Facile preparation of magnetic metal organic frameworks core–shell nanoparticles for stimuli-responsive drug carrier

Ferric chloride, Ferric chloride hexahydrate, Ethylene glycol (EG), Sodium acetate, Ethylene diamine and Dimethyl formamide (DMF) were purchased from Sinopharm Chemical Reagent Co., Ltd 2-Aminoterephthalic acid was purchased from Shanghai Cablebridge biotechnology Co., Ltd Doxorubicin hydrochloride (DOX) was purchased from Wuhan Yuancheng Technology Co., Ltd. Other reagents and materials used were commercially available without any purification. The water used in the experiments was 18.2 MΩ cm⁻¹ ultrapure water.

Fe₃O₄ nanoparticles were prepared according to the method reported by Peng et al [21]. 2.03 g FeCl₃, 3.00 g NaAc and 12.0 ml Ethylene diamine were dissolved in 50.0 ml EG at 70 °C to form a brown solution. The mixed solution was transferred into a 100 ml teflon-lined stainless-steel reaction kettle and maintained at 205 °C for 10 h. Then, the solution was cooled to room temperature. Later, the black magnetic nanoparticles were collected by a magnet and washed several times by water to remove salt. After that, the products were dispersed in a constant volume of water for further use.

The MOFs was prepared according to the method reported by Patricia Horcajada [5] with some modifications. Specifically, 0.13 g FeCl₃ · 6H₂O (0.5 mmol) was dissolved in 50 ml water, and 0.09 g 2-aminoterephthalic acid (0.5 mmol, BDC-NH₂) was dispersed in 50 ml water by ultrasonic for 5 min. Then, Fe₃O₄ suspension solution, FeCl₃ solution and BDC-NH₂ dispersion solution were mixed in a round-bottom flask and ultrasound for another 5 min. The mixture was heated under microwave irradiation at 400 W for several minutes. The obtained magnetic MOFs was recovered by a magnet and washed with water for several times.

Pure MIL101-NH₂ was synthesized using a similar method with the absence of Fe₃O₄. MIL101-NH₂ was collected by centrifugation at 16 000 rpm for 15 min. The solid was washed with absolute DMF for whole night to remove the free acid. Later, the solid was vacuum-dried overnight at 60 °C and stored in a desiccator.

Magnetic MOFs (40 mg) and 25.0 ml DOX solutions with different concentrations were added into 50 ml glass conical flask and then shaken at 90 rpm for 24 h under 30 °C ± 0.5 °C. The DOX loaded composite were separated and collected by the aid of an adscititious magnet and freeze-dried overnight.

The amount of DOX loaded on the magnetic MOFs was estimated by measuring the concentration of DOX before and after adsorption, which was quantitated by measuring the absorbance of DOX at 480 nm with a UV spectrophotometer. The drug loading capacity-D (mg g⁻¹) was calculated according to the following equation:

$$\begin{eqnarray*}&&D=\displaystyle {\textstyle \tfrac{{M}_{1}}{{M}_{0}}},\end{eqnarray*}$$

where M₀ is the quantity of composite including M₁ and M_MOFs (g), M₁ is the quantity of DOX (mg) contained by the composite.

DOX released from the magnetic MOFs composite was analyzed in different pH buffer solutions. The analysis experiment was taken in pH 7.4 PBS buffer if there were no special instructions. Typically, 10 mg composite were added into 25 ml buffer solution at controlled temperature and gently agitated. At predetermined intervals, the magnetic MOFs were separated using an adscititious magnet and 4 ml of the supernatants was withdrawn from the system. 4 ml fresh buffer solution was added into the system to maintain the total volume. The absorbance of the supernatants was monitored at 480 nm with a UV spectrophotometer. The concentration of DOX was calculated from the calibration curve obtained with known amounts of DOX concentrations. Each assay was performed in triplicate and reported as the average value.

The Hela cell lines were provided by the China Center for Type Culture Collection. The Hela cells were cultured in DMEM supplemented with 10% Fetal Bovine Serum at 37 °C and 5% CO₂.

In vitro cytotoxicity was measured by performing protocol of cell counting kit-8 (CCK-8) assay on the Hela cells. Cells were seeded into a 96-well cell culture plate at 5 × 10³ per well, and were cultured at 37 °C and 5% CO₂ for 24 h; different concentrations of magnetic MOFs (0, 5, 10, 25, 50 μg ml⁻¹), DOX (0, 1, 2, 5, 10 μg ml⁻¹) and DOX loaded magnetic MOFs (with equivalent DOX concentrations 0, 1, 2, 5, 10 μg ml⁻¹) diluted in DMEM were then added to each well. The cells were subsequently incubated for 24 h at 37 °C under 5% CO₂. Untreated cells servsed as 100% cell viability. Thereafter, 10 μl CCK-8 was added to each well and the plate was incubated for an additional 4 h at 37 °C under 5% CO₂. The optical density OD value (Abs.) of each well, was measured by means of a Tecan Spark TM 10M monochromator-based multifunction microplate reader at 450 nm. The following formula was used to calculate the inhibition of cell growth:

$$\begin{eqnarray*}&&{\rm{Cell}}\,{\rm{viability}}( \% )=({\rm{mean}}\,{\rm{of}}\,{\rm{Abs}}{\rm{.}}\,{\rm{value}}\,{\rm{of}}\,{\rm{treatment}}\\ &&\,{\rm{group}}/{\rm{mean}}\,{\rm{of}}\,{\rm{Abs}}{\rm{.}}\,{\rm{value}}\,{\rm{of}}\,{\rm{control}})\times 100 \% .\end{eqnarray*}$$

Morphological features and surface characteristics of samples were investigated by taking photographs using scanning electron microscopy (Zeiss, Germany) and transmission electron microscopy (Hitachi H-7000FA, Japan). Element mapping of the MOFs was conducted on HAADF-STEM. X-ray powder diffraction (XRD) patterns of samples were recorded on a PAN-alytical X' Pert Pro x-ray diffractometer with a Cu Ka radiation at a scanning rate of 0.02° s⁻¹ from 4° to 80°. Samples were ground to powders in order to eliminate the influence of crystalline orientation. The patterns of the magnetic hysteresis loops of the samples were characterized at room temperature using a HH-15 model vibrating sample magnetometer. The absorbance of the samples was monitored using UV-2550 spectrophotometer (Shimadzu, Japan). FT-IR spectra of the samples were recorded on a Nicolet 5700 FT-IR spectrometer. The specific surface area of the samples were measured at 77 K through N₂ sorption analysis using the BET method on a Belsorp-mini Ⅱ (Japan).

In a typical process, Fe₃O₄-NH₂ nanoparticles were directly dispersed into the synthetic precursor of MIL101-NH₂ composed of FeCl₃ · 6H₂O , BDC–NH₂ and H₂O, and the reaction was accomplished through a simple microwave-heated procedure. This method avoids time-consuming layer-by-layer MOF growth [16] and further modification of core particles, which is much simpler than most of the previous methods [4] of synthesizing Fe₃O₄ nanoparticle-based core–shell composites.

As figure 1(A1)–(A4) shown, the magnetic MOFs composite presents a spherical morphology with the diameters ranging from 140–330 nm. The pure MIL101-NH₂ shows a type of amorphous form. The diameter of the magnetic MOFs was obviously increased due to the coating of MIL101-NH₂ layer. The average diameter of Fe₃O₄-NH₂ nanoparticles was only 158 nm, while the diameter was increased to 268 nm at the molar ratio of 1:1, demonstrating the covering of MIL101-NH₂ on the surface of the Fe₃O₄-NH₂ nanoparticles. Figure 1(B) showed the statistical analysis of diameter of the nanoparticles with different molar ratio of Fe₃O₄-NH₂ to MIL101-NH₂. The optimal condition to form core–shell structured magnetic MOFs was the molar ratio of Fe₃O₄-NH₂ to MIL101-NH₂ as 1:1, resulting little visible amorphous MIL101-NH₂ and thick MOFs covering layer.

Zoom In Zoom Out Reset image size

Furthermore, HAADF-STEM and elemental mapping analysis were used to verify the core–shell structure of the Fe₃O₄-NH₂@MIL101-NH₂ composite. As figure 1(C) showed, a clear contrast between the core and shell was obtained wherein the core appeared dark, whereas the shell appeared lighter. Moreover, the elemental mapping analysis revealed that Fe elements were distributed in the core and shell. Comparing magnetic MOFs in C4 with Fe₃O₄-NH₂ in C2, the distribution of Fe element was much sparse at the edge of the MIL101-NH₂ shell. All the results demonstrated the successful formation of a core–shell structure in the Fe₃O₄-NH₂@MIL101-NH₂ composite.

In addition, XRD was also employed to demonstrate the success synthesis of magnetic MOFs from 4° to 80° (figure 2(A)). Fe₃O₄ nanoparticles, synthesized through a solvothermal method, were spherical with a mean diameter of 158 nm and in the cubic phase (JCPDS: 19-0629) [4, 21]. The characteristic peaks of Fe₃O₄-NH₂ nanocrystal at 30.2°, 35.5°, 43.2°, 57.3° and 62.8° were also presented in the patterns of the Fe₃O₄-NH₂@MIL101-NH₂ composite, indicating that the core–shell Fe₃O₄-NH₂@MIL101-NH₂ composites were synthesized by an in situ self-assembly of MIL101-NH₂ on the surface of Fe₃O₄-NH₂ nanoparticles. The characteristic peak at 8.5° was not so prominent corresponded to the amorphous MIL101-NH₂ observed by the SEM and TEM images [13].

Zoom In Zoom Out Reset image size

Superparamagnetism is essential to MR imaging [14]. Thus, the magnetic properties of the Fe₃O₄-NH₂@MIL101-NH₂ composites were investigated in a magnetic field range from −10 000 to +10 000 Oe at room temperature (figure 2(B)). The saturation magnetizations (Ms) of the Fe₃O₄-NH₂@MIL101-NH₂(1:1) and Fe₃O₄-NH₂@MIL 101-NH₂(2:1) were 20.47 and 21.32 emu g⁻¹, which were smaller than that of Fe₃O₄-NH₂ nanoparticles (27.67 emu g⁻¹) due to the encapsulation by the MIL101-NH₂ layer. Due to the inclusion of amorphous MIL101-NH₂, the Ms of Fe₃O₄-NH₂@MIL101-NH₂(1:2) was only 3.04 emu g⁻¹. The high Ms value and the existing hysteresis loop without significant coercivity and remanence in the magnetization hysteresis curves demonstrated the strong superparamagnetic character of the as-synthesized Fe₃O₄-NH₂@MIL101-NH₂ composites [16].

In figure 2(C), the characteristic peaks of MIL101-NH₂ at 1593 cm⁻¹, 1498 cm⁻¹, and 1423 cm⁻¹ and Fe₃O₄-NH₂ at 585 cm⁻¹ in the FT-IR spectrum of Fe₃O₄-NH₂@MIL101-NH₂ indicated the formation of the Fe₃O₄-NH₂@MIL101-NH₂ composites [6, 13, 21]. The two peaks at 3525 and 3382 cm⁻¹ in the spectrum of MIL101-NH₂, belonged to the asymmetric and symmetric stretching absorptions of the primary amine groups in the BDC-NH₂ ligands and Fe₃O₄-NH₂. The peak at 760 cm⁻¹ indicated the symmetric stretching of C–N bond of the BDC-NH₂ [22], The FT-IR results further verified the successful growth of the MIL101-NH₂ shell on the Fe₃O₄-NH₂ core.

Taking Fe₃O₄-NH₂@MIL101-NH₂ (1:1) as an example, the porosity of the magnetic MOFs composite was evaluated by nitrogen adsorption-desorption measurements. The composite shows a typical IV type isotherm with H3 hysteresis loop (figure 2(D)) [6]. The BET surface area of the sample was 96.04 m² g⁻¹ with the average pore size 6.68 nm (figure S1(A) is available online at stacks.iop.org/NANO/28/495601/mmedia), which was obviously increased in contrast to 2.96 nm of MIL101-NH₂ (figure S1(B)) [5].

DOX was used as a model drug to test the drug loading capacity of the magnetic MOFs. DOX was loaded to the magnetic MOFs by adsorption and the product was analyzed by FT-IR, as figure 3 shown. The characteristic peaks of magnetic MOFs at 1593, 1423, and 1252 cm⁻¹ and Fe₃O₄-NH₂ at 585 cm⁻¹ were remained in the spectrum of DOX loaded MOFs. The characteristic peaks of DOX at 2933, 1071, 1007 and 991 cm⁻¹ also appeared in the spectrum of DOX loaded magnetic MOFs, indicating the incorporation of DOX molecules in DOX loaded magnetic MOFs composite [4, 23].

Zoom In Zoom Out Reset image size

The DOX loading capacity was largely affected by the composite ratio of Fe₃O₄-NH₂ to MIL101-NH₂. As shown in table 1, The DOX loading capacity of pure Fe₃O₄-NH₂ nanoparticles was 9.46 mg g⁻¹ and the loading capacity of Fe₃O₄-NH₂@MIL101-NH₂ was substantially improved to over 39 mg g⁻¹. However, the DOX loading capacity of pure MIL101-NH₂ was only 5.29 mg g⁻¹, which was probably because that the pore size of the MIL101-NH₂ was small and not suitable for the DOX adsorption [5]. As the results of the porosity listed in table 1, the mean pore diameter of pure MIL101-NH₂ synthesized by this method was only 2.96 nm, while the pore diameter of magnetic MOFs composite were more than 6 nm. According to previous report, the pore size had significant effect on the adsorption property of the MOFs [24, 25]. Although the average pore size of Fe₃O₄-NH₂ nanoparticles was large (14.38 nm), the BET surface area-S was only 34.56 m² g⁻¹. The high drug loading capacity of Fe₃O₄-NH₂@MIL101-NH₂ composites was probably attributed to the large surface area of the MIL101-NH₂ shell and suitable pore size for the entrapment of DOX in MIL101-NH₂ [4].

Next, we showed the microwave time also affected the DOX loading capacity on magnetic MOFs. Table 2 showed the tendency of the specific surface area-S and the total pore volume–Vm of magnetic MOFs (Fe₃O₄-NH₂@MIL101-NH₂ 1:1) changing with reaction time. In the beginning of 5 min, S and Vm increased with the reaction time. Then, they both decreased with the time extended. The DOX loading capacity of the magnetic MOFs correlated well with S and Vm. Thus, the magnetic MOFs microwaved for 5 min had the largest BET surface area 96.04 m² g⁻¹ and total pore volume 22.07 cm³ g⁻¹, resulting an optimum DOX loading capacity of 48.77 mg g⁻¹. As large BET surface area could provide more chemical binding site and pore volume, as a result contribute to the DOX chemical adsorption and physical entrapment [26]. Further increase of the drug load could be accomplished by increasing the concentration of DOX. As the concentration of DOX increased from 200 to 10 000 mg l⁻¹, the drug loading capacity increased from 48.77 to 360.16 mg g⁻¹ (figure S2), which is among the optimal results that reported previously [5, 27].

The release behavior of DOX from the magnetic Fe₃O₄-NH₂@MIL101-NH₂ composite differs from Fe₃O₄-NH₂ nanoparticles and pristine MIL101-NH₂. As shown in figure 4, Fast release of DOX was observed from Fe₃O₄-NH₂ nanoparticles and pristine MIL101-NH₂. On the contrary, the release of DOX from Fe₃O₄-NH₂@MIL101-NH₂ was relatively slow and only 37%–61% of DOX was released in 48 h, suggesting the coating of MOFs layer could effectively reduce the DOX burst release of the Fe₃O₄-NH₂ nanoparticles. Moreover, the release of DOX was affected by the ratio of Fe₃O₄-NH₂ to MIL101-NH_2. It was reported that large pore sizes could reduce the steric diffusion resistance [26, 28]. The average mesopore size of Fe₃O₄-NH₂@MIL101-NH₂ decreased from 8.94 to 6.02 nm with the increased ratio of MOFs, leading to slower DOX release.

Zoom In Zoom Out Reset image size

The stability of the nanocarrier in a wide pH range from basic to acidic is essential for drug loading/release. Thus, we investigated the controlled release of the as-synthesized magnetic Fe₃O₄-NH₂@MIL101-NH₂ composites at different pH values (4.0, 6.5, 7.4). Taking the magnetic MOFs composite-Fe₃O₄-NH₂@MIL101-NH₂ (1:1) as an example, the DOX release curves under various pHs were shown in figure 5. In contrast to the slow and partial release of DOX from the magnetic MOFs composite in simulated body fluid at pH 7.4, 53.9% and 89.7% of the loaded DOX were released in 8 h when the composite was incubated in simulated tumor cell microenvironment (pH 6.5) [29] and weak acidic environment (pH 4.0). This was caused by the decomposition of MIL101-NH₂ in acid buffer due to the protonation of BDC-NH₂ [5]. The results in figure 5 indicated that the release of DOX from Fe₃O₄-NH₂@MIL101-NH₂ could be adjusted by varying pHs, favoring the targeting of tumor cells and on site DOX release.

Zoom In Zoom Out Reset image size

As a potential drug carrier that would be applied in vivo system, the cytotoxicity of magnetic MOFs is of great importance for its bioapplication. As the model drug-DOX was an antitumor antibiotic, we use Hela cell lines to investigate the cytotoxicity of magnetic MOFs. The result in figure 6(A) showed the magnetic MOFs co-incubated with Hela cells for 24 h exhibited little cytotoxicity in the analyzed concentration range (0–50 μg ml⁻¹). DOX loaded magnetic MOFs with equivalent dose of DOX exhibited similar cytotoxic effect as the DOX upon Hela cells in a dose-dependent manner (figure 6(B)). Half of the Hela cells died (IC50) by the chemotherapeutic effect of DOX co-incubated for 24 h at 37 °C, 5% CO₂ when the DOX concentration was above 1 μg ml⁻¹. The cell viability of the DOX loaded magnetic MOFs were 72.52%, 64.62% and 45.99% for 1:2, 1:1 and 2:1 magnetic MOFs samples (molar ratio Fe₃O₄-NH₂ to MIL101-NH₂) at the equivalent DOX concentration 10 μg ml⁻¹, which has a good correspondence with the drug release profiles in figure 4. In contrast to free DOX, DOX loaded magnetic MOFs show weak cytotoxicity due to the slow release of DOX in cell culture medium. All these results were consistent with the previous reports [30]. Hence, the results showed that the magnetic MOFs with low cytotoxicity and high biocompatibility is suitable for drug delivery.

Zoom In Zoom Out Reset image size