Water-soluble ferrocene complexes (WFCs) functionalized silica nanospheres for WFC delivery in HepG2 tumor therapy

Materials Science & Engineering C 90 (2018) 397–406

Contents lists available at ScienceDirect

Materials Science & Engineering C journal homepage: www.elsevier.com/locate/msec

Water-soluble ferrocene complexes (WFCs) functionalized silica nanospheres for WFC delivery in HepG2 tumor therapy Saisai Yan, Fan Hu, Xia Hong, Qi Shuai

T

⁎

Shaanxi Key Laboratory of Natural Products & Chemical Biology, College of Chemistry & Pharmacy, Northwest A&F University, Yangling 712100, Shaanxi Province, PR China

A R T I C LE I N FO

A B S T R A C T

Keywords: Nanocarriers Water-soluble ferrocene complexes Silica-encapsulated nanospheres Drug delivery Cytotoxicity

Silica-encapsulated nanospheres of water-soluble ferrocene complexes WFCs@SiO2 and WFCs@SiO2@glutaraldehyde (GA) were first synthesized by a facile inverse-microemulsion method. The surface functional groups, particle size, and morphologies of nanospheres were characterized by IR spectra, UV–vis absorption spectra, dynamic light scattering (DLS) and SEM images. Single-crystal X-ray diffraction was used to confirm the molecular structure of free ferrocenyl-pyrazol ligand (L) and three WFCs, namely, [Ni(C22H14F6FeN4O4)(H2O)4] (5a), [Mg(C22H14F6FeN4O4)(H2O)4]·3H2O (5b), and [Ba(C22H14F6FeN4O4)(H2O)3] (5c). The electrochemical properties of 5a–5c were explored by cyclic voltammetry. The WFCs-loading capacities of 5a–5c in WFCs@SiO2 were found to be 38.4, 38.2, and 38.1 μg/mg, respectively. Cell studies under two drug delivery modes (free diffusion and endocytosis) were carried out by MTT cell-survival assays and morphological observation of HepG2 cells. It's interesting that the cytotoxicity of WFCs against HepG2 was increased by applying silica nanocarriers. Compared to WFCs@SiO2, the modification of GA on the spherical surface provided not only the better water-dispersity but also additional functional groups for further modification of other pharmacophores. The novel nanocarrier system for WFC delivery present a novel concept-of-proof method to protect varieties of affordable metal-based anticancer agents in physiological conditions and provided experimental basis for future studies focusing on drug delivery of other WFCs.

1. Introduction Nowadays, although the overall mortality of cancer declines due to remarkable breakthroughs made in diagnosis and treatment, cancer is still one of the leading causes of death with a high rate [1]. In addition to radiation and surgical intervention, current cancer treatments rely heavily on cisplatin-based chemotherapy, which has been used for more than 30 years [2,3]. However, the inefficiency on platinum-resistant tumors and severe side effects of these costly cisplatin-based treatments has become a considerable challenge for cancer therapy. In this situation, alternative metal-based anticancer medicines should be developed. Among the non-platinum antitumor agents, ferrocene complexes have recently received increasing attention due to their advantages of hypotoxicity, ease of modification and good redox properties [4,5]. A large number of therapeutic agents which incorporate ferrocene as pharmacophore on the scaffold of biologically active molecule have been found to be one of the most promising antitumor candidates [6–8]. In 2016, Hussain's group designed a ferrocenyl neodymium(III) complex, which shows remarkable photocytotoxicity in cancer cells

⁎

superior to those without the ferrocenyl moiety [9]. Arambula and coworkers reported a series of ferrocenylated N-heterocyclic carbene supported gold(I) complexes as potential anticancer agents in A549 lung cancer cells [10]. Two main anticancer mechanisms were suggested for the notable anticancer advances exhibited by these ferrocene complexes. The first mechanism is that ferrocene unit undergoes a fast and reversible oxidation process and turns into corresponding ferrocenium cation (Fc2+ → Fc3+). This oxidation process can catalyze the production of reactive oxygen species (ROS) and generate oxidative DNA damage [11]. The second mechanism is that ferrocene unit can disturb the normal functioning of the mechanism compromising cell viability and induce tumor cell cycle arrest in some phases and subsequent apoptosis [12]. Nevertheless, further clinical applications of ferrocene complexes in cancer therapy have been hindered by the low solubility and low stability in physiological conditions, which are common problems facing by most metal complexes [13,14]. Much attention should be given to explore more novel water-soluble ferrocene complexes-based drugs with good biocompatibility. Recent advances in nanocarriers, as significant application of

Corresponding author. E-mail address: [email protected] (Q. Shuai).

https://doi.org/10.1016/j.msec.2018.04.079 Received 19 September 2017; Received in revised form 11 March 2018; Accepted 25 April 2018 Available online 30 April 2018 0928-4931/ © 2018 Elsevier B.V. All rights reserved.

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

particle size of nanospheres was analyzed on a Delsa™ Nano system (Beckman Coulter). Scanning electron microscopy (SEM) images were obtained using an S-4800 instrument (Hitachi Ltd.) with the accelerating voltage of 10.0 kV.

nanotechnology, offer an alternative approach to resolve many of challenges in developing metal-based anticancer medicines with high drug-loading ability, low toxicity, and controlled-release properties [1,15]. Additionally, nanocarriers can protect metal complexes drugs from adverse environments, enhance permeability and retention (EPR) effect, improve the delivery of poorly soluble drugs, and lead specific accumulation in targeted cells/tissues, thereby achieving high treatment efficiency with low side effects [14,16]. The development of multifunctional nanocarriers for drug delivery with enhanced anticancer effects has been in the forefront of current cancer therapies and drug formulations. To date, various materials such as micelles, polymers, metal particles, carbon nanotubes, and silica nanoparticles have been used to fabricate the nanocarriers for drug delivery [17–22]. Among these materials, encapsulation with silica to protect metal complexes drugs from adverse environments in physiological conditions has been well developed because of the high biocompatibility, chemical stability, tunable functionalization and low-toxicity degradation pathways of silica material [23,24]. In 2016, Lin and coworkers reported a multifunctional nanocarrier system based on mesoporous silica for targeted delivery of a ruthenium(II) complex, whose further clinical application is limited by low aqueous solubility and great cytotoxicity to normal cell lines [25]. This system dramatically enhanced the anticancer efficacy of ruthenium(II) N-heterocyclic carbene complex and decreased its cytotoxicity to normal cells. Inspired by the recent contributions on ferrocene complexes and silica nanocarriers for cancer therapy, we firstly designed and synthesized one ferrocenyl-pyrazol ligand (L) with acetic acid moiety. Subsequently, corresponding water-soluble ferrocene complexes (WFCs) were prepared by solvent evaporation. Single-crystal X-ray diffraction was used to confirm the molecular structure of free ligand L and three synthesized ferrocene complexes (5a–5c). The electrochemical properties of 5a–5c were explored by cyclic voltammetry. Importantly, three WFCs acted as potential anticancer agents were further encapsulated in silica to form the multifunctional nanospheres (WFCs@SiO2 and WFCs@SiO2@glutaraldehyde (GA)) through a facile inverse-microemulsion method (Fig. 1). The positive charge of the surface and spherical morphology of silica-encapsulated nanospheres can partly enhance the selective uptake by endocytosis in cancer cells and thereby improve their anticancer efficacy. In particular, the obtained WFCs@SiO2 nanospheres were further cross-linked by GA to increase water-dispersity and provide additional functional groups on the spherical surface for further modification of other pharmacophores. Finally, a series of cell studies under two different drug delivery modes (free diffusion and endocytosis) was carried out on their cytotoxicity by both MTT cell-survival assays and morphological observation of HepG2 cells. With the use of this presented strategy on investigating nanocarriers for WFC delivery, other metal-based antitumor agents can also be encapsulated in silica nanospheres and undergo further modification to achieve higher anticancer efficacy.

2.2. Preparation of the ligand L The ligand L was synthesized according to the procedure described in Fig. 2. (i) The synthesis of compound 1 and 2 was performed according to the reported literature [26]. (ii) Synthesis of compound 3: To a solution of 2 (460.0 mg, 1.0 mmol) in 20 mL ethanol, hydrazine hydrate (0.146 mL, 3.0 mmol) was added dropwise over 15 min and then the mixture was refluxed for 6 h. After the reaction mixture was cooled to room temperature, excess hydrazine hydrate and solvent was removed under reduced pressure. The product was obtained as an orange powder after purification by column chromatography over silica gel with eluents of petroleum ether/ethyl acetate (4:1). Yield: 32% (145.3 mg). Found: 1H NMR (500 MHz, DMSO): δ = 13.45 (s, 2H, NH), 6.54 (s, 2H, CCHC), 4.77 (s, 4H, CHFc), 4.33 (s, 4H, CHFc). (iii) Synthesis of compound 4: Powdered 3 (454.0 mg, 1.0 mmol) and tBuOK (280.5 mg, 2.5 mmol) was dissolved in CH3CN (25 mL) and then ethyl bromoacetate (0.333 mL, 3 mmol) was added dropwise. The mixture was stirred and refluxed for 6 h. Then the solvent was removed under reduced pressure. Pure compound, as reddish-orange oil, was obtained by column chromatography with eluents of petroleum ether/ethyl acetate (6:1). Yield: 54% (338.0 mg). Found: 1H NMR (500 MHz, CDCl3): δ = 6.55 (s, 2H, CCHC), 4.92 (s, 4H, CCH2N), 4.42 (t, 4H, CHFc), 4.38 (t, 4H, CHFc), 4.22 (m, 4H, CH3CH2OOC), 1.26 (t, 6H, CH3CH2OOC). (iv) Synthesis of compound 5 (L): The above-mentioned ester 4 (0.626 g, 1.0 mmol) was dissolved in H2O (15 mL), and then NaOH (0.12 g, 3 mmol) was added. The mixture was stirred and refluxed for 5 h in dark. The obtained suspension was cooled down to room temperature and the black precipitate was removed by filtration. Then, the aqueous solution was acidified by the addition of 1 M HCl solution under vigorous stirring. The precipitated brown product was obtained by filtration. Yield: 80% (456.0 mg). Found: 1H NMR (500 MHz, DMSO): δ = 6.79 (s, 2H, CCHC), 5.04 (s, 4H, CCH2N), 4.69 (s, 4H, CHFc), 4.42 (s, 4H, CHFc). IR (cm−1): 3421 (m), 1726 (vs), 1633 (m), 1405 (s), 1079 (s), 936 (s), 814 (vs), 461 (s). 2.3. Preparation of the complex 5a–5c 2.3.1. [Ni(C22H14F6FeN4O4)(H2O)4] 5a Ligand L (285.0 mg, 0.5 mmol) and NaOH (40.0 mg, 1.0 mmol) were dissolved in 10 mL distillated water with stirring until absolutely dissolved, and then mixed with a solution of NiCl2·6H2O (118.8 mg, 0.5 mmol) in H2O (10 mL). Finally, the reaction mixture was adjusted to pH 7 with 0.1 mol·L−1 NaOH solution and heated to 90 °C for 10 h. After cooled to room temperature, the suspension was filtered to remove any undissolved reactants and the obtained filtrate was left to slowly evaporate water in dark at room temperature. Orange transparent crystals suitable for X-ray crystallographic analysis could be obtained after a period of about one month. Yield: 84% (293.6 mg). Found: 1H NMR (500 MHz, DMSO): δ = 6.56 (s, 2H, CCHC), 4.74 (s, 4H, CCH2N), 4.57 (s, 4H, CHFc), 4.27 (s, 4H, CHFc). IR (cm−1): 3417 (vs), 1624 (s), 1513 (m), 1387 (m), 1131 (s), 1081 (s), 979 (m), 810 (m), 466 (s). Anal. calcd for C22H22F6NiFeN4O8 (699.0): C, 37.77; H, 3.15; N, 8.01; Found: C, 37.82; H, 3.34; N, 8.09%.

2. Experimental section 2.1. Materials and general methods All reagents and solvents were obtained from commercial sources and used without further purification unless specified. Elemental analyses (C, H and N) were performed on a Vario MACRO cube elemental analyzer. NMR spectra were recorded on a Bruker AVANCE III 500 MHz spectrometer using the deuterated solvent of CDCl3 and d6-DMSO as internal standard. IR spectra were recorded in the range 4000–400 cm−1 using KBr pellets on a BRUKER TENSOR 27 spectrophotometer. Thermogravimetric analysis (TGA) was operated on SHIMADZU TA-60ws under N2 conditions from room temperature to 1000 °C with a heating rate of 10 °C·min−1. UV–vis absorption spectra of solutions were recorded from 300 to 800 nm at room temperature with a Shimadzu UV-1750 spectrophotometer. The zeta potential and

2.3.2. [Mg(C22H14F6FeN4O4)(H2O)4]·3H2O 5b The synthetic procedure was similar to that of 5a except that 398

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

Fig. 1. Schematic illustration of the preparation of silica-encapsulated nanospheres WFCs@SiO2 and WFCs@SiO2@GA, and the proposed mechanism of WFC delivery in HepG2 cells under two different drug delivery modes (free diffusion and endocytosis).

2.4. Preparation of the nanospheres

MgCl2·6H2O (101.7 mg, 0.5 mmol) was used instead of NiCl2·6H2O. The filtered solution was kept in beakers covered with a porous paper at a dust-free atmosphere. Red-brown block crystals suitable for the X-ray diffraction were harvested after a period of about one month. Yield: 85% (305.4 mg). Found: 1H NMR (500 MHz, DMSO): δ = 6.53 (s, 2H, CCHC), 4.74 (s, 4H, CCH2N), 4.59 (s, 4H, CHFc), 4.28 (s, 4H, CHFc). IR (cm−1): 3403 (vs), 1611 (s), 1511 (m), 1371 (vs), 1131 (m), 1053 (s), 982 (m), 823 (m), 479 (m). Anal. calcd for C22H28F6MgFeN4O11 (718.64): C, 36.74; H, 3.90; N, 7.79; Found: C, 36.52; H, 4.02; N, 7.53%.

2.4.1. Silica-encapsulated ligand (L@SiO2) and complexes nanospheres (WFCs@SiO2: 5a@SiO2, 5b@SiO2 and 5c@SiO2) Under vigorous stirring the ligand (water-soluble sodium salt) or complexes acted as potential anticancer agents were dissolved in H2O (0.5 mL), resulting in the aqueous mixture (0.05 M). Then 12 mL of a premixed solution of 1:1:4 (v/v/v) Triton X-100, hexyl alcohol and cyclohexane were added to the above aqueous phase. The mixture was sonicated for 20 min until it became transparent (the water-in-oil microemulsions were formed) and kept in a vial under vigorous stirring for another 12 h. Then tetraethyl orthosilicate (TEOS, 100 μL) and 3aminopropyltrimethoxysilane (APS, 100 μL) were slowly added into the above mixture solution followed by the addition of ammonium hydroxide solution (60 μL, 2 M). After kept for 24 h at room temperature, the microemulsion was destroyed by acetone (5 mL). The product of silica-encapsulated nanospheres was collected by centrifugation at 8000 rpm for 10 min and thoroughly washed by MeOH and H2O. To estimate the drug-loading capacity of WFCs@SiO2, the absorbance of solution was measured by UV–vis analysis at 450 nm after the removal of silica with NaOH etching.

2.3.3. [Ba(C22H14F6FeN4O4)(H2O)3] 5c The synthetic procedure was similar to that of 5a except that BaCl2·2H2O (122.2 mg, 0.5 mmol) was used instead of NiCl2·6H2O. The subsequent filtered solution was placed in beakers covered with a porous paper and left to slowly evaporate water in dark at room temperature, giving orange crystals that were suitable for X-ray crystallographic analysis after a period of about one month. Yield: 81% (307.6 mg). Found: 1H NMR (500 MHz, DMSO): δ = 6.53 (s, 2H, CCHC), 4.82 (s, 4H, CCH2N), 4.66 (s, 4H, CHFc), 4.31 (s, 4H, CHFc). IR data (cm−1): 3447 (vs), 1601 (s), 1510 (m), 1383 (m), 1132 (s), 1080 (m), 983(m), 812(m), 462 (m). Anal. calcd for C22H20F6BaFeN4O7 (759.56): C, 34.76; H, 2.63; N, 7.37; Found: C, 34.92; H, 2.59; N, 7.44%.

2.4.2. Glutaraldehyde-cross-linked L@SiO2@GA and WFCs@SiO2@GA nanospheres (WFCs@SiO2@GA: 5a@SiO2@GA, 5b@SiO2@GA and 5c@ SiO2@GA) The collected L@SiO2 and WFCs@SiO2 nanospheres were uniformly cross-linked between the amino groups on the surface of nanospheres with GA. Briefly, 40 mg of as-synthesized L@SiO2 or WFCs@SiO2 nanospheres were first dispersed in 1.5% (v/v) acetic acid solution 399

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

Fig. 2. The synthetic route of ligand L (compound 5).

geometrically. Structural plots were generated with Diamond. Table S1 shows a summary of crystallographic data and structure processing parameters of L and 5a–5c. The selected bond lengths and angles for L and 5a–5c are given in Table S2. CCDC 1543359 (for L), 1543355 (for 5a), 1543356 (for 5b), and 1543357 (for 5c) contain the supplementary crystallographic data for this paper. These data are provided free of charge by the Cambridge Crystallographic Data Centre.

(50 mL) by sonication for 20 min. Next, such aqueous suspension was mixed with GA (800 μL, 25%). The resulting mixture was left to react at room temperature overnight under vigorous stirring to produce the orange glutaraldehyde-cross-linked nanospheres, which were washed with water for several times and separated by centrifugation at the speed of 8000 rpm for 10 min. 2.4.3. Glutaraldehyde-cross-linked unloaded SiO2 nanospheres (SiO2@GA) Monodisperse pure SiO2 nanospheres with sizes ranging from 100 to 200 nm were synthesized by hydrolyzing TEOS in an alcoholic medium in the presence of water and ammonia employing a modified Stöber method according to the published procedures [27]. To modify the SiO2 with amino groups on the surface, the obtained nanospheres were modified by APS [28]. Generally, 0.25 g of SiO2 was homogeneously dispersed in 30 mL of methylbenzene at 100 °C for 2 h and stirred using a magnetic stir bar. Subsequently, 0.5 mL of APS was added to the SiO2 suspension and slowly stirred for 24 h at 120 °C. The treated nanoparticles were collected by centrifugation at the speed of 8000 rpm for 10 min and extensively washed with methylbenzene to remove the unreacted APS. Finally, the collected aminated SiO2 nanospheres (SiO2−NH2) were further cross-linked with GA to produce the unloaded SiO2@GA. The procedure was similar to the preparation of WFCs@SiO2@GA.

2.6. Electrochemical study Electrochemical measurements of the synthesized compounds were conducted at room temperature in a conventional three electrode cell consisting of a platinum wire as auxiliary electrode, a glassy carbon working electrode (3.0 mm diameter) and an Ag/Ag+ electrode containing a solution of 0.01 M AgNO3 and 0.1 M tetrabutylammonium perchlorate (n-Bu4NClO4) in N,N-dimethylformamide (DMF) as reference electrode on an electrochemical workstation CHI-660E. Cyclic voltammograms of the investigated compounds were recorded in DMF containing n-Bu4NClO4 (0.1 M) as the supporting electrolyte. In order to ensure the accuracy of the potential measure, ferrocenium/ferrocene redox couple (Fc+/Fc) of ferrocene molecule (Fc) was introduced as an internal standard before and after each set of experiments, but all potentials in the paper are referenced to the reference electrode [29]. The working electrodes in all experiments were polished, cleaned, washed with water and DMF, and then dried under N2 before use. Solutions containing the investigated compounds were purged with high purity argon for at least 15 min before measurement for complete removal of O2. Then the solution was protected from air during the experiment.

2.5. Crystal structure determination Crystallographic measurements were performed on a Bruker SMART CCD area-detector diffractometer at room temperature with graphite monochromatized Mo Kα radiation (λ = 0.71073 Å) in ω scan mode, and the data reduction was collected using Bruker SAINT. The crystal structures were resolved using SHELXL-97 by direct methods and refined on F2 with anisotropic thermal parameters by full-matrix leastsquares technique. Unless stated otherwise, all non-hydrogen atoms were refined anisotropically while the hydrogen atoms were generated

2.7. Cytotoxicity testing Cytotoxicity of the ligand, complexes and prepared nanospheres was evaluated in HepG2 cell lines by conventional methyl thiazole tetrazolium (MTT) cell-survival assay [30]. In a typical experiment, cells 400

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

Fig. 3. The coordination environment and coordination polyhedron of Ni(II) ion in complex 5a (A), Mg(II) ion in complex 5b (B) and Ba(II) ion in complex 5c (C) (symmetry codes for A: 1 − x, 2 − y, −z; B: −1 + x, y, z).

were seeded in 96 well microplate at a density of 5 × 103 cells per well in 1640 medium (100 μL) and grew for 24 h at 37 °C. Subsequently, cells were incubated with the corresponding solutions at different concentrations. After a period of time, the cells were washed and the fresh medium containing 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide was added into each plate, followed by incubation for another 4 h. After that, the medium containing MTT was removed, the resulting formazan was dissolved in dimethylsulphoxide (DMSO, 100 μL) and the plates were gently shaken for 10 min to dissolve the formazan crystals. Finally, the absorbance at 570 nm was recorded with a microplate reader. The relative cell viability was calculated according to the following equation:

3. Results and discussion 3.1. Spectroscopic studies of the ferrocene complexes In order to elucidate the structural information of the synthesized compounds in detail, extensive spectral studies like IR, 1H NMR and UV–vis absorption spectra were recorded. The IR spectra of complexes exhibited a strong signal approximately around 3400–3450 cm−1 corresponding to the existence of coordinated water. By comparing of the IR spectra in the region of v(C]O), it was found that the characteristic absorption of C]O stretching vibration of complexes red-shifted from 1726 (L) to 1624 (5a), 1611 (5b) and 1601 (5c) cm−1, respectively. It was also found that three strong bands due to the ferrocenyl moiety in the spectrum of ligand L (1105, 814 and 461 cm−1). The corresponding frequencies of ferrocene units in complexes, appearing nearly at the same position, showed that the cyclopentadienyl ring was not directly coordinated to the metal center [31]. The 1H NMR spectrum of the ligand L showed four important resonance signals at 6.79, 5.04, 4.69 and 4.42 ppm. These signals were attributed to two pyrazole rings, methylene protons and disubstituted cyclopentadienyl (Cp) rings of ferrocene, respectively. In the 1H NMR spectra of complex 5a–5c, obvious shifts of these proton signals can be observed in comparison with the analogous resonances of free ligand L. An examination of the NMR data revealed that protons of complexes showed larger upfield shifts with reference to the analogous protons present in the ligand L. Fig. S1 represented a clear example of the aforementioned spectral features.

Relative cell viability (%) = 100 × (ODtest − OD0)/(ODcontrol − OD0) ODtest was the optical density of the cell solution cultured with ligand, complexes or prepared nanospheres. ODcontrol was the optical density of the cell solution in the absence of ligand, complexes and prepared nanospheres. OD0 was the optical density of the solution containing cells without MTT. Each experiment was performed in triplicate for each concentration. The results expressed here are from at least three such independent experiments. The morphology images of HepG2 cells were undertaken on an Olympus CKX41 fluorescence microscope.

401

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

3.2.4. Structural description for complex 5c Complex 5c crystallizes in the triclinic system with P-1 space group. The asymmetric unit has half of a Ba(II) ion, one completely deprotonated ligand and three coordinated water molecules. Coordination environment and coordination polyhedron of Ba(II) are presented with atom numbering scheme in Fig. 3C. The eight-coordinated Ba(II) ion lies on the distorted dodecahedral geometry formed by five oxygen atoms from three ligands and three oxygen atoms from coordinated water molecules. The average Ba−O(L) distance for oxygen atoms from ligand is 2.594 Å with the range of 2.476(7)–2.669(7) Å, while the average BaeO(w) distance is 2.592 Å with the range of 2.563(10)–2.611(10) Å. The bond angles around Ba vary in the broad ranges of 47.9(2)–99.4(4)° and 107.9(3)–152.8(2)°. With regard to the three ferrocenyl groups (Fe, FeA, FeB), they are parallel to each other due to the same spatial orientation. They are found to exist almost exactly in the eclipsed conformation with the same torsion angles (2.351°) of the Cp rings.

The UV–vis absorption spectra of Fc, ligand L and complexes 5a–5c in DMF are displayed in Fig. S2. Ferrocene molecule (Fc), which acted as another reference, was measured in order to ensure the accuracy of the experiments and better investigate the chromophore of ferrocene unit. The absorption bands of ligand and complexes displayed a maximum absorption around 287 nm and intense absorption shoulders in the range of 345–360 nm, which were ascribed to the π-π* electron transitions. In addition to these band, another weaker absorption band was visible at around 450 nm, which assigned to another localized excitation with a lower energy produced by two nearly degenerate transitions, by a Fe(II) d-d transition, or by a metal-ligand charge transfer process, consistent with the values reported in the literature for analogous ferrocene containing species [32]. It can be seen that coordination with metal ions slightly shifts the absorption position of the ligand, but the absorption intensity of the peak changes obviously. 3.2. Crystal structures

3.3. Solubility and thermal stability of the synthesized compounds

3.2.1. Structural description for ligand L Single-crystal X-ray diffraction is used to confirm the molecular structure of free ligand L as well as the complex 5a–5c. Orange transparent crystals suitable for X-ray structure analyses of ligand L were obtained from slow evaporation of a CH3OH/H2O mixture (50:50, v/v) at room temperature and crystallized in the orthorhombic system with Pnna space group. The crystal structure of free ligand L is shown in Fig. S3. The distance between the Cp rings is determined at 3.307 Å. The torsion angle (2.17° twist about the edge joining the ring centroids) of Cp rings of ferrocene moiety, as measured with the dihedral angle of 3.31°, shows that the two Cp rings existed in almost fully eclipsed form. Additional, the two external pyrazole units (interplanar angle with 15.38°) are not co-planar with the parent Cp rings.

The water-insoluble ligand L with the carboxylic acid moiety can turn into water-soluble sodium salt with the help of NaOH solution. The ligand L used in cell study and preparation of the nanospheres was processed by NaOH solution (0.1 M) in advance until it is completely dissolved in water. The synthesized complexes (5a–5c) were watersoluble and found to be soluble in PBS, DMF, DMSO, acetone, methanol and ethanol but insoluble in cyclohexane. Meanwhile, the complexes were found to be stable in PBS and DMF even after 48 h as evidenced from the absorption study on solution showing no noticeable spectral change. Thermogravimetric analysis (TGA) for ligand L and corresponding complex (5a–5c) was carried out under a N2 atmosphere at a heating rate of 10 °C·min−1, and the obtained results are shown in Fig. 4. The free ligand L has low thermostability and been decomposed completely before 400 °C. It is obvious that complexes possess better thermostability than free ligand L. Consequently, from the structure characteristics revealed from the crystal structure analysis, we can clearly see that the thermal stability is improved after coordinated with metal ions. Although the complexes have different weight loss ratio at the same temperature, they are similar in the process of thermal decomposition which occurs in three steps compared with ligand L. For ligand L, the decomposition process is simple with a sharp weight loss of 87.532% (calcd: 87.368%) from 30 to 367 °C, following the weight no longer change. For complex 5a, an initial weight loss of 10.364% from 44 to 258 °C is consistent with the release of coordinated water (calcd: 10.300%). Above 260 °C, the anhydrous frameworks start to collapse as a result of thermal decomposition. For complex 5b, the first weight loss from 35 to 274 °C is attributed to the release of coordinated water molecules and lattice water molecules (observed 14.515%, calcd 17.533%). The second weight loss of 22.980% and the third weight loss of 35.347% (observed 70.970%, calcd 71.140%) in the range of 274–980 °C correspond to the collapse of the frameworks. For complex 5c, the first weight loss from 30 to 227 °C is attributed to the loss of coordinated water molecules (observed 7.525%, calcd 7.109%). Above 230 °C, the anhydrous frameworks start to collapse attributed to the thermal decomposition. The initial decomposition temperature of the anhydrous frameworks of the compounds is in the order of 5b (274 °C) > 5a (260 °C) > 5c (230 °C) > L (204 °C).

3.2.2. Structural description for complex 5a Complex 5a crystallizes in the monoclinic system with C2/c space group. There are one Ni(II) ion, one completely deprotonated ligand and four coordinated water molecules in the asymmetric unit. As shown in Fig. 3A, the coordination environment and coordination polyhedron of Ni(II) are presented with atom numbering scheme. The central Ni(II) ion is six-coordinated by two carboxylate group oxygen atoms (O2 and O6) and four oxygen atoms from water molecules with the bond angles around Ni vary from 84.77(8)° to 179.55(12)°, resulting in a distorted octahedral geometry. The apical sites of the octahedron are occupied by the O2 and O4W with O2eNieO4W bond angle of 169.95(8)°, indicating that the distortion of the octahedral geometry is not very high. Four oxygen atoms (O6, O3W, O5W and O8W) from ligand and three coordinated water molecules consist of the equator base. The NieO bond lengths are in the range of 2.050(2)–2.088(2) Å. The average NieO(L) distance for two oxygen atoms from ligand is of 2.062(6) Å, this distance is shorter than the average value of NieO(w) bond distance (2.070(3) Å) from four coordinated water molecules. 3.2.3. Structural description for complex 5b Complex 5b crystallizes in the triclinic system with P-1 space group. The asymmetric unit has one Mg(II) ion, one completely deprotonated ligand, four coordinated water molecules and two lattice water molecules. As shown in Fig. 3B, the coordination environment and coordination polyhedron of Mg(II) are presented with atom numbering scheme. The coordination geometry of six-coordinated Mg(II) ion is comprised of two carboxylate group oxygen atoms (O2 and O7) and four oxygen atoms from water molecules with the bond angles around Mg vary from 78.83(16)° to 176.72(17)°, resulting in a similar distorted octahedral geometry to complex 5a. The MgeO bond lengths are in the range of 2.041(4)–2.111(4) Å. The average MgeO(L) distance for two oxygen atoms from ligand is 2.050(5) Å, this is shorter than the average value of MgeO(w) bond distance (2.090(3) Å) for four oxygen atoms from coordinated water molecules.

3.4. Electrochemistry of the synthesized compounds The electrochemical properties of ferrocenyl-pyrazol ligand L as well as corresponding complex 5a–5c were explored by cyclic voltammetry due to the fact that redox-activity is closely linked with their anticancer performance [9,10]. A summary of key electrochemical data for these compounds obtained in this work appears in Table S3. The 402

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

Fig. 4. TG curves of L, 5a, 5b and 5c.

Comparing the redox features of complexes with free ligand L, cyclic voltammetric studies of complex 5a–5c exhibit three similar well-separated processes labeled I, II and III as shown in Fig. 5. This interesting phenomenon indicates that coordination with different metal ions slightly change their redox properties of complexes. The quasireversible redox process I is assigned to ferrocene group, and formal potentials (E1/2 (n), n = 5a, 5b and 5c) are about 197, 189 and 183 mV, respectively. This process occurs at more positive potentials than found for the free ligand L. The potential shifts (294 mV for complex 5a, 286 mV for complex 5b, 280 mV for complex 5c) of the E1/2 values indicate that the coordination increases the electron withdrawing ability of ferrocene groups, making the complexes harder to oxidize than the free ligand. The other large quasi-reversible redox process II is likely associated with the redox of metal ion, the only redox-active center in the complex other than the ferrocene groups. The formal potentials (E1/2 (n), n = 5a, 5b and 5c) are about 556, 542 and 525 mV, respectively. The irreversible reduction process III with the corresponding peak potential of −824, −751 and − 715 mV is attributed to the reduction process of the pyrazolyl heterocycle.

Fig. 5. Cyclic voltammograms of 1 mmol·L−1 solution of L, 5a, 5b and 5c at the same scan rate of 50 mV·s−1 in DMF with 0.1 mol·L−1 n-Bu4NClO4 at room temperature.

3.5. Characterization of the nanospheres IR spectra, UV–vis absorption spectra, SEM images and photographs were used to confirm the successful synthesis of silica-encapsulated nanospheres. As shown in the IR spectrum of WFCs@SiO2 nanospheres (Fig. S5, black line), the bending vibration of NeH groups at the peaks of 1490 and 1561 cm−1 and the stretching vibration of CeH groups at the peaks of 2930 cm−1 were clearly observed, indicating the availability of amino groups on the surface [33]. In the IR spectrum of WFCs@SiO2@GA nanospheres (Fig. S5, red line), there are some characteristic peaks of formed pyridinium structures appear at 685, 1500, 1578 and 1645 cm−1, which are significant in comparison with those of WFCs@SiO2 nanospheres [34]. Similar comparing information was also observed in the UV–vis absorption spectra (Fig. 6A). WFCs@ SiO2@GA nanospheres (red curve) showed their featured peak only at around 279 nm without the peak at 218 nm after cross-linked with GA. In addition, the GA-involved cross-linked process with amino groups on the surface of WFCs@SiO2 nanospheres was accompanied by considerable reduction of absorptivity for the aqueous suspension. The morphologies of pure SiO2, WFCs@SiO2 and WFCs@SiO2@GA nanospheres were characterized by SEM images. Fig. 6C shows the SEM image of the WFCs@SiO2 nanospheres, whose sizes (100–200 nm) are almost the same as those of pure SiO2 nanospheres (Fig. 6B).

cyclic voltammograms of ligand L and complex 5a–5c are shown in Fig. 5. In order to compare the stability and reversibility of electrochemically oxidized products, cyclic voltammogram of Fc was taken at first in the same experimental conditions. It can be seen that the cyclic voltammogram of a 1 mM solution of Fc in DMF shows a typical oneelectron reversible oxidation process (E1/2 = −2 mV, Ipa/Ipc = 0.99, black line in Fig. S4), corresponding to the oxidation of ferrocene to ferrocenium ion (Fc2+ → Fc3+), with a peak-to-peak separation (ΔEp) of about 106 mV at a scan rate of 50 mV·s−1. The ligand L undergoes a quasi-reversible one-electron redox process labeled I (Fig. S4, red line), typical of a ferrocene derivative, assigned to the ferrocenium/ferrocene redox couple. In comparison with Fc, the half-wave potential (E1/ 2 = −97 mV) of ferrocene moiety in ligand L shifts to more negative potential because of the partial charge transfer from pyrazolyl group to the ferrocene caused by conjugation in the molecule, which means the ferrocene moiety is easier to oxidize. Additional, the other oxidation process II with a peak potential of 326 mV is observed when the potential is scanned to more positive values, which is associated with the pyrazolyl heterocycle in ligand L. No reduction wave of pyrazolyl heterocycle appears in the cathodic sweep. 403

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

Fig. 6. (A) UV–vis absorption spectra of aqueous suspensions of WFCs@SiO2 (black curve), and WFCs@SiO2@GA (red curve) nanospheres. SEM images of (B) pure SiO2 nanospheres (smooth surface), (C) WFCs@SiO2 nanospheres (rough surface), (D) WFCs@SiO2@GA nanospheres (villous spherical surface). Insets are their SEM images at high magnification. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

modification with GA can provide excellent water-dispersity for WFCs@SiO2 nanospheres. The aqueous suspensions of WFCs@SiO2 nanospheres in phosphate-buffered saline (PBS) have relatively poor water-dispersity, leading to the significant aggregation or sedimentation behaviors after setting overnight (Fig. S8). In contrast, the WFCs@ SiO2@GA nanospheres can be readily dispersed in PBS to form a homogeneous orange suspension, which is so stable that no evident precipitate appeared after setting overnight (Fig. S8), displaying highly water-dispersible performance. The excellent water-dispersity of WFCs@SiO2@GA nanospheres is largely attributed to the strong electrostatic repulsion caused by the increase of zeta potential (from 11.58 to 25.18 mV) after surface modification with GA. The particle size of WFCs@SiO2@GA measured by dynamic laser scattering (DLS) was relatively narrow, and the average particle size was estimated to be 300 nm (Fig. S9). The bigger particle size than the result revealed from SEM image was partly due to the mutual cross-linking of WFCs@ SiO2@GA nanospheres.

Nevertheless, compared with the smooth spherical surface of SiO2 nanospheres, more external surface roughness was observed for WFCs@ SiO2 nanospheres, displaying a unique appearance like leechee. The SEM images of WFCs@SiO2@GA nanospheres in Fig. 6D displayed a villous spherical surface which should be ascribed to the cross-linked with large amount of GA. Photographs of aqueous suspension of WFCs@SiO2 and pure SiO2 nanospheres before and after treatment with NaOH etching demonstrated that the WFCs have been successfully encapsulated in silica as evidenced from color contrast of resulting transparent solutions (Fig. S6). In order to further estimate the drug-loading capacity in WFCs@ SiO2 nanospheres, take complex 5b as example, the linear relationship between the optical absorbance data and drug concentrations by the UV–vis analysis at 450 nm were investigated firstly (Fig. S7). Then the absorbance of solution was measured after the removal of silica with NaOH etching. When the initial concentration of 5b for synthesis of nanospheres was 0.05 M, the drug-loading amount in 5b@SiO2 was estimated to be 38.2 μg/mg. Take complex 5a and 5c as examples, the drug-loading capacity was similar and the amounts in 5a@SiO2 and 5c@SiO2 were found to be 38.4 and 38.1 μg/mg, respectively.

3.7. In vitro cytotoxicity assays In this section, we firstly assessed the potential of WFCs dissolved directly in PBS (pH 7.4) solution as potential anticancer agents. Secondly, to determine the roles of silica nanocarriers for WFC delivery, control experiments were conducted on the silica-encapsulated nanospheres (WFCs@SiO2 and WFCs@SiO2@GA). Finally, in order to further quantitatively evaluate the cytotoxicity of unloaded nanospheres, the viability of HepG2 cells after 48 h treatment with SiO2@GA was tested as well under the same conditions. For the first drug delivery mode, WFCs which dissolved in PBS were directly introduced to the media and encountered with cells through free diffusion. Prior to testing, the solubility of each solution was evaluated in media at 37 °C over 72 h and no precipitation was observed. In order to assess their anticancer efficacy in HepG2 tumor

3.6. Influence of GA on WFCs@SiO2 nanospheres The glutaraldehyde-involved cross-linked process of WFCs@SiO2 nanospheres can contribute to the enhancement of functional groups on the spherical surface, resulting in novel WFCs@SiO2@GA nanospheres for further application and modification [33,35]. During the procedure of cross-linked with GA, the color of aqueous suspensions of the WFCs@ SiO2 nanospheres changes from original light yellow to orange. The corresponding color of solid product for WFCs@SiO2@GA nanospheres is red (inset of Fig. S5). The reaction mechanism of forming pyridinium structures between amines and GA has been reported to elucidate the dramatic color change in acidic water [34]. Additionally, surface 404

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

therapy, cell viabilities after 72 h incubation with different complexes were assessed using the well-known MTT assay. Key data were summarized in Table S4. It was found that Ba-containing complex 5c displayed greater inhibition of cell proliferation than 5a and 5b, giving IC50 values of 574.8 ± 7.4 μM. Complex 5a and 5b dissolved directly in PBS had low cytotoxicity to HepG2 cells. To evaluate whether the cytotoxicity could be attributed to the ligand L containing ferrocenyl moiety or not, we also tested the cytotoxicity of ligand alone under the same conditions. Interestingly, the free ligand L (water-soluble sodium salt) can obviously accelerate the proliferation as to the cancer cells instead of inhibition, while it can disturb the normal functioning mechanism. For the second drug delivery mode, synthesized ferrocene complexes acted as potential anticancer agents were encapsulated in silica to form multifunctional nanospheres (WFCs@SiO2 and WFCs@ SiO2@GA) at first. In this drug delivery mode, it is important to investigate the cytotoxicity of the vehicles. As the result of MTT assay of pure SiO2 nanospheres (Fig. S10), low cytotoxicity was observed even the concentration up to 300 μg/mL, suggesting their good biocompatibility. Furthermore, the ligand L (water-soluble sodium salt) was also encapsulated in silica to form L@SiO2 and L@SiO2@GA in order to compare with the complexes. Cytotoxicity was evaluated by both MTT assay and morphological observation with optical microscopy. The presence of amino groups conferred the surface of WFCs@SiO2 positively charged [28,36]. More importantly, cross-linked with GA cannot change the charged properties of WFCs@SiO2 according to the literature [34]. As the result of DLS measurements, the zeta potential of WFCs@SiO2 was 11.58 mV, which increased to 25.18 mV after surface modification with GA. The adhesion of the nanospheres to the cell surface was attributed to the electrostatic attraction of the positively charged WFCs@SiO2 and the negatively charged cytomembrane [19,37]. While it is a nontargeted system, the positive charge of the surface can partly enhance the selective uptake by endocytosis in cancer cells and thereby increase their cytotoxicity. The nanospheres were proposed to be internalized into the cytosol via cellular endocytosis, after that the encapsulated WFCs were released through lysosomal enzymatic metabolism (Fig. 1) [38]. The morphologies were also an important factor since the spherical morphology was one facile shape for efficient endocytosis [39]. As shown in Fig. 7, cell viabilities of the HepG2 cells after 48 h incubation with different concentrations (based on the total particle weight) of WFCs@SiO2 nanospheres were evaluated. After treatment with WFCs@SiO2, the cell viability was decreased, exhibiting a dosage-dependent toxic effect. L@SiO2 and 5b@

SiO2 have relatively low cytotoxicity with the increase of concentration when compared with 5a@SiO2 and 5c@SiO2. Comparing to phenomenon mentioned above in the first drug delivery mode of free diffusion, it was worth noting that the cytotoxicity of complexes delivered by silica-encapsulated nanospheres displayed a startling change. 5a@SiO2 exhibited great inhibition of cell proliferation, while complex 5a had low cytotoxicity when it was directly dissolved in PBS. The cytotoxicity of 5a was inconsistent under two different drug delivery modes. The 5a@SiO2 obtained from encapsulation of 5a with silica, which lead to the change of drug delivery mode from free diffusion to endocytosis, was likely to be responsible for the enhanced cytotoxicity. The presence of nanocarriers greatly increased the cytotoxicity of complex 5a. The optical microscopy image was shown in Fig. S11B to evaluate their influence on cell morphology. Cells incubation with 5a@SiO2 after 48 h at a concentration of 120 μg/mL separated from each other in a shrinking morphology and some dead cells were found in sight. The cytotoxicity of different nanospheres was in the order of 5a@ SiO2 > 5c@SiO2 > 5b@SiO2 > L@SiO2 within 120 μg/mL. Their distinct performance in cytotoxicity for different ferrocene complexes may mainly depend on the coordinated metal ions when they were delivered by silica nanocarriers. Moreover, the silica nanocarriers can improve the performance of encapsulated ferrocene complexes due to their good biocompatibility and cellular membrane-penetrating capacity. On the basis of the experimental evidence, the roles of nanocarriers for drug delivery should also be taken into consideration in the study of potential anticancer agents. Cell viabilities of HepG2 cells after 48 h incubation with different concentrations (based on the total particle weight) of SiO2@GA, L@ SiO2@GA and WFCs@SiO2@GA nanospheres were shown in Fig. 8. Equivalent amount of SiO2@GA under the same conditions as control was treated in order to better evaluate the cytotoxicity of nanospheres without complexes encapsulated. It was found that the SiO2@GA have low cytotoxicity, even maintaining high cell viability almost 60% even if the concentration up to 300 μg/mL. As shown in Fig. S11C, no obvious morphological change of HepG2 cells was observed. Take the cell viability of SiO2@GA as control group, both 5a@SiO2@GA and 5c@ SiO2@GA showed remarkable cytotoxicity. In contrast, the cell viability incubated with L@SiO2@GA remained high. Complex 5a and 5c delivered by WFCs@SiO2@GA showed promising inhibition of cells proliferation while 5b was weak. The cytotoxicity of individual WFC in WFCs@SiO2@GA exhibited no obvious change compared with WFCs@ SiO2, although a certain amount of GA was cross-linked on the surface. Interestingly, as shown in Fig. S11D, unique red signal from WFCs@

Fig. 7. Cell viabilities of HepG2 cells after 48 h incubation with different concentrations of L@SiO2 and WFCs@SiO2 nanospheres (WFCs@SiO2: 5a@SiO2, 5b@SiO2 and 5c@SiO2).

Fig. 8. Cell viabilities of HepG2 cells after 48 h incubation with different concentrations of SiO2@GA, L@SiO2@GA and WFCs@SiO2@GA nanospheres (WFCs@SiO2@GA: 5a@SiO2@GA, 5b@SiO2@GA and 5c@SiO2@GA). 405

Materials Science & Engineering C 90 (2018) 397–406

S. Yan et al.

SiO2@GA was observed in some cells after 24 h incubation under the normal light. After the cells were washed three times with fresh medium, this unique red signal remained the same. The massive accumulation of nanospheres was probably attributed to the endocytosis by HepG2 cells, suggesting that the cells internalize nanospheres.

[12]

[13]

4. Conclusions

[14]

Herein we reported two series of silica-encapsulated nanospheres (WFCs@SiO2 and WFCs@SiO2@GA) via a facile inverse-microemulsion method for WFC delivery. The synthesized ferrocene complexes (5a–5c) have been successfully encapsulated in silica to form the multifunctional nanospheres, thereby changing drug delivery mode and protecting them from the adverse environments in physiological conditions. The WFCs-loading capacities of 5a–5c in WFCs@SiO2 were found to be 38.4, 38.2, and 38.1 μg/mg, respectively. The zeta potential of WFCs@SiO2 was 11.58 mV, which increased to 25.18 mV after surface modification with GA. Single-crystal X-ray diffraction analyses revealed that both Ni(II) in 5a and Mg(II) in 5b were six-coordinated with octahedral arrangements, while Ba(II) in 5c was totally different from them with dodecahedral geometry in eight-coordinated. As the result of cell studies, it was worth noting that both 5a and 5c delivering by silica-encapsulated nanospheres displayed obvious cytotoxicity. Compared to WFCs@SiO2, the modification of GA on the spherical surface provided not only the better water-dispersity but also more functional groups for further modification of other pharmacophores. The novel nanocarrier system for WFC delivery presented a novel concept-of-proof method to protect varieties of affordable metal-based anticancer agents in physiological conditions and provided great potential to increase their anticancer efficacy.

[15] [16]

[17]

[18]

[19]

[20]

[21]

[22] [23] [24]

Acknowledgements [25]

This work was financially supported by the National Natural Science Foundation of China (grant numbers 21473136 and 21103140).

[26]

Appendix A. Supplementary data

[27]

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.msec.2018.04.079.

[28]

[29]

References

[30]

[1] T. Sun, Y.S. Zhang, B. Pang, D.C. Hyun, M. Yang, Y. Xia, Engineered nanoparticles for drug delivery in cancer therapy, Angew. Chem. Int. Ed. 53 (2014) 12320–12364. [2] N.J. Wheate, S. Walker, G.E. Craig, R. Oun, The status of platinum anticancer drugs in the clinic and in clinical trials, Dalton Trans. 39 (2010) 8113–8127. [3] A. Grozav, O. Balacescu, L. Balacescu, T. Cheminel, I. Berindan-Neagoe, B. Therrien, Synthesis, anticancer activity, and genome profiling of thiazolo arene ruthenium complexes, J. Med. Chem. 58 (2015) 8475–8490. [4] C. Ornelas, Application of ferrocene and its derivatives in cancer research, New J. Chem. 35 (2011) 1973–1985. [5] C.G. Hartinger, N. Metzler-Nolte, P.J. Dyson, Challenges and opportunities in the development of organometallic anticancer drugs, Organometallics 31 (2012) 5677–5685. [6] E. Hillard, A. Vessieres, L. Thouin, G. Jaouen, C. Amatore, Ferrocene-mediated proton-coupled electron transfer in a series of ferrocifen-type breast-cancer drug candidates, Angew. Chem. Int. Ed. 45 (2005) 285–290. [7] F.A. Larik, A. Saeed, T.A. Fattah, U. Muqadar, P.A. Channar, Recent advances in the synthesis, biological activities and various applications of ferrocene derivatives, Appl. Organomet. Chem. 31 (2017) 1–22. [8] C. Mu, S.W. Chang, K.E. Prosser, A.W. Leung, S. Santacruz, T. Jang, J.R. Thompson, D.T. Yapp, J.J. Warren, M.B. Bally, T.V. Beischlag, C.J. Walsby, Induction of cytotoxicity in pyridine analogues of the anti-metastatic Ru(III) complex NAMI-A by ferrocene functionalization, Inorg. Chem. 55 (2016) 177–190. [9] T. Sarkar, S. Banerjee, S. Mukherjee, A. Hussain, Mitochondrial selectivity and remarkable photocytotoxicity of a ferrocenyl neodymium(III) complex of terpyridine and curcumin in cancer cells, Dalton Trans. 45 (2016) 6424–6438. [10] J.F. Arambula, R. McCall, K.J. Sidoran, D. Magda, N.A. Mitchell, C.W. Bielawski, V.M. Lynch, J.L. Sessler, K. Arumugam, Targeting antioxidant pathways with ferrocenylated N-heterocyclic carbene supported gold(I) complexes in A549 lung cancer cells, Chem. Sci. 7 (2016) 1245–1256. [11] H.-Y. Zhou, M. Li, J. Qu, S. Jing, H. Xu, J.-Z. Zhao, J. Zhang, M.-F. He, Effective antitumor candidates based upon ferrocenylseleno-dopamine derivatives: growth

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38] [39]

406

inhibition by induction cell apoptosis and antivascular effects, Organometallics 35 (2016) 1866–1875. A. Sun, J. Lin, C. Pi, R. Xu, X. Cui, Biological evaluation of ferrocenyl olefins: cancer cell growth inhibition, ROS production, and apoptosis activity, Arch. Pharm. 349 (2016) 186–192. L. He, T. Chen, Y. You, H. Hu, W. Zheng, W.L. Kwong, T. Zou, C.M. Che, A cancertargeted nanosystem for delivery of gold(III) complexes: enhanced selectivity and apoptosis-inducing efficacy of a gold(III) porphyrin complex, Angew. Chem. Int. Ed. 53 (2014) 12532–12536. C.N. Lok, T. Zou, J.J. Zhang, I.W. Lin, C.M. Che, Controlled-release systems for metal-based nanomedicine: encapsulated/self-assembled nanoparticles of anticancer gold(III)/platinum(II) complexes and antimicrobial silver nanoparticles, Adv. Mater. 26 (2014) 5550–5557. H. Zazo, C.I. Colino, J.M. Lanao, Current applications of nanoparticles in infectious diseases, J. Control. Release 224 (2016) 86–102. L. He, Y. Huang, H. Zhu, G. Pang, W. Zheng, Y.-S. Wong, T. Chen, Cancer-targeted monodisperse mesoporous silica nanoparticles as carrier of ruthenium polypyridyl complexes to enhance theranostic effects, Adv. Funct. Mater. 24 (2014) 2754–2763. Y. Gao, Y. Li, Y. Li, L. Yuan, Y. Zhou, J. Li, L. Zhao, C. Zhang, X. Li, Y. Liu, PSMAmediated endosome escape-accelerating polymeric micelles for targeted therapy of prostate cancer and the real time tracing of their intracellular trafficking, Nanoscale 7 (2015) 597–612. S. Ruan, X. Cao, X. Cun, G. Hu, Y. Zhou, Y. Zhang, L. Lu, Q. He, H. Gao, Matrix metalloproteinase-sensitive size-shrinkable nanoparticles for deep tumor penetration and pH triggered doxorubicin release, Biomaterials 60 (2015) 100–110. N. Wang, Y. Feng, L. Zeng, Z. Zhao, T. Chen, Functionalized multiwalled carbon nanotubes as carriers of ruthenium complexes to antagonize cancer multidrug resistance and radioresistance, ACS Appl. Mater. Interfaces 7 (2015) 14933–14945. T. Yin, L. Wang, L. Yin, J. Zhou, M. Huo, Co-delivery of hydrophobic paclitaxel and hydrophilic AURKA specific siRNA by redox-sensitive micelles for effective treatment of breast cancer, Biomaterials 61 (2015) 10–25. X. Yao, X. Niu, K. Ma, P. Huang, J. Grothe, S. Kaskel, Y. Zhu, Graphene quantum dots-capped magnetic mesoporous silica nanoparticles as a multifunctional platform for controlled drug delivery, magnetic hyperthermia, and photothermal therapy, Small 13 (2017) 1602225–1602236. S. Giret, M. Wong Chi Man, C. Carcel, Mesoporous-silica-functionalized nanoparticles for drug delivery, Chemistry 21 (2015) 13850–13865. V. Mamaeva, C. Sahlgren, M. Linden, Mesoporous silica nanoparticles in medicine—recent advances, Adv. Drug Deliv. Rev. 65 (2013) 689–702. N.Ž. Knežević, V. Stojanovic, A. Chaix, E. Bouffard, K.E. Cheikh, A. Morère, M. Maynadier, G. Lemercier, M. Garcia, M. Gary-Bobo, J.-O. Durand, F. Cunin, Ruthenium(ii) complex-photosensitized multifunctionalized porous silicon nanoparticles for two-photon near-infrared light responsive imaging and photodynamic cancer therapy, J. Mater. Chem. B 4 (2016) 1337–1342. G. Lv, L. Qiu, G. Liu, W. Wang, K. Li, X. Zhao, J. Lin, pH sensitive chitosan-mesoporous silica nanoparticles for targeted delivery of a ruthenium complex with enhanced anticancer effects, Dalton Trans. 45 (2016) 18147–18155. M. Veronelli, S. Dechert, S. Demeshko, F. Meyer, 1,1′-Bis(pyrazol-3-yl)ferrocene: a clip ligand that forms supramolecular aggregates and prismatic hexanuclear coinage metal complexes, Inorg. Chem. 54 (2015) 6917–6927. Y.-S. Lin, C.L. Haynes, Impacts of mesoporous silica nanoparticle size, pore ordering, and pore integrity on hemolytic activity, J. Am. Chem. Soc. 132 (2010) 4834–4842. S. Yang, X. Feng, S. Ivanovici, K. Mullen, Fabrication of graphene-encapsulated oxide nanoparticles: towards high-performance anode materials for lithium storage, Angew. Chem. Int. Ed. 49 (2010) 8408–8411. J.H.K. Yip, J. Wu, K. Wong, K. Yeung, J.J. Vittal, Syntheses, structures, and electrochemistry of polynuclear CuI, AgI, and PtII complexes bearing ferrocenyl group, Organometallics 21 (2002) 1612–1621. Y. Chang, K. Yang, P. Wei, S. Huang, Y. Pei, W. Zhao, Z. Pei, Cationic vesicles based on amphiphilic pillar[5]arene capped with ferrocenium: a redox-responsive system for drug/siRNA co-delivery, Angew. Chem. Int. Ed. 53 (2014) 13126–13130. S. Yan, L. Jing, Q. Shuai, Mn(II) and Co(II) complexes functionalized by imidazolylferrocene: electrochemical property and different supramolecular architectures driven by intriguing hydrogen-bonded system, Polyhedron 126 (2017) 252–261. T. Romero, R.A. Orenes, A. Tárraga, P. Molina, Preparation, structural characterization, electrochemistry, and sensing properties toward anions and cations of ferrocene-triazole derivatives, Organometallics 32 (2013) 5740–5753. X. Yao, Z. Tian, J. Liu, Y. Zhu, N. Hanagata, Mesoporous silica nanoparticles capped with graphene quantum dots for potential chemo-photothermal synergistic cancer therapy, Langmuir 33 (2017) 591–599. K. Wang, X. Yuan, Z. Guo, J. Xu, Y. Chen, Red emissive cross-linked chitosan and their nanoparticles for imaging the nucleoli of living cells, Carbohydr. Polym. 102 (2014) 699–707. K. Wang, X. Zhang, X. Zhang, X. Fan, Z. Huang, Y. Chen, Y. Wei, Preparation of biocompatible and photostable PEGylated red fluorescent nanoparticles for cellular imaging, Polym. Chem. 6 (2015) 5891–5898. Y. Tang, H. Hu, M.G. Zhang, J. Song, L. Nie, S. Wang, G. Niu, P. Huang, G. Lu, X. Chen, An aptamer-targeting photoresponsive drug delivery system using “off-on” graphene oxide wrapped mesoporous silica nanoparticles, Nanoscale 7 (2015) 6304–6310. B. Yu, Y. Zhang, W. Zheng, C. Fan, T. Chen, Positive surface charge enhances selective cellular uptake and anticancer efficacy of selenium nanoparticles, Inorg. Chem. 51 (2012) 8956–8963. P.T. Wong, S.K. Choi, Mechanisms of drug release in nanotherapeutic delivery systems, Chem. Rev. 115 (2015) 3388–3432. C. Xu, D. Yang, L. Mei, B. Lu, L. Chen, Q. Li, H. Zhu, T. Wang, Encapsulating gold nanoparticles or nanorods in graphene oxide shells as a novel gene vector, ACS Appl. Mater. Interfaces 5 (2013) 2715–2724.