoscale coordination polymers for targeted delivery of antifolates to cancer cells. Chem Sci 2012;3: 198e204. [30] Imaz I, Rubio-Martínez M, García-Fernández L, García F, Ruiz-Molina D, Hernando J, et al. Coordination polymer particles as potential drug delivery systems. Chem Commun 2010;46:4737e9. [31] Rieter WJ, Pott KM, Taylor KM, Lin W. Nanoscale coordination polymers for platinum-based anticancer drug delivery. J Am Chem Soc 2008;130:11584e5. [32] Xing L, Cao Y, Che S. Synthesis of core-shell coordination polymer nanoparticles (CPNs) for pH-responsive controlled drug release. Chem Commun 2012;48:5995e7. [33] Xing L, Zheng H, Cao Y, Che S. Coordination polymer coated mesoporous silica nanoparticles for pH-responsive drug release. Adv Mater 2012;24:6433e7. [34] Xing L, Zheng H, Che S. A pH-responsive cleavage route based on a metalorganic coordination bond. Chem Eur J 2011;17:7271e5. [35] Li H, Sun X. Fluorescence-enhanced nucleic acid detection: using coordination polymer colloids as a sensing platform. Chem Commun 2011;47:2625e7. [36] Aimé C, Nishiyabu R, Gondo R, Kimizuka N. Switching on luminescence in nucleotide/lanthanide coordination nanoparticles via synergistic interactions with a cofactor ligand. Chem Eur J 2010;16:3604e7. [37] Imaz I, Hernando J, Ruiz-Molina D, Maspoch D. Metal-organic spheres as functional systems for guest encapsulation. Angew Chem Int Ed 2009;48: 2325e9. [38] Liu D, Huxford RC, Lin W. Phosphorescent nanoscale coordination polymers as contrast agents for optical imaging. Angew Chem Int Ed 2011;50:3696e700. [39] Nishiyabu R, Aimé C, Gondo R, Kaneko K, Kimizuka N. Selective inclusion of anionic quantum dots in coordination network shells of nucleotides and lanthanide ions. Chem Commun 2010;46:4333e5.

S. Xu et al. / Biomaterials 35 (2014) 1676e1685 [40] Nishiyabu R, Hashimoto N, Cho T, Watanabe K, Yasunaga T, Endo A, et al. Nanoparticles of adaptive supramolecular networks self-assembled from nucleotides and lanthanide ions. J Am Chem Soc 2009;131:2151e8. [41] Xu S, You L, Zhang P, Zhang Y, Guo J, Wang C. Fe3O4@coordination polymer microspheres with self-supported polyoxometalates in shells exhibiting highperformance supercapacitive energy storage. Chem Commun 2013;49:2427e9. [42] Choi J, Yang HY, Kim HJ, Son SU. Organometallic hollow spheres bearing bis (N-heterocyclic carbene)-palladium species: catalytic application in threecomponent strecker reactions. Angew Chem Int Ed 2010;49:7718e22. [43] Park KH, Jang K, Son SU, Sweigart DA. Self-supported organometallic rhodium quinonoid nanocatalysts for stereoselective polymerization of phenylacetylene. J Am Chem Soc 2006;128:8740e1. [44] Kitaura R, Onoyama G, Sakamoto H, Matsuda R, Si Noro, Kitagawa S. Immobilization of a metallo schiff base into a microporous coordination polymer. Angew Chem Int Ed 2004;43:2684e7. [45] Jo C, Lee HJ, Oh M. One-pot synthesis of silica@coordination polymer core-shell microspheres with controlled shell thickness. Adv Mater 2011;23:1716e9. [46] Liu J, Sun Z, Deng Y, Zou Y, Li C, Guo X, et al. Highly water-dispersible biocompatible magnetite particles with low cytotoxicity stabilized by citrate groups. Angew Chem Int Ed 2009;48:5875e9. [47] Graf N, Lippard SJ. Redox activation of metal-based prodrugs as a strategy for drug delivery. Adv Drug Deliv Rev 2012;64:993e1004. [48] Luo B, Xu S, Luo A, Wang WR, Wang SL, Guo J, et al. Mesoporous biocompatible and acid-degradable magnetic colloidal nanocrystal clusters with sustainable stability and high hydrophobic drug loading capacity. ACS Nano 2011;5:1428e35. [49] Guo J, Yang WL, Wang CC. Magnetic colloidal supraparticles: design, fabrication and biomedical applications. Adv Mater 2013;25:5196e214.

1685

[50] Kang X-J, Dai Y-L, Ma P-A, Yang D-M, Li C-X, Hou Z-Y, et al. Poly(acrylic acid)modified Fe3O4 microspheres for magnetic-targeted and pH-triggered anticancer drug delivery. Chem Eur J 2012;18:15676e82. [51] Chikazumi S. Physics of ferromagnetism. 2nd ed. New York: Oxford University Press; 1997. [52] Chen Z, Yin J-J, Zhou YT, Zhang Y, Song L, Song M, et al. Dual enzyme-like activities of iron oxide nanoparticles and their implication for diminishing cytotoxicity. ACS Nano 2012;6:4001e12. [53] Li D, Tang J, Wei C, Guo J, Wang S, Chaudhary D, et al. Doxorubicin-conjugated mesoporous magnetic colloidal nanocrystal clusters stabilized by polysaccharide as a smart anticancer drug vehicle. Small 2012;8:2690e7. [54] Xu Y, Meng J, Meng L, Dong Y, Cheng Y, Zhu C. A highly selective fluorescencebased polymer sensor incorporating an (R, R)-salen moiety for Zn2þ detection. Chem Eur J 2010;16:12898e903. [55] Chandler J, Messer H, Ellender G. Cytotoxicity of gallium and indium ions compared with mercuric ion. J Dent Res 1994;73:1554e9. [56] Han J, Burgess K. Fluorescent indicators for intracellular pH. Chem Rev 2009;110:2709e28. [57] Liu J, Lu W, Reigada D, Nguyen J, Laties AM, Mitchell CH. Restoration of lysosomal pH in RPE cells from cultured human and ABCA4/ mice: pharmacologic approaches and functional recovery. Invest Ophthalmol Vis Sci 2008;49:772e80. [58] von Haefen C, Wieder T, Gillissen B, Stärck L, Graupner V, Dörken B, et al. Ceramide induces mitochondrial activation and apoptosis via a baxdependent pathway in human carcinoma cells. Oncogene 2002;21:4009e19. [59] Johansson A-C, Steen H, Öllinger K, Roberg K. Cathepsin D mediates cytochrome c release and caspase activation in human fibroblast apoptosis induced by staurosporine. Cell Death Differ 2003;10:1253e9.