Folic acid decorated magnetic nanosponge: An efficient nanosystem for targeted curcumin delivery and magnetic resonance imaging

Journal of Colloid and Interface Science 556 (2019) 128–139

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Folic acid decorated magnetic nanosponge: An efficient nanosystem for targeted curcumin delivery and magnetic resonance imaging Elham Gholibegloo a,b, Tohid Mortezazadeh c, Fatemeh Salehian a, Hamid Forootanfar d, Loghman Firoozpour a, Alireza Foroumadi a, Ali Ramazani b,⇑, Mehdi Khoobi a,e,⇑ a

The Institute of Pharmaceutical Sciences (TIPS), Tehran University of Medical Sciences, Tehran 1417614411, Iran Department of Chemistry, Faculty of Science, University of Zanjan, Zanjan, Iran Department of Medical Physics, School of Medicine, Tabriz University of Medical Science, Tabriz, Iran d Food, Drug, and Cosmetics Safety Research Center, Kerman University of Medical Sciences, Kerman, Iran e Department of Pharmaceutical Biomaterials and Medical Biomaterials Research Center, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran b c

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 8 May 2019 Revised 11 August 2019 Accepted 12 August 2019 Available online 13 August 2019 Keywords: Theranostic system Cyclodextrin nanosponge Magnetic resonance imaging Folic acid Curcumin

a b s t r a c t Magnetic drug delivery system is one of the most important strategies for cancer diagnosis and treatment. In this study, a novel theranostic system was fabricated based on cyclodextrin nanosponge (CDNS) polymer anchored on the surface of Magnetite nanoparticles (Fe3O4/CDNS NPs) which was then decorated with folic acid (FA) as a targeting agent (Fe3O4/CDNS-FA). Curcumin (CUR), a hydrophobic model drug, was next loaded into the cyclodextrin cavity and polymeric matrix of CDNS (Fe3O4/CDNSFA@CUR). The system was fully characterized. The in vitro release study revealed pH-sensitive behavior. Cytotoxicity assays indicated a negligible toxicity for CUR free Fe3O4/CDNS-FA NPs against both of M109 cancerous cells and MCF 10A normal cells. CUR-loaded Fe3O4/CDNS-FA NPs exhibited higher toxicity against M109 cancerous cells than MCF 10A normal cells (p < 0.05). Fe3O4/CDNS-FA@CUR NPs resulted in much more cell viability on normal cells than pure CUR (p < 0.05). Moreover, blood compatibility study showed minor hemolytic activity. In vitro MRI studies illustrated negative signal increase in cells affirming acceptable diagnostic ability of the nanocarrier. The T2 MR signal intensity for Fe3O4/CDNS-FA@CUR NPs in M109 cells was around 2-fold higher than that of MCF 10A cells. This implies two times higher

⇑ Corresponding authors. E-mail addresses: [email protected] (A. Ramazani), [email protected] (M. Khoobi). https://doi.org/10.1016/j.jcis.2019.08.046 0021-9797/Ó 2019 Elsevier Inc. All rights reserved.

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

129

selective cellular uptake of the Fe3O4/CDNS-FA@CUR NPs into M109 cell compared to MCF 10A. The multifunctional nanocarrier could be considered as promising candidate for cancer theranostics because of the smart drug release, selective cytotoxicity, suitable hemocompatibility, and proper T2 MRI contrast efficiency. Ó 2019 Elsevier Inc. All rights reserved.

1. Introduction In recent years, a great attention has been paid to the synthesis of Magnetite nanoparticles (Fe3O4 NPs) because of their potential application in cancer diagnosis and treatment [1–3]. This potential is due to the low toxicity, easy production and separation, targeting capability and excellent magnetic characteristics of Fe3O4 NPs [4–6]. Fe3O4 NPs are one of the well-known negative contrast agents having the ability to produce improved visualization of cellular and sub-cellular compartments [7,8]. However, high surface area and small size of naked Fe3O4 NPs can result in aggregation reducing their magnetism and dispersibility [9,10]. To improve their colloidal stability, bioavailability, cytocompatibility and anticancer drug loading capacity, Fe3O4 NPs are decorated using biocompatible compounds and natural polymers such as chitosan, starch, dextran, cellulose, alginate and cyclodextirns [11–16]. Cyclodextrin (CD) is one of the organic supramolecules widely applied as hydrophobic drug trapping agent. The lipophilic central cavity of CD provides a perfect media for host-guest complexation between CD and different hydrophobic anticancer drugs. The hydrophilic character of the outer space of CD is related to the existence of high hydroxyl groups providing enhanced aqueous solubility and stability for hydrophobic anticancer drugs [17,18]. Furthermore, hydroxyl groups of CD participate in polymerization with various bi-functional cross-linking agents such as diisocyanates, dianhydrides, and other activated carbonyls [19,20]. Epiclon B-4400 (5-(2, 5-dioxotetrahydrofuryl)-3-methyl3-cyclohexene-1, 2-dicarbooxylic anhydride, EPI) is an inexpensive alicyclic dianhydride having high biocompatibility for biological applications [21] compared to the toxic aromatic dianhydrides. The polymerization process is performed during less than 5 min between EPI and bCD [22], while this time is around 3 h for ethylenediaminetetraacetic acid dianhydride (EDTA) and pyromellitic dianhydride (PMA) [23,24]. These valuable features make EPI as one of the best cross-linking agents for preparation of CD-based nanosponges (CDNSs). CDNSs provide porous and three-dimensional polymeric structure bearing improved surface area and lipophilic drug loading capability compared to the bare CDs [25]. In addition, the CDNSs decrease the diffusion rate of loaded drug and provide a sustained drug release more than free CD, which is due to the high porosity produced by polymeric network [26,27]. Interestingly, CDNSs are coated on the surface of Fe3O4 NPs without the need for prefunctionalization or applying proper cross-linker [28,29]. This can be attributed to the existence of large amount of carbonyl and hydroxyl groups in the structure of CDNSs having high ability to complex and graft on the surface of Fe3O4 NPs. Notwithstanding all outstanding features resulted from nanonization of the anticancer agents and proper surface functionalization of the nanocarrier boosting high accumulation of the drug in cancer tissues, targeted delivery is an important factor for perfect cancer therapy and diagnosis. Among various active targeting ligands applied for targeted delivery, Folic acid (FA) is one the most optimal ligands for selective localization of therapeutic or diagnostic agents in cancer cells. This is due to its low cost and toxicity, as well as high stability without harmful efficacy on normal tissue [30]. High affinity of FA to folate receptors on the surface of wide

range of cancerous cells makes this active targeting agent as first candidate in several studies [31]. Recently we introduced a nontoxic CDNS bearing significant capability for Curcumin (CUR) loading with the ability of pH sensitive release of the drug making the polymer as a perfect nanocarrier in cancer therapy [22]. Herein, EPI was exploited as crosslinking agent for the preparation of magnetic CDNS. The myriad amount of carboxylic acid groups originated from EPI after polymerization could be employed as capping agent for stabilization of Fe3O4 NPs as well as folic acid-hydrazide conjugation. Fe3O4 NPs were served as core endowing the target system with T2 contrast ability in magnetic resonance imaging (MRI). The CDNS with high capacity for CUR loading and controllable release of the drug was coated on the surface of Fe3O4 NPs to improve the solubility, stability and biocompatibility of Fe3O4 NPs. FA was finally conjugated to Fe3O4/CDNS NPs providing a smart drug vehicle for selective chemotherapy. CUR loading capacity and release profile, cytotoxicity of the samples against positive folate-receptor cancerous cell (M109) and negative folate-receptor normal cell (MCF 10A), and blood-compatibility of the samples were fully studied. In vitro MRI studies were also performed to evaluate the potential of the designed targeted magnetic CDNS in MRI. Although, pure CDNSs have been employed as a nanocarrier in drug delivery [22,25,26], there is no study related to the magnetic CD-based nanosponge (Fe3O4/CDNS) for therapeutic and diagnostic applications.

2. Experimental 2.1. Materials Ferric chloride hexahydrate (
98%, FeCl36H2O), ferrous chloride tetrahydrate (
99%, FeCl24H2O), b-cyclodextrin (
97%, bCD), 5-(2, 5-dioxotetrahydrofuryl)-3-methyl-3-cyclohexene-1, 2-dicarbooxylic anhydride (
98%, Epiclon B-4400, EPI), 1-ethyl-3(3-dimethylaminopropyl) carbodiimide (
98%, EDC), Nhydroxysuccinimide (98%, NHS), folic acid (
97%, FA), curcumin (
98%, CUR), ethylenediaminetetraacetic acid (99%, EDTA), ammonia (25%, NH4OH), triethylamine (
99.5%, Et3N), dimethyl sulfoxide (
99%, DMSO), hydrazine hydrate (50–60%, NH2.NH2), Tween-80 (a non-ionic surfactant), phosphate buffered saline powder (pH = 7.4, PBS) and acetone (
99.5%) were obtained from Merck (Darmstadt, Germany). Thiazolyl Blue Tetrazolium Bromide (98%, MTT) was purchased from Sigma–Aldrich (St. Louis, MO, USA). All chemicals were used without further purification. 2.2. Cell culture Normal human mammary epithelial cell line (MCF 10A), as a folate receptor-negative cell, and Madison lung carcinoma cell line (M109), as a folate receptor-positive cell from the National Cell Bank of Iran (Pasteur Institute; Iran) were provided and grown at 37 °C in moist condition under 5% CO2. MCF 10A and M109 cells were cultured in DMEM and RPMI 1640 (Gibco, Germany) media, respectively, supplemented with 10% heat-inactivated fetal bovine serum (FBS, Gibco), penicillin and streptomycin.

130

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

2.3. Synthesis of Fe3O4 NPs Fe3O4 NPs were synthesized by co-precipitation procedure [32]. Typically, FeCl36H2O (2 mmol) and FeCl24H2O (1 mmol) were dissolved in 45 mL of deionized (DI) water. Under vigorous stirring by a mechanical stirrer and argon atmosphere, 30 mL of NH4OH solution was added dropwise to the iron solution. The mixture of reaction was stirred at 60 °C for 1 h. The black product was separated by an external magnet and washed with DI water and ethanol three times. Finally, it was dried at 60 °C for 12 h. 2.4. Synthesis of CDNS Briefly, anhydrous bCD (1 mmol) was dissolved in anhydrous DMSO (4 mL) including 1 mL of anhydrous Et3N. EPI (8 mmol) was then added under magnetic stirring. The polymerization reaction was completed after five minutes. The solid product was crashed with spatula and rinsed with acetone in a Soxhlet apparatus within 24 h. Finally, the yellowish product (CDNS) was dried under vacuum [22,24]. 2.5. Synthesis of Fe3O4 NPs coated with CDNS

2.9. Loading of CUR into the Fe3O4/CDNS-FA NP CUR was loaded in Fe3O4/CDNS-FA NPs according to the reported method [36]. 2 mg of the NPs were dispersed in PBS (1 mL, pH = 7.4). A solution of CUR in acetone (1 mL containing 2 mg of CUR) was added dropwise to the suspension of the NPs, and the resulting mixture was shaken at room temperature overnight in dark. The final product was separated by external magnet. Then, it was washed two times with DI water to eliminate unloaded CUR. To calculate CUR loading efficiency and capacity, the content of the unloaded CUR in the supernatant was obtained from a calibration curve of CUR standard concentrations using an UV–Visible spectroscopy at 420 nm. CUR loading ratios of the samples were determined as below equations (2 and 3):

CUR loading efficiency ðLE; %Þ ¼ ðmass of loaded CUR=mass of initial CURÞ  100

ð2Þ

CUR loading capacity ðLC; %Þ

Fe3O4 NPs (1 g) were initially dispersed in DMSO (10 mL). bCD (1 mmol) and Et3N (1 mL) were then added to the latter dispersion. It was stirred at room temperature for 3 h. EPI (8 mmol) was next added and the reaction was stirred for more 3 h. Fe3O4 NPs coated with CDNS (Fe3O4/CDNS NPs) were isolated with external magnet and washed with ethanol three times and dried in vacuum for 24 h [28,29]. 2.6. Back-titration assay Quantification of the carboxyl groups on the surface of Fe3O4/ CDNS NPs was performed by back titration method [33]. Firstly, Fe3O4/CDNS NPs (0.1 g) were dispersed in a solution of sodium hydroxide (NaOH 0.1 N, 100 mL) and stirred for 6 h. Then, the remaining NaOH were determined via titration using hydrochloric acid solution (HCl 0.1 N, 100 mL). The amount of carboxyl (COOH) groups on the surface of Fe3O4/CDNS NPs was evaluated according to the following formula (1):

Number of COOH groups ¼ ððVb  Va Þ  M  1000Þ=m

was added and the reaction was kept on for 6 h [35]. The product (Fe3O4/CDNS-FA NPs) was separated by external magnet, rinsed with ethanol three times and dried in vacuum for 24 h.

ð1Þ

where Vb is the total volume of NaOH solution (mL), Va is the consumption volume of HCl solution (mL), M is concentration of NaOH solution (mol L1), and m is the weight of sample (g). 2.7. Synthesis of FA-hydrazide (FA-NH-NH2) Firstly, FA (1 g) was dissolved in 20 mL of DMSO. Excess amount of NHS and EDC (in 1:1 M ratio) were then added to the solution. The solution was stirred for 24 h at room temperature under inert atmosphere. To prepare FA-hydrazide, the activated FA solution was added to 2 mL of hydrazine hydrate under constant stirring for 6 h. Folate-hydrazide was converted into hydrochloride salt using addition of 6 mL of HCl solution (0.5 M) and precipitated by (1:1) diethyl ether/acetonitrile. The sediment was centrifuged, redissolved in a little amount of water and reparticipated with ethanol. Finally, the yellow pellet was washed with ethanol and diethyl ether, respectively, and dried under vacuum [34]. 2.8. Conjugation of FA-hydrazide to Fe3O4/CDNS NPs (Fe3O4/CDNS-FA) Fe3O4/CDNS NPs (1 g) were dispersed in DMSO (20 mL). EDC (6 mmol) and NHS (6 mmol) were added and the solution was stirred at room temperature for 24 h. Then, FA-hydrazide (6 mmol)

¼ ðmass of loaded CUR=mass of the final productÞ  100 ð3Þ 2.10. In vitro CUR release assay The release profile of CUR from Fe3O4/CDNS-FA@CUR NPs was assessed using the reported method [36,37]. 5 mg of Fe3O4/ CDNS-FA@CUR NPs was immersed into 10 mL of PBS solution containing Tween-80 (0.1% w/v) at pH = 5.5, or 7.4. The release assay was carried out at 37 °C in dark condition by shaking at 100 rpm. Sampling was performed at predestined time points. In each time, the samples were centrifuged at 10,000 rpm for 10 min and removed supernatant was replaced with the same volume of fresh buffer solution. The cumulative CUR release was determined by UV–visible spectrophotometry (at 420 nm). Standard solution of CUR was used to create the calibration curve of the released CUR. 2.11. MTT cell viability assay Cytotoxicity of Fe3O4/CDNS-FA NPs, Fe3O4/CDNS-FA@CUR NPs and free CUR was determined by MTT assay against MCF 10A and M109 cell lines as normal and cancerous cells, respectively. MCF 10A and M109 cells were seeded in 96-well plates at the density of 5  10 3 cells per well and culture mediums of RPMI and DMEM (200 lL). The cells were incubated for 24 h to attach properly. The media were then replaced with fresh media containing Fe3O4/CDNS-FA NPs, Fe3O4/CDNS-FA@CUR NPs, and free CUR (solved in DMSO with safety concentration of 0.1%) at different concentration (0.1–100 lg mL1) with a total volume of 200 lL. In the next step, the plates were incubated at 37 °C for 48 h. After that, 100 lL of PBS containing MTT (0.5 mg mL1) was poured to each well. The cells were incubated for 4 h and 100 lL of DMSO was then added into each well. In the next step, Bio Tek microplate reader (USA) was applied to read the absorbance of each well at 570 nm (test wavelength) and 630 nm (reference wavelength) [38,39]. Cell viability was calculated via the below formula (4):

Cell Viabilityð%Þ ¼ ½ðOD sample  OD BlankÞ=ð

ODsample ðmeanÞ  ODBlank ðmeanÞ ODControl ðmeanÞ  ODBlank ðmeanÞ

 100 OD Control  OD BlankÞ  100 ð4Þ

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

2.12. Relaxivity measurements T2-weighted phantom MRI images were obtained at dispersion of Fe3O4 NPs and Fe3O4/CDNS-FA@CUR NPs in DI water with various concentrations of 0.04, 0.08, 0.16, 0.32, and 0.64 mM, (amount of Fe was determined by ICP analysis). Phantom agar gel of samples were prepared in agar solution (2% w/v). The T2-weighted images were evaluated by a multi-spin echo (MSE) sequence with following parameters: repetition times (TR) = 2500 ms, diverse echo time (TE) = 13, 28, 39, 52, 66, 79, 92, 105 and 118 ms, slice thickness = 3 mm, field of view (FOV) = 250 mm  250 mm, and matrix = 384  512. The magnetic relaxation measurements were carried out via a 3.0 T human clinical scanner by head coil at room temperature and aqueous dispersion of the NPs. 2.13. In vitro MRI MCF 10A and M109 cells (5  106 cells well1) were incubated with Fe3O4/CDNS-FA@CUR NPs and Fe3O4 NPs (control) in different Fe concentrations (0, 1 and 10 lg mL1) for 6 h at 37 °C to study the targeting ability of the NPs. After incubation, the cells were washed with PBS three times and resuspended in PBS with a cell density of 1  106 cells mL1. All MRI measurements were performed with a 3.0 T system (3 T Siemens Prisma). T2-weighted MRI using a fast spin-echo sequence were used to reduce acquisition time under the following parameters: TR/TE = 3600/90 ms, 220  320 matrix, 82  120 mm2 field of view, 220 Hz/Px bandwidth and slice thickness of 3 mm. 2.14. Hemolysis assay The hemolysis assay of Fe3O4/CDNS-FA@CUR NPs was tested according to the reported methods [2,40]. The fresh human blood was stabilized with EDTA, centrifuged at 4 °C, 2000 rpm for 10 min to separate the red blood cells (RBCs). The RBCs precipitate was rinsed 5 times with PBS (pH = 7.4). RBCs suspension was prepared and diluted 10 times with PBS, and then 200 lL of RBCs was added to 800 lL of different concentrations of the sample solution (1.95–2000 lg mL1). To obtain the positive or negative controls, 200 lL of RBCs was added to 800 lL of Triton X100 (2% v/v) or PBS, respectively. Subsequently, all specimens were shaken at 37 °C for 2 h, moderately. The samples were centrifuged at 8000 rpm for 2 min, and the hemoglobin absorption in the supernatant was investigated by UV–visible spectrophotometer at 541 nm. The hemolytic percentage was calculated according to the following formula (5):


Hemolysisð%Þ ¼ ½ Asample  Actrl =ðActrlþ  Actrl Þ  100

ð5Þ

2.15. Characterization instrument Fourier Transform Infrared (FT-IR) spectra were recorded using FT-IR spectrometer (Magna 550, Nicolet) at the wavelength range of 400–4000 cm1 with the KBr pellet method. 1H and 13C nuclear magnetic resonance (NMR) spectra were obtained using Avance III ultrasheild NMR spectrometer (Bruker, 500 MHz) using deuterated-D2O. The X-ray diffraction (XRD) patterns were collected via an X´Pert Pro MPD (PANalytical Co.) diffractometer with Cu-Ka radiation source (k = 1.54060 Å) in the range of 2h = 25–80°. Thermogravimetric analysis (TGA) was performed by simultaneous thermal analyzer SDT-Q600 V20.9 Build 20 at the heating rate of 10 °C/min under inert condition. UV–Visible spectra were recorded by a spectrophotometer (UV–Visible Jasco-530) in the range of k = 200–500 nm. Inductively coupled plasma mass spectrometry, ICP-MS (7900, Agilent) was utilized to determine elemental

131

composition. The zeta potential and hydrodynamic size distribution of the prepared samples were measured by a ZEN3600 Zetasizer (Malvern Instruments) in DI water at room temperature. Transmission Electron Microscopy (TEM) and Field Emission Scanning Electron Microscopy (FE-SEM) images were taken with CM30, Philips and VEGA-II TESCAN, respectively. Vibrating Sample Magnetometer (VSM, Meghnatis Kavir Kashan Co., Kashan, Iran) with a maximum magnetic field of 8 kOe at 25 °C was used to detect of the magnetic property of the samples. 2.16. Statistical analysis The one-way analysis of variance (ANOVA) and Student’s t-test were applied to evaluate the obtained data statistically. The p < 0.05 value was accepted for statistical significance. All experiments were performed at least three times, and the results were reported as mean ± standard deviation (SD). 3. Results and discussion The sequential steps for the synthesis of Fe3O4/CDNS-FA@CUR NPs are shown in Fig. 1. Firstly, Fe3O4 NPs were prepared via coprecipitation method. The surface of Fe3O4 NPs was coated and functionalized with CDNS polymeric network. CDNS was prepared by polymerization of bCD and EPI as a cross-linking agent. Fe3O4/ CDNS NPs containing a large number of carboxylic acid groups were obtained. FA was converted to folic acid-hydrazide and then conjugated to Fe3O4/CDNS NPs to form Fe3O4/CDNS-FA NPs. Finally, CUR, as an anticancer drug, was loaded into the hydrophobic cavity of CD and porous network of CDNS (Fe3O4/CDNSFA@CUR NPs). 3.1. FT-IR Fig. 2 shows FT-IR spectra of Fe3O4 NPs, free CDNS, Fe3O4/CDNS NPs and Fe3O4/CDNS-FA NPs. The sharp adsorption peak at 582 cm1 is the characteristic of Fe-O bond, and the bands at 3409 and 1617 cm1 are corresponded to the stretching and bending vibration of OAH groups, respectively [29,32]. The FT-IR spectrum of free CDNS shows two main peaks at 1730 and 1574 cm1 corresponding to the carbonyl (C@O) ester bonds and carboxylate groups, respectively. These bonds could be assigned to the opening of EPI anhydride ring [22,29,41]. The spectrum of Fe3O4/CDNS NPs indicates a band at 1710 cm1 attributed to the carbonyl ester bond in CDNSs. This bond has been shifted to the lower frequencies compared with free CDNS, which could be due to chelation of Fe3O4 NPs with oxygen atoms in the CDNS network. Also, in the spectrum of Fe3O4/CDNS-FA NPs, the peak at 1646 cm1 could be attributed to the new carbonyl bond (C@O) produced by the chemical reaction between carboxylic acid groups of CDNS structure and hydrazide groups of FA-NH-NH2 [39]. 3.2. NMR analysis The 1H and 13C NMR spectra of CDNS are shown in Fig. S1a and b, respectively. The peaks related to the anomeric proton of bCD (H1) and vinylic proton of EPI (Ha) were observed at 5.01 and 5.50 ppm, respectively. The rest of the protons related to the bCD moiety (H2-H6) were appeared at 3.26–3.79 ppm. In addition, the peaks related to the aliphatic protons of EPI (Hb-Hh) were appeared as multiplet signals in the range of 1.66 to 2.60 ppm [42,43]. The carboxylic acid groups formed during the synthesis of CDNS could be stabilized by triethylammonium cations. The presence of triplet–quartet pattern at 1.15 ppm and 3.07 ppm having high intensity could be related to the ethyl group of the triethylammonium

132

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

Fig. 1. Sequential steps for the preparation of Fe3O4/CDNS-FA@CUR NPs.

carbons were appeared in the region from 21.1 to 71.8 ppm and the peak related to the anomeric carbon was observed at 102.0. The peaks observed in the range of 123.4–146.7 ppm could be assigned to the vinylic carbons of EPI. The peaks appeared at 174.1–179.7 ppm could be attributed to the carbonyl carbons of the ester or carboxylic acid groups in the structure of CDNS [44]. The presence of various peaks could be due to the asymmetrical structure of EPI and the possibility of the EPI cross-linking with different hydroxyl groups of bCD providing various electronic environments for the carbons resulting in difference in chemical shifts. 3.3. XRD

Fig. 2. FT-IR spectra of Fe3O4 NPs, free CDNS, Fe3O4/CDNS NPs and Fe3O4/CDNS-FA NPs.

cations (Et3NH+) as counterions of high amount of produced carboxylate groups in the structure of the CDNS [20]. 13CNMR spectrum also confirmed the successfully formation of the CDNS. The presence of the peaks at 8.2 and 46.7 ppm could be ascribed to the ethyl groups of the triethylammonium cations. All aliphatic

The X-ray diffraction patterns of the pure Fe3O4 NPs and Fe3O4/ CDNS NPs are seen in Fig. 3. The characteristic peaks of Fe3O4 NPs at 2h = 30.1, 35.6, 43.3, 54.0, 57.2, 62.9 and 74.4° are corresponded to the (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1), (4 4 0) and (6 2 0) crystal planes, respectively [45]. These peaks are suitably agreed with reflection patterns of JCPDS file no: 19-0629 corresponding to the inverse spinel cubic structure of Fe3O4 NPs. As shown in Fig. 3, the Bragg diffraction angles display the structure of Fe3O4 NPs that are retained in Fe3O4/CDNS NPs. Also, no peaks of impurities are revealed in the XRD pattern. 3.4. TGA TGA curves of the bare Fe3O4 NPs, free CDNS and Fe3O4/CDNS NPs were displayed in Fig. 4. The first weight loss of 1.6% below 150 °C in Fe3O4 thermogram can be related to the evaporation of

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

133

position for naked Fe3O4 NPs, which is comparable with the result achieved in this study (about 3% having the same pattern of decomposition). They also reported the weight loss of around 16% related to the CDNS decomposition, while it was 9% in our study. 3.5. VSM Fig. 5 illustrates magnetization curves of bare Fe3O4 NPs and Fe3O4/CDNS NPs between 8 kOe and +8 kOe at room temperature. The saturation magnetization (Ms) values of the Fe3O4 NPs and Fe3O4/CDNS NPs were 50 and 40 emu/g, respectively. The saturation magnetization of Fe3O4 NPs was higher than Fe3O4/CDNS NPs. This could be due to the existence of non-magnetic polymeric layer (CDNS) on the surface of Fe3O4 NPs [29]. In addition, no hysteresis loop in curves was observed, expressing superparamagnetic properties of the NPs [41]. 3.6. UV–Vis

Fig. 3. XRD patterns of Fe3O4 NPs and Fe3O4/CDNS NPs.

Fig. 4. TGA graphs of Fe3O4 NPs, free CDNS and Fe3O4/CDNS NPs.

physisorbed water. A partial weight gain can be seen in thermogram of bare Fe3O4 NPs over a range of 150 to 200 °C which may be assigned to the oxidation of Fe3O4 to Fe2O3. The final weight loss of 0.9% at the range of 200–600 °C could be ascribed to the phase transition of Fe3O4 to FeO. The same results were also obtained in several previously published works [46,47]. For free CDNS, the first weight loss of 5.5% was observed in the range of 25–140 °C, assigning to the vaporization of the hydroxyl groups. The second step (68.5%) at temperature above 140 °C can be attributed to the decomposition of CDNS structure. The same pattern of decomposition has been reported for CDNS prepared by diphenyl carbonate as cross-linking agent [48]. The first and second weight losses (about 3 and 9%, respectively) in thermogram of Fe3O4/CDNS NPs between 25 and 140 °C and 150–600 °C could be probably due to the desorption of water molecule and decomposition of the CDNS layer coated on the surface of magnetic nanoparticles, respectively. By comparison between thermograms of the samples, it could be concluded that amount of CDNS is 9% in the Fe3O4/CDNS. Liang et al was applied PMA as cross-linking agent for the preparation of CDNS and then Fe3O4 NPs were synthesized in the presence of the CDNS [29]. They reported the weight loss of 3.5% and three steps decom-

UV–Vis technique was used to distinguish the presence of FA on the surface of Fe3O4/CDNS NPs. As shown in Fig. 6, FA showed absorption peaks at 233, 280 and 360 nm. In the spectrum of Fe3O4/CDNS NPs, no characteristic peaks were appeared. However,

Fig. 5. Magnetization versus applied magnetic field for Fe3O4 NPs and Fe3O4/CDNS NPs.

Fig. 6. UV–Vis spectra of FA, Fe3O4/CDNS NPs and Fe3O4/CDNS-FA NPs.

134

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

the spectrum of Fe3O4/CDNS-FA NPs indicated k values at 235, 282 and 365 nm with a slight shift to the longer wavelengths compared with free FA [3,34]. The quantity of FA in Fe3O4/CDNS-FA NPs was determined 32.12%. This result confirms successfully conjugation of FA onto the surface of Fe3O4/CDNS NPs. 3.7. Dynamic light scattering (DLS) and zeta potential analyses Hydrodynamic particle size distribution of Fe3O4 NPs and Fe3O4/CDNS NPs was determined by DLS measurement. Mean sizes (diameter) of Fe3O4 NPs and Fe3O4/CDNS NPs were 42 nm and 68 nm, respectively. The CDNS layer enhanced the mean diameter of Fe3O4/CDNS NPs compared to bare Fe3O4 NPs. The low polydispersity index (PdI  0.2) of the prepared samples illustrated a narrow particle size distribution implying the formation of uniform NPs. The surface zeta potential values of Fe3O4 NPs, Fe3O4/CDNS NPs and Fe3O4/CDNS-FA NPs were 26.6, 35.7 and 21.4 mV, respectively. The negative charge value of Fe3O4/CDNS NPs was higher than that of bare Fe3O4 NPs. This could be assigned to the existence of the negatively hydroxyl and carboxyl groups of CDNS layer. It has been reported that negatively charged NPs have low tendency to interact with plasma proteins and hence show higher stability in blood circulation compared to the positively charged NPs [49,50].

3.10. CUR loading CUR, as a lipophilic anticancer drug, was loaded into the hydrophobic cavity of CD and mesoporous structure of the polymeric structure [22]. The UV–Vis spectra of CUR, and Fe3O4/ CDNS-FA@CUR NPs are seen in Fig. 8. An absorption band at 432 nm was observed for the solution of free CUR in DMSO. This band was also observed in the spectrum of Fe3O4/CDNS-FA@CUR NPs indicating the appropriate CUR loading into Fe3O4/CDNS-FA NPs. The best CUR loading values were achieved at a drug to carrier ratio of 1:2. CUR loading efficiency and capacity were 96% and 45%, respectively, showing valuable efficacy of the target system for CUR loading. 3.11. In vitro CUR release study Fig. 9 represents the release profile of CUR from Fe3O4/CDNSFA@CUR NPs at pH = 5.5 and 7.4 expressing endosomal compartments of cancer cells and biological media (bloodstream), respectively [53]. As shown in Fig. 9, release of CUR at pH = 5.5 was around 32%, 41% and 50% after 24, 72 and 120 h, respectively, whereas the release values were about 18%, 22% and 31% at pH = 7.4 after the same times. CUR release percentage under natural condition was lower than that of in acidic environment. This

3.8. FE-SEM and TEM analyses Figs. S3a and 7b present the FE-SEM images of Fe3O4 NPs and Fe3O4/CDNS NPs showing the spherical shape with proper size distribution. The mean size of Fe3O4 NPs and Fe3O4/CDNS NPs was determined 35 and 55 nm, respectively. However, there was a difference between the obtained size from DLS and FE-SEM. This could be due to the fact that DLS shows the hydrodynamic size of the particles and water swellability of polymers can increase the size of the particles; while FE-SEM shows the exact size of NPs in dried condition. Another morphological measurement was carried out by TEM analysis of Fe3O4 NPs and Fe3O4/CDNS NPs implying approximately spherical morphology of the target NPs (Figs. S3b and 7b) [51,52]. 3.9. The number of carboxyl groups on Fe3O4/CDNS NPs The number of carboxyl groups on the surface of Fe3O4/CDNS NPs was obtained by back titration method. The number of carboxyl groups was 0.3 mmol mg1.

Fig. 8. UV–Vis spectra of Fe3O4/CDNS-FA NPs, Fe3O4/CDNS-FA@CUR NPs and CUR.

Fig. 7. FE-SEM image of Fe3O4/CDNS NPs (a) and TEM image of Fe3O4/CDNS NPs (b).

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

behavior could be assigned to CUR protonation decreasing the interaction between CUR and CD hydrophobic cavity in the CDNS structure [54]. Microenvironment of cancerous tissues and intracellular organelles such as endosomes and lysosomes are mildly acidic, but normal tissues are natural [55]. Therefore, pH-dependent of drug release is appropriate in the treatment of cancers and suppression of drug side effects to healthy tissues.

Fig. 9. Release profiles of CUR from Fe3O4/CDNS-FA@CUR NPs at pH = 5.5 and 7.4. Data is expressed as mean ± S.D. (n = 3).

135

3.12. Toxicity To consider the targeting ability of the synthesized NPs, the in vitro toxicity of Fe3O4/CDNS-FA@CUR NPs was evaluated against MCF 10A normal cells having low expression of folate receptors and M109 cancerous cells bearing over expression of folate receptors [56]. The cell viability assays were determined at 24 and 48 h. The cytotoxicity of Fe3O4/CDNS-FA NPs (without anticancer agent) against both cell lines was low indicating biocompatibility of the carrier (Fig. 10). The least cell viability was higher than 80% at maximum concentration (100 lg mL1) after 48 h. The viability of both cell lines decreased dramatically with the enhancement of the incubation time and CUR content. As shown in Fig. 10b and 10d, the cell viability for both M109 and MCF 10A cells was about 15.2% and 15.6%, respectively, after 48 h treatment with CUR at concentration of 100 lg mL1; while these values for Fe3O4/CDNS-FA@CUR NPs were around 29.5% and 44.8%, respectively (p < 0.05). This higher toxicity of free CUR could be attributed to the hydrophobicity of free CUR leading to the more penetration through lipid cell wall [39,57]. In addition, the lower toxicity of Fe3O4/CDNS-FA@CUR NPs in comparison with free CUR could be appropriate for healthy cells and tissues. In addition, the cytotoxicity values of Fe3O4/CDNS-FA@CUR NPs on M109 cells (averagely 37.8% and 29.5% viability at top concentration after 24 and 48 h, respectively) was more than that of on MCF 10A cells (about 57.2% and 44.8% viability, after 24 and 48 h, respectively (p < 0.05)). This might be assigned to the higher CUR release from Fe3O4/CDNS-FA@CUR NPs in acidic environment of tumor cells than natural medium of normal ones. Furthermore, the more interaction of FA with high level of folate receptors on the surface of M109 cells could be another reason for higher inhibition rate of Fe3O4/CDNS-FA@CUR NPs, which led to its specific cell uptake through the mechanism of receptor-mediated endocytosis [31,58–60].

Fig. 10. Cytotoxicity of CUR, Fe3O4/CDNS-FA NPs and Fe3O4/CDNS-FA@CUR NPs measured by MTT assay in MCF 10A after: (a) 24 h, (b) 48 h; and M109 after: (c) 24 h, (d) 48 h incubation. The significance level was p < 0.05. Data is expressed as mean ± S.D. (n = 3).

136

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

3.13. Hemolysis assay The hemocompatibility analysis has been recognized as one of the most important assay to evaluate the biosafety of NPs on blood

erythrocytes. As shown in Fig. 11b, RBCs in the presence of different concentration of Fe3O4/CDNS-FA@CUR NPs behaves like to that of in PBS solution as a negative control. While, treatment of RBCs with Triton X100 solution (2% v/v) as positive control resulted in

Fig. 11. Hemolytic activity and hemolysis percentage of Fe3O4/CDNS-FA@CUR NPs (a), Photographs of RBCs treated with Fe3O4/CDNS-FA@CUR NPs at different concentrations (b). Data is expressed as mean ± S.D. (n = 3).

Fig. 12. T2 relaxivity plots of aqueous suspension of Fe3O4 NPs (a) and Fe3O4/CDNS-FA@CUR NPs (b). T2-weighted MR images of Fe3O4 NPs and Fe3O4/CDNS-FA@CUR NPs in DI water at 3.0 T MR system (c).

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

no precipitation implying severe hemolysis phenomenon. Fig. 11a depicts that percentage of erythrocytes lysis activity for various concentration of Fe3O4/CDNS-FA@CUR NPs (1.95–1000 lg mL1) was desirable and less than 5% (standard acceptance limit). Therefore, the prepared nanocarrier at the concentration level of 1000 lg mL1 revealed a suitable hemocompatibility [61,62]. 3.14. Relaxivities and map images The magnetic properties of the target system was evaluated using relaxivity and map images tests to estimate ability of the designed system as MRI contrast agent. Five Fe concentrations was applied to measure transverse (T2) relaxation times. The inverse relaxation times (1/T2) were depicted versus Fe concentration. The slopes of the curves were derived to obtain the transverse (r2) relaxivity of Fe3O4/CDNS-FA@CUR NPs and Fe3O4 NPs. The relaxivity values for Fe3O4 NPs and Fe3O4/CDNS-FA@CUR NPs were 64.95 and 54.28 s1 mM1, respectively, implying enough diagnosis capability of the prepared NPs for MRI applications [63]. Different parameters including the type and amount of the coating layer trapping Fe ions affect the r2 values. It has been reported that the polar groups like carbonyl and hydroxyl bonds in the hydrophilic shell surrounding the Fe3O4 NPs dedicate contrast agents with proper water availability [64]. It has been also revealed that the relaxation rate (r2) of magnetic NPs could be increased in the pres-

137

ence of color agent, such as CUR [36]. As shown in Fig. 12c, T2 map images show that Fe concentration affect directly on the relaxation rate of the water protons and subsequently on the contrast [65]. The ability of the bare CDNS in MRI was also studied. The free CDNS showed no ability as contrast agent in MRI. Changing in the concentration of CDNS did not alter the slop of signal intensity (SI) curve versus Echo Time (TE) and the slop of relaxivity curve was approximately zero (r2 = 0.0002) (Fig. S4a and b). As shown in Fig. S4c, enhancement of CDNS concentration had no effect on the signal intensity for images. 3.15. Targeted T2-weighted MRI of cancer cells in vitro MRI efficiency and cancer cell targeting capability of Fe3O4/ CDNS-FA@CUR NPs and Fe3O4 NPs were evaluated in the presence of normal and cancerous cells. M109 and MCF 10A cell lines were initially incubated with various concentrations of the NPs (0, 1 and 10 lg mL1) for 6 h. T2-weighted MRI and the MR signal intensities of the cells were then recorded (Fig. 13). As shown in Fig. 13a, the T2-weighted MR images of Fe3O4/CDNS-FA@CUR NPs showed a remarkable signal decline in M109 cells, while this decrease of signal was moderate in the case of MCF 10A cell. The quantitative measurement of the MR signal variation for Fe3O4/CDNS-FA@CUR NPs showed that the T2 MR signal intensity in M109 cells was 4.5 times lower than that of in the control group. However, the

Fig. 13. T2-weighted MR images of Fe3O4/CDNS-FA@CUR NPs in MCF 10A and M109 cells at different concentrations after incubation for 6 h on a 3 T MR system (a). Signal intensity analysis for T2-weighted MR images (b). The significance level was p < 0.05. Data is expressed as mean ± S.D. (n = 3).

138

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139

T2 MR signal intensity for MCF 10A cells treated with the NPs was only 2.5 times lower than that of in the same control groups (Fig. 13b). The considerable difference in the MR imaging between M109 and MCF 10A cells could be attributed to the selective cellular uptake of Fe3O4/CDNS-FA@CUR NPs into M109 cell with high folate receptor expression [38]. This study disclosed that specific interaction between FA on the surface of the NPs and folate receptor on the outer surface of M109 cell leads to the selective cell internalization of Fe3O4/CDNS-FA@CUR NPs as well as improved T2 MR signals [66]. In addition, T2-weighted MR images of Fe3O4 NPs showed similar behavior for both cells. This could be due to the fact that the cell uptake of Fe3O4 NPs was done only by the enhanced permeability and retention (EPR) mechanism and active targeting has no role in cell penetration.

[3]

[4]

[5]

[6]

[7] [8]

4. Conclusion [9]

In this study, a novel smart theranostic agent (Fe3O4/CDNSFA@CUR NPs) was prepared using surface modification of Fe3O4 magnetic NPs with CDNS-FA conjugate for targeted delivery of CUR and T2-weighted magnetic resonance imaging. Although a few studies have focused on the preparation of magnetic CDNSs and their application as adsorbent for contaminant removal [28,29] or as catalyst [67], to the best of our knowledge, this is the first study focusing on the application of novel magnetic CDNS as theranostic agent in cancer. Moreover, the introduced nanosystem exhibited acceptable drug loading capacity and release profile for cancer therapy. Cytotoxicity results showed that CUR-free nanocarrier had no apparent toxic effect against both cell lines. In addition, the proliferation inhibitory effect of Fe3O4/CDNSFA@CUR NPs against folate receptor-positive M109 cancerous line was more than that of against folate receptor-negative MCF 10A normal cell. The drug delivery system revealed excellent hemocompatibility, indicating its biocompatibility and safety for further studies. Furthermore, in vitro MRI findings disclosed that Fe3O4/ CDNS-FA@CUR NPs could be applied as a negative contrast agent because of sufficient magnetic characters and selective targeting capability. These valuable features suggest Fe3O4/CDNS-FA@CUR NPs as promising candidate for targeted other hydrophobic anticancer delivery. More in vivo studies on the capability of the nanosystem in tumor shrinking will be the objective of future research. Acknowledgments This study was supported financially by a grant from Tehran University of Medical Sciences (Grant number: 33124) and Iran Science Elites Federation.

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18] [19] [20] [21]

[22]

[23]

Disclosure statement No conflict of interest, financial or other, exists. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcis.2019.08.046.

[24]

[25]

[26]

References [1] A.M. Malekzadeh, A. Ramazani, S.J.T. Rezaei, H. Niknejad, Design and construction of multifunctional hyperbranched polymers coated magnetite nanoparticles for both targeting magnetic resonance imaging and cancer therapy, J. Colloid Interface Sci. 490 (2017) 64–73. [2] M. Ghorbani, B. Bigdeli, L. Jalili-baleh, H. Baharifar, M. Akrami, S. Dehghani, B. Goliaei, A. Amani, A. Lotfabadi, H. Rashedi, Curcumin-lipoic acid conjugate as a promising anticancer agent on the surface of gold-iron oxide nanocomposites:

[27]

[28]

a pH-sensitive targeted drug delivery system for brain cancer theranostics, Eur. J. Pharm. Sci. 114 (2018) 175–188. Q.L. Jiang, S.W. Zheng, R.Y. Hong, S.M. Deng, L. Guo, R.L. Hu, B. Gao, M. Huang, L. F. Cheng, G.H. Liu, Y.Q. Wang, Folic acid-conjugated Fe3O4 magnetic nanoparticles for hyperthermia and MRI in vitro and in vivo, Appl. Surf. Sci. 307 (2014) 224–233. Z. Gao, X. Liu, Y. Wang, G. Deng, F. Zhou, Q. Wang, L. Zhang, J. Lu, Facile one-pot synthesis of Fe3O4@chitosan nanospheres for MRI and fluorescence imaging guided chemo-photothermal combinational cancer therapy, Dalton Trans. 45 (2016) 19519–19528. J. Yu, X. Li, Y. Luo, W. Lu, J. Huang, S. Liu, Poly (ethylene glycol) shell-sheddable magnetic nanomicelle as the carrier of doxorubicin with enhanced cellular uptake, Colloids Surf. B. 107 (2013) 213–219. C. Tudisco, F. Bertani, M. Cambria, F. Sinatra, E. Fantechi, C. Innocenti, C. Sangregorio, E. Dalcanale, G. Condorelli, Functionalization of PEGylated Fe3O4 magnetic nanoparticles with tetraphosphonate cavitand for biomedical application, Nanoscale 5 (2013) 11438–11446. H.B. Na, I.C. Song, T. Hyeon, Inorganic nanoparticles for MRI contrast agents, Adv. Mater. 21 (2009) 2133–2148. B.R. Smith, S.S. Gambhir, Nanomaterials for in vivo imaging, Chem. Rev. 117 (2017) 901–986. X. Guo, W. Li, L. Luo, Z. Wang, Q. Li, F. Kong, H. Zhang, J. Yang, C. Zhu, Y. Du, External magnetic field-enhanced chemo-photothermal combination tumor therapy via iron oxide nanoparticles, ACS Appl. Mater. Interfaces. 9 (2017) 16581–16593. B. Shen, Y. Ma, S. Yu, C. Ji, Smart multifunctional magnetic nanoparticle-based drug delivery system for cancer thermo-chemotherapy and intracellular imaging, ACS Appl. Mater. Interfaces. 8 (2016) 24502–24508. Y. Ding, S.Z. Shen, H. Sun, K. Sun, F. Liu, Y. Qi, J. Yan, Design and construction of polymerized-chitosan coated Fe3O4 magnetic nanoparticles and its application for hydrophobic drug delivery, Mater. Sci. Eng. C. 48 (2015) 487– 498. X. Xu, T. Qu, L. Fan, X. Chen, M. Gao, J. Zhang, T. Guo, Preparation of pH-and magnetism-responsive sodium alginate/Fe3O4@ HNTs nanocomposite beads for controlled release of granulysin, RSC Adv. 6 (2016) 111747–111753. G. Liu, R. Hong, L. Guo, G. Liu, B. Feng, Y. Li, Exothermic effect of dextran-coated Fe3O4 magnetic fluid and its compatibility with blood, Colloids Surf. A. 380 (2011) 327–333. H. Hamidian, T. Tavakoli, Preparation of a new Fe3O4/starch-g-polyester nanocomposite hydrogel and a study on swelling and drug delivery properties, Carbohydr. Polym. 144 (2016) 140–148. R. Elumalai, S. Patil, N. Maliyakkal, A. Rangarajan, P. Kondaiah, A.M. Raichur, Protamine-carboxymethyl cellulose magnetic nanocapsules for enhanced delivery of anticancer drugs against drug resistant cancers, Nanomedicine 11 (2015) 969–981. C. Tudisco, V. Oliveri, M. Cantarella, G. Vecchio, G.G. Condorelli, Cyclodextrin anchoring on magnetic Fe3O4 nanoparticles modified with phosphonic linkers, Eur. J. Inorg. Chem. 2012 (2012) 5323–5331. J. Zhang, P.X. Ma, Cyclodextrin-based supramolecular systems for drug delivery: recent progress and future perspective, Adv. Drug Deliv. Rev. 65 (2013) 1215–1233. R. Challa, A. Ahuja, J. Ali, R. Khar, Cyclodextrins in drug delivery: an updated review, AAPS PharmSciTech 6 (2005) 329–357. R. Cavalli, F. Trotta, W. Tumiatti, Cyclodextrin-based nanosponges for drug delivery, J. Incl. Phenom. Macrocycl. Chem. 56 (2006) 209–213. F. Trotta, M. Zanetti, R. Cavalli, Cyclodextrin-based nanosponges as drug carriers, Beilstein J. Org. Chem. 8 (2012) 2091–2099. I. Stoica, A.I. Barzic, M. Butnaru, F. Doroftei, C. Hulubei, Surface topography effect on fibroblasts population on epiclon-based polyimide films, J. Adhes. Sci. Technol. 29 (2015) 2190–2207. E. Gholibegloo, T. Mortezazadeh, F. Salehian, A. Ramazani, M. Amanlou, M. Khoobi, Improved curcumin loading, release, solubility and toxicity by tuning the molar ratio of cross-linker to b-cyclodextrin, Carbohydr. Polym. 213 (2019) 70–78. B. Rossi, V. Venuti, A. Paciaroni, A. Mele, S. Longeville, F. Natali, V. Crupi, D. Majolino, F. Trotta, Thermal fluctuations in chemically cross-linked polymers of cyclodextrins, Soft Matter. 11 (2015) 2183–2192. M. Ferro, F. Castiglione, C. Punta, L. Melone, W. Panzeri, B. Rossi, F. Trotta, A. Mele, Anomalous diffusion of Ibuprofen in cyclodextrin nanosponge hydrogels: an HRMAS NMR study, Beilstein J. Org. Chem. 10 (2014) 2715– 2723. S. Swaminathan, L. Pastero, L. Serpe, F. Trotta, P. Vavia, D. Aquilano, M. Trotta, G. Zara, R. Cavalli, Cyclodextrin-based nanosponges encapsulating camptothecin: physicochemical characterization, stability and cytotoxicity, Eur. J. Pharm. Biopharm. 74 (2010) 193–201. C.P. Dora, F. Trotta, V. Kushwah, N. Devasari, C. Singh, S. Suresh, S. Jain, Potential of erlotinib cyclodextrin nanosponge complex to enhance solubility, dissolution rate, in vitro cytotoxicity and oral bioavailability, Carbohydr. Polym. 137 (2016) 339–349. V. Venuti, B. Rossi, A. Mele, L. Melone, C. Punta, D. Majolino, C. Masciovecchio, F. Caldera, F. Trotta, Tuning structural parameters for the optimization of drug delivery performance of cyclodextrin-based nanosponges, Expert Opin. Drug Deliv. 14 (2017) 331–340. S. Salazar, D. Guerra, N. Yutronic, P. Jara, Removal of aromatic chlorinated pesticides from aqueous solution using b-cyclodextrin polymers decorated with Fe3O4 nanoparticles, Polymers 10 (2018) 1038.

E. Gholibegloo et al. / Journal of Colloid and Interface Science 556 (2019) 128–139 [29] W.T. Liang, D. Li, X.W. Ma, W.J. Dong, J. Li, R.F. Wu, C. Dong, Q.C. Dong, Surface b-cyclodextrin polymer coated Fe3O4 magnetic nanoparticles: synthesis, characterization and application on efficient adsorption of malachite green, J. Nanoparticle Res. 54 (2018) 54–65. [30] A.M. Khattabi, W.H. Talib, D.A. Alqdeimat, A targeted drug delivery system of anti-cancer agents based on folic acid-cyclodextrin-long polymer functionalized silica nanoparticles, J. Drug Deliv. Sci. Tec. 41 (2017) 367–374. [31] N. Ma, J. Liu, W. He, Z. Li, Y. Luan, Y. Song, S. Garg, Folic acid-grafted bovine serum albumin decorated graphene oxide: an efficient drug carrier for targeted cancer therapy, J. Colloid Interface Sci. 490 (2017) 598–607. [32] H. Cai, X. An, J. Cui, J. Li, S. Wen, K. Li, M. Shen, L. Zheng, G. Zhang, X. Shi, Facile hydrothermal synthesis and surface functionalization of polyethyleneiminecoated iron oxide nanoparticles for biomedical applications, ACS Appl. Mater. Interf. 5 (2013) 1722–1731. [33] C.Y. Wen, J.Y. Sun, Quantitative determination of the carboxyl groups on individual nanoparticles by acid-base titrimetry, ChemistrySelect 2 (2017) 10885–10888. [34] B. Altiparmak, F.Y. Lambrecht, E. Bayrak, K. Durkan, Design and synthesis of 99m Tc-citro-folate for use as a tumor-targeted radiopharmaceutical, Int. J. Pharm. 400 (2010) 8–14. [35] S. Rana, N.G. Shetake, K.C. Barick, B.N. Pandey, H.G. Salunke, P.A. Hassan, Folic acid conjugated Fe3O4 magnetic nanoparticles for targeted delivery of doxorubicin, Dalton Trans. 45 (2016) 17401–17408. [36] M.M. Yallapu, S.F. Othman, E.T. Curtis, N.A. Bauer, N. Chauhan, D. Kumar, M. Jaggi, S.C. Chauhan, Curcumin-loaded magnetic nanoparticles for breast cancer therapeutics and imaging applications, Int. J. Nanomed. 7 (2012) 1761. [37] N.S. Rejinold, R.G. Thomas, M. Muthiah, H.J. Lee, Y.Y. Jeong, I.-K. Park, R. Jayakumar, Breast tumor targetable Fe3O4 embedded thermo-responsive nanoparticles for radiofrequency assisted drug delivery, J. Biomed. Nanotech. 12 (2016) 43–55. [38] N. Parker, M.J. Turk, E. Westrick, J.D. Lewis, P.S. Low, C.P. Leamon, Folate receptor expression in carcinomas and normal tissues determined by a quantitative radioligand binding assay, Anal. Biochem. 338 (2005) 284–293. [39] M.P. Daryasari, M.R. Akhgar, F. Mamashli, B. Bigdeli, M. Khoobi, Chitosan-folate coated mesoporous silica nanoparticles as a smart and pH-sensitive system for curcumin delivery, RSC Adv. 6 (2016) 105578–105588. [40] S.-J. Yang, F.-H. Lin, K.-C. Tsai, M.-F. Wei, H.-M. Tsai, J.-M. Wong, M.-J. Shieh, Folic acid-conjugated chitosan nanoparticles enhanced protoporphyrin IX accumulation in colorectal cancer cells, Bioconjugate Chem. 21 (2010) 679–689. [41] S. Vahedi, O. Tavakoli, M. Khoobi, A. Ansari, M. Ali Faramarzi, Application of novel magnetic b-cyclodextrin-anhydride polymer nano-adsorbent in cationic dye removal from aqueous solution, J. Taiwan Inst. Chem. E 80 (2017) 452–463. [42] M.I. Slavkova, D. Momekova, B. Kostova, G.T. Momekov, P. Petrov, Novel dextran/b-cyclodextrin and dextran macroporous cryogels for topical delivery of curcumin in the treatment of cutaneous T-cell lymphoma, Bulg. Chem. Commun. 49 (2017) 792–799. [43] S. Mallakpour, M. Reza Zamanlou, Synthesis of new optically active poly (amide-imide) s containing EPICLON and L-phenylalanine in the main chain by microwave irradiation and classical heating, J. Appl. Polym. Sci. 91 (2004) 3281–3291. [44] R. Pushpalatha, S. Selvamuthukumar, D. Kilimozhi, Cross-linked, cyclodextrinbased nanosponges for curcumin delivery-physicochemical characterization, drug release, stability and cytotoxicity, J. Drug Deliv. Sci. Tec. 45 (2018) 45–53. [45] L. Zhuang, W. Zhang, Y. Zhao, H. Shen, H. Lin, J. Liang, Preparation and characterization of Fe3O4 particles with novel nanosheets morphology and magnetochromatic property by a modified solvothermal method, Sci. Rep. 5 (2015) 9320. [46] M. Khoobi, T.M. Delshad, M. Vosooghi, M. Alipour, H. Hamadi, E. Alipour, M.P. Hamedani, Z. Safaei, A. Foroumadi, A. Shafiee, Polyethyleneimine-modified superparamagnetic Fe3O4 nanoparticles: an efficient, reusable and water tolerance nanocatalyst, J. Magn. Magn. Mater. 375 (2015) 217–226. [47] O. ur. Rahman, S.C. Mohapatra, S. Ahmad, Fe3O4 inverse spinal super paramagnetic nanoparticles, Mater. Chem. Phys. 132 (2012) 196–202. [48] P. Singh, X. Ren, T. Guo, L. Wu, S. Shakya, Y. He, C. Wang, A. Maharjan, V. Singh, J. Zhang, Biofunctionalization of b-cyclodextrin nanosponges using cholesterol, Carbohydr. Polym. 190 (2018) 23–30.

139

[49] S. Honary, F. Zahir, Effect of zeta potential on the properties of nano-drug delivery systems-a review (Part 1), Trop. J. Pharm. Res. 12 (2013) 255–264. [50] B. Sarmento, D. Mazzaglia, M.C. Bonferoni, A.P. Neto, M. do Céu Monteiro, V. Seabra, Effect of chitosan coating in overcoming the phagocytosis of insulin loaded solid lipid nanoparticles by mononuclear phagocyte system, Carbohydr. Polym. 84 (2011) 919–925. [51] L. Zhuang, Y. Zhao, H. Zhong, J. Liang, J. Zhou, H. Shen, Hydrophilic magnetochromatic nanoparticles with controllable sizes and super-high magnetization for visualization of magnetic field intensity, Sci. Rep. 5 (2015) 17063. [52] H.S. Chae, S.H. Piao, H.J. Choi, Fabrication of spherical Fe3O4 particles with a solvothermal method and their magnetorheological characteristics, J. Ind. Eng. Chem. 29 (2015) 129–133. [53] Y. Xiao, H. Hong, A. Javadi, J.W. Engle, W. Xu, Y. Yang, Y. Zhang, T.E. Barnhart, W. Cai, S. Gong, Multifunctional unimolecular micelles for cancer-targeted drug delivery and positron emission tomography imaging, Biomaterials 33 (2012) 3071–3082. [54] S. Some, A.-R. Gwon, E. Hwang, G.-H. Bahn, Y. Yoon, Y. Kim, S.-H. Kim, S. Bak, J. Yang, D.-G. Jo, Cancer therapy using ultrahigh hydrophobic drug-loaded graphene derivatives, Sci. Rep. 4 (2014) 6314. [55] V. Estrella, T. Chen, M. Lloyd, J. Wojtkowiak, H.H. Cornnell, A. Ibrahim-Hashim, K. Bailey, Y. Balagurunathan, J.M. Rothberg, B.F. Sloane, Acidity generated by the tumor microenvironment drives local invasion, Cancer Res. 73 (2013) 1524–1535. [56] T. Mortezazadeh, E. Gholibegloo, N.R. Alam, S. Dehghani, S. Haghgoo, H. Ghanaati, M. Khoobi, Gadolinium (III) oxide nanoparticles coated with folic acid-functionalized poly (b-cyclodextrin-co-pentetic acid) as a biocompatible targeted nano-contrast agent for cancer diagnostic: in vitro and in vivo studies, MAGMA 32 (2019) 487–500. [57] L. Jiang, Z.-M. Gao, L. Ye, A.-Y. Zhang, Z.-G. Feng, A pH-sensitive nano drug delivery system of doxorubicin-conjugated amphiphilic polyrotaxane-based block copolymers, Biomater. Sci. 1 (2013) 1282–1291. [58] J. Soleymani, M. Hasanzadeh, M.H. Somi, N. Shadjou, A. Jouyban, Probing the specific binding of folic acid to folate receptor using amino-functionalized mesoporous silica nanoparticles for differentiation of MCF 7 tumoral cells from MCF 10A, Biosens. Bioelectron. 115 (2018) 61–69. [59] C. Müller, R. Schibli, Folic acid conjugates for nuclear imaging of folate receptor–positive cancer, J. Nucl. Med. 52 (2011) 1–4. [60] D. Feng, Y. Song, W. Shi, X. Li, H. Ma, Distinguishing folate-receptor-positive cells from folate-receptor-negative cells using a fluorescence off–on nanoprobe, Anal. Chem. 85 (2013) 6530–6535. [61] J. Li, L. Zheng, H. Cai, W. Sun, M. Shen, G. Zhang, X. Shi, Polyethyleneiminemediated synthesis of folic acid-targeted iron oxide nanoparticles for in vivo tumor MR imaging, Biomaterials 34 (2013) 8382–8392. [62] D. Chen, Q. Tang, X. Li, X. Zhou, J. Zang, W.-Q. Xue, J.-Y. Xiang, C.-Q. Guo, Biocompatibility of magnetic Fe3O4 nanoparticles and their cytotoxic effect on MCF-7 cells, Int. J. Nanomed. 7 (2012) 4973–4982. [63] C.W. Jung, P. Jacobs, Physical and chemical properties of superparamagnetic iron oxide MR contrast agents: ferumoxides, ferumoxtran, ferumoxsil, Magn. Reson. Imaging 13 (1995) 661–674. [64] C.-Y. Chou, M. Abdesselem, C. Bouzigues, M. Chu, A. Guiga, T.-H. Huang, F. Ferrage, T. Gacoin, A. Alexandrou, D. Sakellariou, Ultra-wide range fielddependent measurements of the relaxivity of Gd1 x Eux VO4 nanoparticle contrast agents using a mechanical sample-shuttling relaxometer, Sci. Rep. 7 (2017) 44770. [65] V. Nejadshafiee, H. Naeimi, B. Goliaei, B. Bigdeli, A. Sadighi, S. Dehghani, A. Lotfabadi, M. Hosseini, M.S. Nezamtaheri, M. Amanlou, Magnetic bio-metal– organic framework nanocomposites decorated with folic acid conjugated chitosan as a promising biocompatible targeted theranostic system for cancer treatment, Mater. Sci. Eng. C. 99 (2019) 805–815. [66] C. Loo, A. Lowery, N. Halas, J. West, R. Drezek, Immunotargeted nanoshells for integrated cancer imaging and therapy, Nano Lett. 5 (2005) 709–711. [67] S. Sadjaji, M. Malmir, M.M. Heravi, M. Raja, Magnetic hybrid of cyclodextrin nanosponge and polyhedral oligomeric silsesquioxane: efficient catalytic support for immobilization of Pd nanoparticles, Int. J. Biol. Macromol. 128 (2019) 638–647.