AMF-responsive doxorubicin loaded β-cyclodextrin-decorated superparamagnetic nanoparticles

Abstract

An alternating magnetic field (AMF)-responsive controlled release system has been developed by the binding of mono-6-deoxy-6-(p-tolylsulfonyl)-β-cyclodextrin (βCD-Ts) onto amine-modified superparamagnetic iron oxide nanoparticles (MNP-NH₂), resulting in a MNP-βCD nanocarrier. The structural, chemical and colloidal properties of the nanocarrier MNP-βCD were fully evaluated by several techniques. Doxorubicin hydrochloride (DOX), as a model anticancer drug, was loaded onto MNP-βCD resulting in MNP-βCD-DOX and the release process was investigated by varying the temperature (37 and 45 °C), pH (7.4 and 5.0) and presence of an AMF (with and without). DOX can interact with the negatively charged surface of MNP-βCD and also with the cavity of βCD forming inclusion complexes. Under acidic conditions, both the negatively charged surface of MNP-βCD and the ionizable groups of DOX are protonated and the interactions between DOX and the nanocarrier MNP-βCD are weakened, thus, accelerating drug release. Temperature increase can reduce the supramolecular interactions between DOX and the nanocarrier MNP-βCD, favoring DOX release. Thermo-induced burst release at 45 °C has been investigated either by applying an alternating magnetic field (AMF, f = 307 kHz) or heating the solution in a thermostatic cuvette holder. In the absence of an AMF at 45 °C and at the pH value of the lysosome (5.0), 92% of DOX was released into the solution within 6 h. Importantly, in an AMF, at 45 (±1) °C and pH = 5.0, the same percentage of DOX release was observed after 50 min. Furthermore, in vitro assays revealed that the unloaded MNP-βCD nanoparticles (100 μg mL⁻¹) displayed negligible cytotoxicity against A549 human lung adenocarcinoma cells either in the absence or presence of an AMF (H × f = 4.9 × 10⁹ A m⁻¹ s⁻¹) applied for 30 min on the cells. In the presence of an AMF no macroscopic temperature variation was detected in the wells containing the cells incubated with DOX unloaded MNP-βCD nanoparticles. The same amount of DOX loaded nanoparticles (MNP-βCD-DOX) showed cytotoxicity only in the presence of an AMF (H × f = 4.9 × 10⁹ A m⁻¹ s⁻¹ applied for 30 min), inducing cell death. Also, no macroscopic temperature variation was detected; therefore, DOX release might be associated with local heating generated from the rotation movements of the nanoparticles (Brown relaxation) upon AMF application. Thus, the DOX loaded β-cyclodextrin-decorated superparamagnetic nanoparticles (MNP-βCD-DOX) are a potential controlled drug release agent with cancer-killing properties under an AMF.

---

Introduction

The development of new alternative therapies for cancer treatment that are selective, less invasive and with lower adverse effects to the patient than the existing ones is urgently needed. Drug delivery systems (DDSs) have been extensively investigated because they can enhance drug solubility, drug circulation half-life and drug specific localisation, and reduce the therapeutic doses.^1–3 Particularly, stimuli-responsive DDSs, e.g., those in which the drug release is controlled by a given stimulus, such as temperature,⁴ redox activation,⁵ light radiation,⁶ pH changes,⁷ enzymes⁸ and an alternating magnetic field (AMF), are promising smart materials for cancer treatment.⁹

Iron oxide nanoparticles (IONPs) are approved by the US Food and Drug Administration (FDA) for biomedical applications^10,11 and are excellent candidates for DDSs because they exhibit low toxicity and strong magnetic responses.² IONPs can be introduced and concentrated in a specific tissue of the body by applying a magnetic field. IONPs smaller than 50 nm can exhibit superparamagnetic behavior, which is an important property for biomedical applications. Because superparamagnetic iron oxide nanoparticles (SPIONs) do not show residual magnetization in the absence of an external magnetic field,¹¹ they do not act as permanent magnets inside the body and are easily dispersed. As such they do not tend to agglomerate and present no danger of thrombosis.^2,12,13

For drug delivery applications, the SPIONs must be biocompatible and stable under physiological pH to improve their residence time in the body.^12,13 This can be achieved by coating the SPION surface with a variety of molecules, such as polyethylene glycol (PEG),¹⁴ dextran,¹⁵ citric acid¹⁶ and chitosan.¹⁷ Particularly interesting, cyclodextrins (CDs), a class of cyclic oligosaccharides, have been coupled to several DDSs,^5,18,19 thus acting as carriers for hydrophobic drugs and enhancing their solubility in water.²⁰ It is well known from the literature that CDs can form inclusion complexes with slightly water soluble drugs, e.g. doxorubicin and tamoxifen, through host–guest interaction with their cavities.^7,21 Doxorubicin (DOX), widely used in cancer treatment, acts by intercalating into DNA, as well as inhibiting the enzyme topoisomerase II, which is responsible for cell division and growth,⁴ leading to DNA damage and cell death. However, the continuous use of DOX, among numerous problems, leads to cardiomyopathy.²² Therefore, the development of efficient DDSs for antineoplastic agents to improve chemotherapy efficiency is still a challenge.

In this work, we successfully prepared superparamagnetic iron oxide nanoparticles (SPIONs) coated with β-cyclodextrin, MNP-βCD, as a drug delivery system (Scheme 1). Nanocarrier MNP-βCD was prepared in four steps: (i) monotosylation of the βCDs in basic solution, (ii) synthesis of Fe₃O₄ nanoparticles via coprecipitation of Fe(II) and Fe(III) ions from basic solution (MNP), (iii) Fe₃O₄ surface modification with aminopropyl groups (MNP-NH₂) and (iv) coupling of the monotosylated βCDs with the amino-functionalized Fe₃O₄ nanoparticles through S_N2 nucleophilic substitution (MNP-βCD). The nanocarrier was fully characterized and loaded with DOX (MNP-βCD-DOX), as a model anticancer drug. DOX release profiles were monitored using UV-Vis spectroscopy under various conditions, namely, at different pH (7.4 and 5.0) and temperature values (37 and 45 °C) and in the presence and absence of an AMF. Moreover, cell viability assays using small amounts (100 μg mL⁻¹) of the nanoparticles were investigated towards human lung cancer cells (A549) in the presence and absence of an AMF. It is important to highlight that the literature does not report the cytotoxicity against cells of drug delivery systems based on magnetic nanoparticles decorated with βCD in the presence of an AMF.^{7,20,21,23,24} Furthermore, it should be stressed that good cytotoxicity results employing SPIONs have generally been obtained by using either high amounts of magnetic nanoparticles or harsh AMF conditions (high magnetic field and frequency), for long periods of time. However, this is not recommended because high magnetic fields can generate eddy currents and damage healthy tissues.^25,26

Scheme 1 Representation of the synthetic procedure used to prepare the doxorubicin loaded magnetic nanocarrier (MNP-βCD-DOX).

Experimental

Materials

The following chemicals were used as received: 3-aminopropyltriethoxisilane (99%) (APTES), iron(II) chloride tetrahydrate, iron(III) chloride anhydrous, p-toluenesulfonyl chloride (TsCl), doxorubicin hydrochloride, β-cyclodextrin (>97%) (βCD), potassium chloride, sodium chloride, sodium phosphate dibasic and potassium phosphate monobasic from Sigma-Aldrich and ammonium hydroxide (25–30%), acetone, ethanol, acetonitrile, dimethylformamide (DMF) and sodium hydroxide from Vetec.

Structural analysis

Elemental analyses (CHN) were carried out using a Perkin-Elmer CHN 240 C analyser at the Analytical Center of the Institute of Chemistry, University of São Paulo, Brazil. Fourier-transform infrared (FTIR) spectra were recorded in the range of 400–4000 cm⁻¹ on a Varian 660-IR FT-IR spectrophotometer using KBr pellets. The TGA of the samples was carried out under flowing nitrogen (50 cm³ min⁻¹) on a TG Instruments system, model Shimadzu DTG-60/60H at a heating rate of 5 °C min⁻¹; the samples were heated from 30 °C to 600 °C at a heating rate of 5 °C min⁻¹. Powder XRD patterns were collected using a Rigaku Miniflex X-ray diffractometer equipped with a scintillation counter detector and CuK_α1 = 1.5405 Å and CuK_α2 = 1.5444 Å radiation source within a 2θ range of 10–80°. The following databases were used: PDF 082-1533 for magnetite and PDF 039-1346 for maghemite, both from ICCD, 2003. Raman spectra were recorded using a Witec Alpha 300 confocal Raman imaging microscope. The experiments were performed at 25 °C using an Nd:YAG green laser with a wavelength of 532 nm. The laser light was focused onto the sample using a 100× objective lens (NA = 0.95); the integration time was 0.5 s and the number of scans was 600. TEM images were obtained using a JEOL JEM-1011 operating at an accelerating voltage of 80 kV. The magnetic measurements were carried out in a magnetometer using a SQUID sensor (MPMS3, Quantum Design). The NMR spectra were recorded on a Varian VNMRS 500 MHz for ¹H and 75 MHz for ¹³C. The size of the hydrodynamic diameter and the charge surface were measured using dynamic light scattering (DLS) and zeta potential, respectively, both using Zetasizer Nano ZS, Malvern instruments. UV-Vis absorption spectra were obtained using a Cary 5000 UV-Vis-NIR spectrophotometer (Varian) equipped with a thermoelectric (Peltier system) temperature-controlled cuvette holder (TLC 40) coupled with a temperature controller (Quantum Northwest TC 125).

Methods

Preparation of the SPIONs. SPIONs were synthesized by the chemical coprecipitation method.²⁷ FeCl₂·4H₂O (11.7 g, 92.5 mmol) and anhydrous FeCl₃ (30.0 g, 184.9 mmol) were dissolved in 500 mL of water under Ar atmosphere and mechanical stirring (400 rpm) at RT. Then, 100 mL of 30–25% NH₄OH was quickly added in one portion to the mixture. A black precipitate appeared immediately and the reaction mixture remained under mechanical stirring and Ar atmosphere at RT for 1 h. The black precipitate was separated from the solution using a magnet and subsequently washed with deionized water until neutral pH. The resulting black powder was finally dried under vacuum at RT for 8 h, yielding magnetic iron oxide nanoparticles, named MNPs (27.0 g).

Preparation of APTES-coated-SPIONs. SPIONs were coated with APTES following a procedure from the literature with minor modifications.^28,29 MNPs (13.0 g) were dispersed in EtOH : H₂O (180 : 1) by sonication under mechanical stirring (500 rpm) at RT for 30 min. The dispersion was transferred to a round-bottom flask and heated to 70 °C, and then the first aliquot of APTES (10 mL) was added. The reaction mixture remained under stirring for 4 h, after which period three aliquots of 5 mL of APTES were added during an interval of 5 h. The reaction mixture was left stirring for another 5 h, and the solid was collected using a magnet and washed with EtOH (3×) and acetone (3×). The final product was dried under vacuum for 8 h and named MNP-NH₂ (8.0 g).

Synthesis of mono-6-deoxy-6-(p-tolylsulfonyl)-cyclodextrin

This compound was synthesized according to a procedure described in the literature and named βCD-Ts (2.50 g, 9%).³⁰¹H NMR (500 MHz, DMSO-d₆): δ 7.75 (d, J = 8.3 Hz, 2H); 7.43 (d, J = 7.8 Hz, 2H); 5.56–5.68 (m, 14H); 4.76–4.85 (m, 7H); 4.21–4.37 (m, 9H); 3.52–3.72 (m, 27H); 3.20–3.45 (m, overlaps with water) 2.43 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ 144.7 (s), 132.8 (s), 129.8 (d), 127.4 (d), 101.9 (m), 81.5 (d), 73.1–72.4 (m), 69.6, 68.9, 59.6 (t), 21.1 (s).

Synthesis of SPIONs coupled with βCD

βCDs were grafted onto the MNP-NH₂ surface following a procedure described in the literature with minor modifications.³¹ Briefly, MNP-NH₂ (5.0 g) was suspended in 50 mL of dry DMF under mechanical stirring (400 rpm) in a two-neck flask. The suspension was heated to 50 °C under Ar atmosphere and then βCD-Ts (2.3 g, 2.0 mmol) dissolved in 30 mL of dry DMF was added to the suspension. The reaction mixture remained at 50 °C for 8 h and after this period left stirring overnight at RT. The black solid obtained was separated using a magnet and successively washed with DMF and acetone (3×). The resulting material was dried under vacuum at RT and named MNP-βCD (5.0 g).

Purification of the MNP-βCD by dialysis

MNP-βCD (4.6 g) was dispersed in 100 mL of deionized water using an ultrasound bath for 10 min. The suspension of MNP-βCD was added into a cellulose membrane (14 kDa, Sigma-Aldrich) forming a bag. This bag was placed in a beaker containing 3 L of deionized water and remained under magnetic stirring for 4 days. The water was changed 3 times. After this period, the suspension was rota-evaporated and dried under vacuum for 8 h (4.5 g).

DOX loading of MNP-βCD

MNP-βCD (1.0 g) was soaked in 20 mL of 50 mg L⁻¹ of DOX phosphate-buffered saline solution (PBS). The mixture was protected from light and gently shaken (160 rpm) for 24 h at RT. MNP (1.0 g) and MNP-NH₂ (1.0 g) were also loaded with DOX under the same conditions for comparison. The materials were separated from the solution using a magnet, dried under vacuum at RT for 6 h and named MNP-βCD-DOX, MNP-DOX and MNP-NH₂-DOX. The supernatant was collected to calculate the drug loading efficiency. DOX concentration was determined by UV-Vis standard curve, according to the measurement of a series of concentrations of DOX solutions (1, 10, 20, 30, 40, 50 mg mL⁻¹), at λ = 254 nm.

DOX release behavior in the absence of AMF

A series of experiments were performed to investigate the pH and temperature dependence of DOX release without an alternating magnetic field (AMF). The studies were conducted at 37 °C (body temperature) and 45 °C (hyperthermia temperature). For each temperature, the samples were kept at pH 7.4 (physiological pH) and pH 5.0 (lysosome). Initially, 10 mg of MNP-βCD-DOX was transferred to a quartz cuvette containing 1 mL of PBS buffer pH 7.4 or 5.0. A permanent magnet was placed below the cuvette to decant the nanoparticles. The cuvette was then inserted into the preheated thermoelectric controlled cuvette holder of the UV-Vis spectrophotometer, at 37 °C and 45 °C. A spectrum was acquired every minute during 6 h at each temperature for both pH values.

Magnetic hyperthermia capability of the nanocarrier MNP-βCD

The magnetic hyperthermia ability of MNP-βCD was investigated in a magnetic induction heating system (Nanoscale Biomagnetics DM2-s53) equipped with a fiber optic temperature probe and vacuum shield as thermal insulation. The nanoparticles were transferred to a glass vial containing PBS buffer pH = 7.4 or 5.0, resulting in concentrations of 1, 5, 10, 15 or 20 mg mL⁻¹. The systems were heated at 25 °C for 8 min, in an AMF at a fixed frequency (f) of 307 kHz and strength (H) of 200 Oe.

DOX release behavior in an AMF

The effect of an AMF on DOX release was investigated in the same magnetic induction heating used for hyperthermia analysis at fixed frequency (f) = 300 kHz. Two mixtures of MNP-βCD-DOX (10 mg) and PBS buffer (1 mL) at pH 7.4 and pH 5.0, respectively, were prepared. A glass vial containing each mixture was placed in the center of the coil and the measurements started when the temperature of the system was 23 °C (RT). The samples were exposed to the AMF for 10, 20, 30, 40 and 50 minutes. For the first 6 min, the sample temperature increased up to 45 (±1) °C (hyperthermia temperature) and then, was maintained constant by controlling the field amplitude (H). Immediately after AMF exposure, the supernatant was collected by placing a magnet on the bottom of the vial and analyzed by UV-Vis spectroscopy.

Cell culture

The A549 human adenocarcinomic lung cells were obtained from the American Type Culture Collection (ATCC® CCL-185™). They were maintained in high glucose Dulbecco's modified Eagle's medium, supplemented with 10% fetal bovine serum (FBS), in a humidified atmosphere of 95% air–5% CO₂, at 37 °C. Sub-confluent cells were seeded using 0.01% trypsin–1 mM EDTA for use in all experiments.

In vitro cell viability assay with AMF treatment

The viability of the cells seeded with the samples was assessed using a commercial kit (Presto Blue® Cell Viability Reagent, purchased from Invitrogen) following the instructions provided by the supplier. The cells were seeded into an adapted 96-well plate at a density of 10⁴ cells per well, and were cultured for 24 h, as described in the cell culture conditions. Aliquots of PBS buffer solution containing free DOX (0.1 μg), and DOX loaded and unloaded MNP-βCD (100 μg) were added to each well containing the cells (100 μg mL⁻¹), in sextuplicate, and incubated for 24 h. The adapted plate was placed in the middle of the coil of the magnetic induction heating system and exposed to the AMF (f = 307 kHz and H = 200 Oe) for 30 min. As a control, the same assay was carried out in the absence of an AMF. Immediately, fluorescence was detected at 485 nm (excitation) and 590 nm (emission) using a fluorometer (Synergy 2, BioTek Instruments). The sodium dodecyl sulfate solution (1% w/v in PBS buffer) was used as a positive control (cytotoxic), and (DMEM) culture medium was used as a negative control (noncytotoxic).

Results and discussion

Characterization of the nanocarrier

The mono-6-deoxy-6-(p-tolylsulfonyl)-β-cyclodextrin (βCD-Ts) was synthesized according to a procedure previously reported in the literature with modifications.^30,321H and ¹³C NMR spectra of βCD-Ts in DMSO-d₆ confirmed the monotosylation of βCD (Fig. S1 and S2, see ESI†).

Fig. S3 (see ESI†) shows the FTIR spectra of the nanocarrier MNP-βCD and of its precursors. The FTIR spectrum of Fe₃O₄ shows a broad band at 3300–3450 cm⁻¹ associated with the stretching vibration mode of the surface hydroxyl groups.²⁷ An intense band due to Fe–O bond vibrations at 590 cm⁻¹ from magnetite bulk appears in all materials. The FTIR spectrum of the MNP-NH₂ shows bands at 1006 cm⁻¹, 1103 cm⁻¹, 1627 cm⁻¹ and 2849 to 2920 cm⁻¹, which are assigned to the Fe–O–Si,³³ Si–O,^28,34 N–H³⁴ (in the plane) and C–H²³ stretching vibrations, respectively. These bands indicate that the MNP surface was coated with (3-aminopropyl)triethoxysilane (APTES).

In the spectrum of βCD-Ts, the βCD fingerprint region is located between 900 and 1200 cm⁻¹.²³ The characteristic peaks of βCD appear at 941 cm⁻¹ (R-1,4 bond skeleton vibrations), 1031 and 1082 (coupled C–C/C–O stretching vibration) and 1157 cm⁻¹ (asymmetric glycosidic vibration C–O–C).^23,31 In the same region overlapping bands related to the stretching vibrations of the bonds Fe–O–Si and Si–O are also observed. The bands at 2849–2920 cm⁻¹ in the spectra of MNP-NH₂ and MNP-βCD are due to C–H stretching modes.

Thermogravimetric analysis (TGA) was carried out to verify the thermal stability of the nanocarrier MNP-βCD and its precursor MNP-NH₂ (Fig. S4, see ESI†). The TGA curves show two main decomposition steps. The initial weight loss of approximately 1% for MNP-NH₂ and 2% for MNP-βCD is likely due to the presence of physisorbed water or solvent molecules on the surface of the materials. The highest weight loss for the nanocarrier MNP-βCD is attributed to the presence of hydroxyl groups in βCD that can form hydrogen bonds with water molecules. Two weight losses observed at temperatures ranging from 200 to 600 °C are related to the decomposition of the organosilane group (APTES) and the βCD grafted onto the surface of SPIONs.²³ The weight loss in this step is approximately 6 and 8 wt% for MNP-NH₂ and MNP-βCD, respectively. These results corroborate with the carbon elemental analyses for the nanocarrier and its precursor (1.6 mmol g⁻¹ MNP-NH₂ and 2.1 mmol g⁻¹ MNP-βCD). Thus, the higher weight loss in MNP-βCD than in MNP-NH₂ is due to the presence of βCD groups.

Powder X-ray diffraction (PXRD) patterns show the cubic spinel-structured oxide of magnetite (Fe₃O₄) or maghemite (γ-Fe₂O₃) phases (Fig. 1). Confocal Raman microscopy was used, therefore, to distinguish both phases of iron oxide nanoparticles (Fig. S5, see ESI†). The spectra exhibit broad bands at approximately 717 (A_1g), 500 (T_2g) and 354 (E_g) cm⁻¹ assigned to the stretch modes of oxygen atoms along Fe–O bonds associated to the maghemite phase.^27,35 The band related to magnetite at 666 cm⁻¹ is also present in all spectra. Therefore, the nanocarrier and its precursors are composed of both magnetite and maghemite phases of iron oxide.

Fig. 1 PXRD patterns of the nanocarrier MNP-βCD and its precursors.

Transmission electron microscopy (TEM) and the particle-size distribution (PSD) curve for MNP-βCD are shown in Fig. 2. The particles obtained by the coprecipitation method are spherical with an average particle diameter of 14 nm and narrow size distribution.

Fig. 2 TEM image and PSD (inset) of the nanocarrier MNP-βCD.

The magnetization property of MNP-βCD was investigated at room temperature using a SQUID vibration sample magnetometer (VSM). The absence of hysteresis in the curve shown in Fig. 3 confirms the superparamagnetic nature of the nanocarrier MNP-βCD at room temperature. There is neither remnant magnetization nor coercivity. The saturation value of the magnetization corresponds to approximately 65 emu g⁻¹ for MNP-βCD.

Fig. 3 Field dependence of the magnetization measured at 300 K for MNP-βCD (the inset above shows photographs of the MNP-βCD suspensions in PBS buffer in the presence and absence of a permanent magnet and the inset below shows a magnification around zero field at the same temperature).

Colloidal properties of the nanocarrier MNP-βCD and its precursors were investigated to evaluate the surface functionalization of SPIONs as well as the electrostatic stability of the nanocarrier in PBS buffer. Zeta potential and hydrodynamic diameter measurements as a function of pH are displayed in Fig. 4 and Fig. S6 (see ESI†), respectively. The zeta potentials (ζ) of MNP, MNP-NH₂, and MNP-βCD (0.1 mg mL⁻¹) were measured in PBS buffer solution at pH ranging from 3 to 11, using 0.03 mol L⁻¹ KNO₃ as the background electrolyte. The values of zeta potentials of the nanocarrier and its precursors increase with decreasing pH because of the protonation of the hydroxyl and amino groups present on the surface of the MNP, MNP-NH₂, and MNP-βCD.

Fig. 4 Zeta potential curves as a function of pH (3–11) for MNP, MNP-NH₂ and MNP-βCD.

The isoelectric points (pI) of the SPIONs are 6.9 for MNP, 9.5 for MNP-NH₂ and 5.5 for MNP-βCD. The precursor MNP-NH₂ shows positive surface charges at pH < 9.5, indicating that the aminopropyl groups are protonated (NH₃⁺).

Zeta potential values at pH = 7.4 for MNP, MNP-NH₂ and MNP-βCD are −28, +17 and −26 mV, respectively. For biological applications highly negative or positive values of zeta potential are desirable in order to avoid nanoparticle aggregation. The functionalization of the SPIONs with βCD increases the hydrodynamic diameters, measured at pH = 7.4, from 134 for MNP to 185 nm for MNP-βCD (Fig. S6, see ESI†).

The nanocarrier MNP-βCD was loaded with DOX in PBS buffer at pH 7.4. The nanocarrier precursors MNP and MNP-NH₂ were also loaded with DOX at pH 7.4 for comparison purposes. DOX is an anthracycline antibiotic presently used in the treatment of a wide variety of cancers, such as acute lymphoblastic leukaemia, lymphomas, multiple myeloma, sarcomas, mesotheliomas, germ cell tumours of the ovary or testis, carcinomas of the head and neck, breast, pancreas, stomach, liver, ovaries, lung, prostate, and uterus, and neuroblastomas.³⁶ DOX can assume several prototropic forms at different pH values due to the presence of hydroxyl, carbonyl and amino groups on the sugar sidechain.³⁷ For example, in PBS buffer solution (pH = 7.4) the primary amino groups of DOX molecules are mostly protonated.³⁸ Furthermore DOX, as the hydrochloride salt, can interact with βCD in a 1 : 1 stoichiometry by hydrogen bonding and hydrophobic interactions, resulting in an inclusion complex (DOX⊂βCD). The formation constant (K_f) in phosphate buffered saline solution, at 25 °C and pH 7.1, was reported to be 210 M⁻¹.³⁷ The protonated form of DOX can also interact with the negative surface of the nanocarrier MNP-βCD (ζ = −28 mV at pH = 7.4) by electrostatic interaction and hydrogen bonding.³⁹

DOX aqueous solution absorbs in the UV-Vis region of the spectra and emits at 560–590 nm, therefore, DOX release can be monitored using either UV-Vis or fluorescence spectroscopies. In this work we have used UV-Vis spectroscopy. The absorption spectra of DOX solutions in PBS buffer (pH = 7.4) and the standard curve are shown in Fig. S7 and S8 (see ESI†).

DOX loading efficiency was calculated using eqn (1); where M is the initial mass of DOX in the solution (1 mg) and m is the loaded DOX mass per gram of the SPIONs. The calculated loaded amount of DOX in the nanocarrier was 0.81 μg mg⁻¹ of MNP-βCD nanoparticles. The quantities were determined using the DOX standard curve from the difference between the initial DOX concentration (50 mg L⁻¹) in PBS buffer solution and the remaining DOX concentration in the supernatant after a period of 24 h in contact with the SPIONs (Fig. S9, see ESI†).

DOX loading efficiency (%) = m/M × 100 (1)

The DOX loading efficiency in MNP (ζ = −26 mV at pH = 7.4), MNP-NH₂ (ζ = +17 mV at pH = 7.4) and MNP-βCD (ζ = −28 mV at pH = 7.4) was 60%, 34% and 81%, respectively (Fig. S9, see ESI†). The DOX loading efficiency in the nanocarrier MNP-βCD was higher than that of MNP and MNP-NH₂ both because of the strong interaction between the positively charged DOX molecules and the negatively charged surface of this nanocarrier, and the interaction between DOX and the cavity of βCD, resulting in a DOX⊂βCD inclusion complex.^38,40 After DOX loading, the zeta potential of the nanocarrier decreased from −28 mV to −9 mV (pH = 7.4), confirming the electrostatic interaction between the species. MNP (ζ = −26 mV) may also interact with DOX by electrostatic interactions. The interaction between DOX and the positively charged MNP-NH₂ nanoparticles suggests that DOX is not fully protonated at pH = 7.4. The DOX loading efficiency of MNP-βCD was higher compared to other magnetic DDSs described in the literature.⁴¹

DOX release profile in the absence of an AMF

DOX release from an aqueous dispersion of loaded MNP-βCD at the concentration of 10 mg mL⁻¹ was monitored by UV-Vis spectroscopy, at 37 and 45 °C, in a thermostatic cuvette holder, in the absence of an AMF. DOX release at both temperatures was investigated at physiological pH (7.4) and at lysosomal pH (5.0) (Fig. S10, see ESI†) and the absorption intensities of DOX were measured as a function of time to obtain the release profiles (Fig. S11, see ESI†). The percentage of DOX release was calculated using the standard curve (Fig. S8, see ESI†).

DOX release profiles are sensitive to the temperature and pH values of the solution (Fig. S10, see ESI†). From this figure, it is straightforward to conclude that the DOX release is faster at 45 °C than at 37 °C at both pH values and also that it is faster at pH = 5.0 at both temperatures. Considering the pH = 5.0, at both temperatures, 60% of the whole DOX was released within 120 min, and reaches the maximum amount of DOX released within 360 min. The maximum released at 45 °C is greater than at 37 °C, reaching an amount of 92% and 77%, respectively. The maximum amount of DOX released at pH = 7.4 is 83% at 45 °C and 49% at 37 °C within 360 min.

Under acid conditions, the negatively charged surface of the nanocarrier MNP-βCD is protonated (the zeta potential changed from −26 mV at pH −7.4 to +6 mV at pH 5.0) and the interaction with DOX is weakened, thus, accelerating drug release.^42,43 The temperature increase can reduce the supramolecular interactions between DOX and the nanocarrier MNP-βCD, favoring DOX release.

Magnetic heating efficiency of the MNP-βCD nanoparticles in an AMF

To confirm the potential of the nanocarrier MNP-βCD as a magnetic hyperthermia agent, the magnetic heating efficiency of this material in an alternating magnetic field was investigated (Fig. 5). The magnetic heating capability of different concentrations of MNP-βCD dispersed in PBS buffer solution was evaluated in a magnetic field strength (H) of 200 Oe (1.6 × 10⁴ A m⁻¹) and frequency (f) = 307 kHz (H × f = 4.9 × 10⁹ A m⁻¹ s⁻¹). It is worth noting that for a physiologically safe hyperthermia treatment and to preserve the healthy tissues from eddy currents, the product H × f must be <5 × 10⁹ A m⁻¹ s⁻¹.^25,44

Fig. 5 The magnetic heating efficiency of different concentrations of the nanocarrier MNP-βCD dispersed in PBS buffer solution (H × f = 4.9 × 10⁹ A m⁻¹ s⁻¹).

The temperature of the MNP-βCD suspensions in PBS buffer in an AMF depends on the concentration of the nanoparticles (Fig. 5). The higher the concentration, the faster the heating of the system, for the same H and f, e.g., at 10 mg mL⁻¹, the MNP-βCD dispersion takes 251 s to reach the hyperthermia temperature of 45 °C. At lower MNP-βCD concentrations (1 and 5 mg mL⁻¹), the dispersions do not reach the temperature of 45 °C in the conditions employed. Notably, however, Pellegrino and coworkers, have detected significant local heating increase, without increase of the solution temperature, by employing a molecular temperature probe attached to SPIONs of similar average diameter and saturation magnetization to ours, at the concentration of 5 nM (much lower than the one employed herein), in the presence of slightly milder AMF conditions than described in this work.⁴⁵ Therefore, based on Pellegrino's work, we assume that even at low MNP-βCD concentrations, local heating is generated.

The magnetic heating efficiency results shown in Fig. 5 are related to the specific absorption rate (SAR) of the nanocarrier MNP-βCD. The SAR is the amount of magnetic energy converted into heat per unit of time and mass of the magnetic nanoparticles under a particular magnetic field strength and frequency.⁴⁶ The SAR depends on the size, shape, composition, magnetic interaction and concentration of the magnetic nanoparticles in the dispersion.⁴⁷ To calculate the SAR of the samples we used the initial slope method (eqn (2)). The variables m_NP and m₁ are the mass (kg) of the nanoparticles and of the liquid, respectively; C_NP and C₁ are the specific heat (J kg⁻¹ K⁻¹) of the nanoparticles and the liquid; and ΔT/Δt is the initial slope of the temperature vs. time curve, which is the temperature increase per time (K s⁻¹).⁴⁸

SAR = (m_NPC_NP + m₁C₁/m_NP) × (ΔT/Δt) (2)

The calculated SAR values are 51 and 33 W g⁻¹ using the concentrations of 20 and 10 mg mL⁻¹ of the MNP-βCD, respectively. Therefore, the higher the concentration of the nanoparticles, the larger their heating efficiency.

DOX release behavior in an AMF

DOX release from the loaded nanocarrier MNP-βCD at the concentration of 10 mg mL⁻¹ (SAR = 51 W g⁻¹) in PBS buffer (pH 7.4 and 5.0) in an AMF was monitored by UV-Vis spectroscopy. At this concentration of MNP-βCD, the magnetic field strength (H) could be controlled during the experiment to keep the temperature constant at 45 (±1) °C (Fig. S12, see ESI†). The absorption intensities of DOX were collected as a function of time and used to generate the release curves (Fig. 6). DOX release curves obtained from experiments carried out in a thermostatic cuvette holder, at 45 °C (pH 7.4 and 5.0), in the absence of an AMF, (Fig. S10, see ESI†) were also plotted in Fig. 6 for comparison.

Fig. 6 DOX release curves obtained at 45 (±1) °C from MNP-βCD-DOX in the presence and absence of an AMF as a function of time.

It is important to stress that DOX release at 45 °C is clearly much more efficient when the sample is in an AMF (at both pH values), than in its absence, in which case only 45% and 29% of DOX was released within 50 min at pH = 5 and pH = 7.4, respectively. This percentage increased to 92% (pH = 5) and 85% (pH = 7.4) upon application of an AMF for 50 min (Fig. 6 and Scheme 2). Moreover, the pH effect is less pronounced in an AMF. In the experiments presented in Fig. 6, DOX release at pH = 7.4 and 5.0 was triggered by heat from different heating sources. In the absence of an AMF the loaded nanocarrier dispersions were heated in a thermostatic cuvette. Upon application of an AMF the temperature increase of the dispersions is mainly associated with Néel and Brown relaxation losses (Scheme 2). In AMF-response, the heat generation occurs through interaction of the magnetic moments of the SPIONs of the MNP-βCD-DOX with the AMF, thus converting magnetic energy into thermal energy.⁴⁸ On the one hand, the Néel relaxation is related to the SPIONs magnetic moments that switch, driven by the action of the AMF (Scheme 2b). On the other hand, the Brown relaxation is due to the torque over the SPIONs, caused by the field gradient. This torque will spin the MNPs giving rise to friction between the SPION's surface and the medium, generating local heat (Scheme 2a).

Scheme 2 Representation of the DOX release process from MNP-βCD-DOX upon application of an AMF: (a) Brown and (b) Néel relaxations. In Brown relaxation the magnetic moment does not relax, but the particle rotates following the AMF; whereas in Néel relaxation the magnetic moment relaxes following the AMF direction and the particle does not move.

Therefore, at the same temperature (45 °C) DOX release is faster in an AMF than in the absence of an AMF. This is, somehow, related to both Néel and Brown relaxation processes. However, taking into account that DOX is weakly bonded to βCD and to the negatively charged surface of the nanoparticles, we could suggest that the mechanical friction generated by the movement of MNP-βCD-DOX in the medium could wrest the DOX from both the cavity of βCD and from the negatively charged surface of the nanoparticles. Thus, Brown relaxation should play a preponderant role in the DOX release process.

Viability assay in an AMF

In vitro experiments were carried out to evaluate the cytotoxicity of DOX unloaded and loaded MNP-βCD towards human lung carcinoma A549 cells in an AMF and without it (Fig. 7). The cells were incubated with 100 μg mL⁻¹ of DOX unloaded and loaded MNP-βCD, for 24 h. DOX unloaded MNP-βCD nanoparticles show negligible cytotoxicity both without and with AMF exposure for 30 min (f = 307 kHz and H = 200 Oe). As expected (see Fig. 5), upon AMF application, no macroscopic temperature variation was detected due to the low nanoparticle concentration. These results indicate that the local heating that might have been generated, as previously discussed,⁴⁵ does not cause cell damage. The A549 cells were treated with free DOX at the concentration of 0.1 μg mL⁻¹, which is slightly higher than the concentration of DOX in the MNP-βCD nanocarrier (0.08 μg mL⁻¹). Cell viability was 65% of the control cell level, thus showing the cytotoxicity of DOX for this type of cell line (Fig. 7). The cell viability with DOX loaded MNP-βCD (MNP-βCD-DOX) was analyzed with and without AMF exposure (Fig. 7). MNP-βCD-DOX exhibits low cytotoxicity towards A549 cells without AMF. This result can be explained considering that only approximately 40% of DOX is released within 30 min, at pH = 5.0 and 37 °C in the absence of AMF (Fig. S10, see ESI†). This percentage of DOX corresponds to a concentration of 0.0058 μmol L⁻¹, which is ten times smaller than the IC₅₀ (0.05 μM),⁴⁹ therefore, this result was expected. Nonetheless, when MNP-βCD-DOX was exposed to an AMF for 30 min, the cell viability was 67% of the control cell level, revealing that the cytotoxicity of MNP-βCD-DOX is improved in an AMF. Since the unloaded MNP-βCD exposed to the AMF was not cytotoxic, the effect observed is due to the released DOX. Importantly, no temperature variation was detected upon application of an AMF in the wells containing the cells, because of the low concentration of the SPIONs (100 μg mL⁻¹). Therefore at low concentration of MNP-βCD-DOX, in the presence of an AMF, DOX release might result from local heating generated by the rotation movements of the SPIONs, which do not yield macroscopic heating of the medium.

Fig. 7 Prestoblue viability assay of A549 cells not exposed and exposed to the AMF after 24 h incubation with DOX (0.1 μg mL⁻¹) and DOX loaded and unloaded MNP-βCD systems (100 μg mL⁻¹). Negative control (cells): cells exposed to the culture medium (DMEM); positive control (C+): cells exposed to sodium dodecyl sulfate (SDS) 1%. ANOVA followed by Dunnet post hoc test was performed *P < 0.01.

Conclusions

In this study, the preparation and operation of a nanocarrier based on superparamagnetic nanoparticles decorated with β-cyclodextrin and loaded with doxorubicin has been described. The magnetic nanoparticles were obtained by the coprecipitation method yielding crystalline particles composed of both magnetite and maghemite phases with a magnetic core size of around 14 nm. The particles also displayed superparamagnetic behavior at room temperature. The surface of the SPIONs was functionalized with βCD, which improved both the colloidal stability of the particles and the DOX loading capacity by supramolecular host–guest and electrostatic interactions between the drug and the nanocarrier. Investigation of DOX release by UV-Vis spectroscopy showed that heat (45 °C) accelerates drug release. At physiological pH and 37 °C, drug release is low. Furthermore, a burst release was achieved by magnetic hyperthermia (45 °C) under AMF (f = 307 kHz and H = 200 Oe) exposure. Finally, the use of the nanocarrier as a possible drug delivery system against A549 cells, in an AMF has been demonstrated. In vitro cytotoxicity assays showed that the unloaded material at the concentration of 100 μg mL⁻¹ did not reduce cell viability under an AMF. Therefore the cytotoxicity caused by the loaded nanocarrier in an AMF was exclusively due to DOX release from MNP-βCD-DOX triggered by local heating resulting from the particle movements. The results reported herein have demonstrated that the proposed nanocarrier is a potential drug delivery system for cancer therapy.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Council for Scientific and Technological Development (CNPq grant number 550572/2012-0 and research fellowships), the Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES, T.C.S. fellowship) and the Rio de Janeiro Research Foundation (FAPERJ). We are grateful to the Material Characterization (http://www.uff.br/lamate/), Molecular Spectroscopy (http://www.uff.br/lame/), NMR (http://www.laremn.uff.br) and X-Ray Diffraction (http://www.ldrx.uff.br) Multiuser Laboratories from Universidade Federal Fluminense. The authors also thank MSc Alan Moraes for assistance with the TEM images, Dr Pedro Pablo Florez Rodrigues for confocal Raman measurements and Prof. Priscilla Finotelli for helpful discussions.

Notes and references

1. A. P. Subramanian, S. K. Jaganathan, A. Manikandan, K. N. Pandiaraj, N. Gomathi and E. Supriyanto, RSC Adv., 2016, 6, 48294–48314.
2. M. Mahmoudi, S. Sant, B. Wang, S. Laurent and T. Sen, Adv. Drug Delivery Rev., 2011, 63, 24–46.
3. C. Rejeeth, R. Vivek and S. Kannan, RSC Adv., 2015, 5, 94534–94538.
4. A. Hervault, A. E. Dunn, M. Lim, C. Boyer, D. Mott, S. Maenosono and N. T. K. Thanh, Nanoscale, 2016, 8, 21–24.
5. G. Q. Silveira, M. D. Vargas and C. M. Ronconi, J. Mater. Chem., 2011, 21, 6034–6039.
6. D. Tarn, D. P. Ferris, J. C. Barnes, M. W. Ambrogio, J. F. Stoddart and J. I. Zink, Nanoscale, 2014, 6, 3335–3343.
7. C. Wang, L. Huang, S. Song, B. Saif, Y. Zhou, C. Dong and S. Shuang, Appl. Surf. Sci., 2015, 357, 2077–2086.
8. C. Yang, W. Guo, N. An, L. Cui, T. Zhang, R. Tong, Y. Chen, H. Lin and F. Qu, RSC Adv., 2015, 5, 80728–80738.
9. N. Griffete, J. Fresnais, A. Espinosa, C. Wilhelm, A. Bée and C. Ménager, Nanoscale, 2015, 7, 18891–18896.
10. L. Gutiérrez, S. Romero, G. B. Da Silva, R. Costo, M. D. Vargas, C. M. Ronconi, C. J. Serna, S. Veintemillas-Verdaguer and M. Del Puerto Morales, Biomed. Tech., 2015, 60, 417–425.
11. M. Iafisco, C. Drouet, A. Adamiano, P. Pascaud, M. Montesi, S. Panseri, S. Sarda and A. Tampieri, J. Mater. Chem. B, 2016, 4, 57–70.
12. R. A. Bohara, N. D. Thorat and S. H. Pawar, RSC Adv., 2016, 6, 43989–44012.
13. G. S. Demirer, A. C. Okur and S. Kizilel, J. Mater. Chem. B, 2015, 3, 7831–7849.
14. C. Sun, K. Du, C. Fang, N. Bhattarai, O. Veiseh, F. Kievit, Z. Stephen, D. Lee, R. G. Ellenbogen, B. Ratner and M. Zhang, ACS Nano, 2010, 4, 2402–2410.
15. M. Marciello, V. Connord, S. Veintemillas-Verdaguer, M. A. Vergés, J. Carrey, M. Respaud, C. J. Serna and M. P. Morales, J. Mater. Chem. B, 2013, 1, 5995–6004.
16. G. B. da Silva, M. Marciello, M. P. Morales, C. J. Serna, M. D. Vargas, C. M. Ronconi and R. Costo, J. Braz. Chem. Soc., 2017, 28, 731–739.
17. G. Prabha and V. Raj, J. Magn. Magn. Mater., 2016, 408, 26–34.
18. Y.-M. Zhang, Y. Cao, Y. Yang, J.-T. Chen and Y. Liu, Chem. Commun., 2014, 50, 13066–13069.
19. G. Q. Silveira, R. S. Da Silva, L. P. Franco, M. D. Vargas and C. M. Ronconi, Microporous Mesoporous Mater., 2015, 206, 226–233.
20. J. Zhu, J. He, X. Du, R. Lu, L. Huang and X. Ge, Appl. Surf. Sci., 2011, 257, 9056–9062.
21. K. Hayashi, K. Ono, H. Suzuki, M. Sawada, M. Moriya, W. Sakamoto and T. Yogo, ACS Appl. Mater. Interfaces, 2010, 2, 1903–1911.
22. S. Licata, A. Saponiero, A. Mordente and G. Minotti, Chem. Res. Toxicol., 2000, 13, 414–420.
23. L. Huang, H. Wang, B. Li, E. Li, Y. Zhou, Y. Yang, C. Dong and S. Shuang, J. Inclusion Phenom. Macrocyclic Chem., 2014, 1–7.
24. P. Chen, H. Song, S. Yao, X. Tu, M. Su, L. Zhou and S. M. Liu, RSC Adv., 2017, 7, 29025–29034.
25. P. Guardia, R. Di Corato, L. Lartigue, C. Wilhelm, A. Espinosa, M. Garcia, F. Gazeau, L. Manna and T. Pellegrino, ACS Nano, 2012, 6, 3080–3091.
26. R. Hergt, S. Dutz, R. Michael and M. Röder, J. Phys.: Condens. Matter, 2008, 20, 385214.
27. E. C. S. Santos, T. C. dos Santos, R. B. Guimarães, L. Ishida, R. S. Freitas and C. M. Ronconi, RSC Adv., 2015, 5, 48031–48038.
28. M. Ma, Y. Zhang, W. Yu, H. Y. Shen, H. Q. Zhang and N. Gu, Colloids Surf., A, 2003, 212, 219–226.
29. G. Q. Silveira, C. M. Ronconi, M. D. Vargas, R. A. S. San Gil and A. Magalhães, J. Braz. Chem. Soc., 2011, 22, 961–967.
30. R. C. Petter, J. S. Salek, C. T. Sikorski, G. Kumaravel and F. Lin, J. Am. Chem. Soc., 1990, 112, 3860–3868.
31. G. H. Gaber Ahmed, R. Badía Laíño, J. A. García Calzón and M. E. Díaz García, Microchim. Acta, 2014, 181, 941–948.
32. G. Tripodo, C. Wischke, A. T. Neffe and A. Lendlein, Carbohydr. Res., 2013, 381, 59–63.
33. B. Atashkar, A. Rostami and B. Tahmasbi, Catal. Sci. Technol., 2013, 3, 2140.
34. Y. Liu, Y. Li, X. M. Li and T. He, Langmuir, 2013, 29, 15275–15282.
35. Raman Spectroscopy for Nanomaterials Characterization, ed. C. S. S. R. Kumar, Springer-Verlag, Baton Rouge, LA, 2012.
36. R. S. Benjamin, C. E. Riggs and F. R. Bachur, Clin. Pharmacol. Ther., 1973, 14, 592–600.
37. N. Husain, T. T. Ndou, A. M. de la Peña and I. M. Warnert, Appl. Spectrosc., 1992, 46, 652–658.
38. N. V. Roik and L. A. Belyakova, Interface Focus, 2016, 6, 1–10.
39. K. Nawara, J. Romiszewski, K. Kijewska, J. Szczytko, A. Twardowski, M. Mazur and P. Krysinski, J. Phys. Chem. C, 2012, 116, 5598–5609.
40. Q. Da Hu, G. P. Tang and P. K. Chu, Acc. Chem. Res., 2014, 47, 2017–2025.
41. S. Dutta, S. Parida, C. Maiti, R. Banerjee, M. Mandal and D. Dhara, J. Colloid Interface Sci., 2016, 467, 70–80.
42. S. Nigam, K. C. Barick and D. Bahadur, J. Magn. Magn. Mater., 2011, 323, 237–243.
43. E. Munnier, S. Cohen-Jonathan, C. Linassier, L. Douziech-Eyrolles, H. Marchais, M. Soucé, K. Hervé, P. Dubois and I. Chourpa, Int. J. Pharm., 2008, 363, 170–176.
44. R. Hergt, S. Dutz and R. Michael, J. Phys.: Condens. Matter, 2008, 20, 385214.
45. A. Riedinger, P. Guardia, A. Curcio, M. A. Garcia, R. Cingolani, L. Manna and T. Pellegrino, Nano Lett., 2013, 13, 2399–2406.
46. Z. Nemati, J. Alonso, H. Khurshid, M. H. Phan and H. Srikanth, RSC Adv., 2016, 6, 38697–38702.
47. F. Shubitidze, K. Kekalo, R. Stigliano and I. Baker, J. Appl. Phys., 2015, 117, 1–12.
48. A. E. Deatsch and B. A. Evans, J. Magn. Magn. Mater., 2014, 354, 163–172.
49. R. A. El-Awady, M. H. Semreen, M. M. Saber, F. Cyprian, V. Menon and T. H. Al-Tel, DNA Repair, 2016, 37, 1–11.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nj02860a

---