Abstract
This paper describes the development of mitoxantrone-loaded PEGylated graphene oxide/magnetite nanoparticles (PEG-GO/Fe3O4-MTX), and investigation of its preliminary drug delivery performance. For this, the GO was synthesized through oxidizing graphite powder, and subsequently carboxylated using a substitution nucleophilic reaction. The carboxylated GO (GO-COOH) was then conjugated with amine end-caped PEG chains by Steglich esterification. Afterward, GO-PEG/Fe3O4 nanocomposite was synthesized through the anchoring of Fe3O4 nanoparticles onto the surface of GO-PEG during the sonication. The biocompatibility and MTX-loading capacity of the synthesized GO-PEG/Fe3O4 nanocomposite were evaluated. The pH dependent drug release behavior and cytotoxicity effect of the MTX-loaded GO-PEG/Fe3O4 nanocomposite were also studied. According to biocompatibility, pH dependent drug release behavior as well as superior physicochemical and biological characteristics of graphene and magnetite nanoparticles, it is expected that the GO-PEG/Fe3O4 nanocomposite may be applied as de novo drug delivery system (DDS) for cancer therapy using both chemo- and photothermal therapy approaches.
Introduction
Since the first use of nanoparticles (NPs) for medicinal applications, defined as nanomedicines, by Speiser and co-workers in the late 1960s [1,2], more and more research efforts have been devoted toward the design and development of efficient nanosystems for enhancing the targeted delivery of therapeutic molecules. This intense interest is arises from their targetability, multifunctionality, possibility to entrap poorly soluble drugs, overcome to multi-drug resistance as well as potential for innovative diagnostic and therapeutic methods (known as theranostic) for a number of diseases [3–5]. In addition, controlled release of drugs is the way for meeting a number of problems in drug delivery systems. In the past few years controlled release of drugs from NPs-based drug delivery systems have been extensively studied and some convincing data have been obtained [6,7]. Nevertheless, this field is still growing and many pivotal issues remain to be addressed.
In this context, superparamagnetic iron oxide nanoparticles (SPIONPs) [3,8,9], graphene and its derivatives (especially graphene oxide; GO) [10–12] have been extensively applied for development of de novo drug delivery systems (DDSs). Furthermore, the theranostic potentials of both mentioned nanomaterials are well documented [13,14]. The biomedical application of SPIONPs is supported by U.S Food and Drug Administration (FDA) approval, and due to its inherent physicochemical characteristics including, superparamagnetism, high field irreversibility, high saturation field, extra anisotropy contributions, and shifted loops after field cooling more research efforts devoted on the design and development of DDSs as well as theranostic nanomedicines using SPIONPs as magnetic resonance (MR) imaging agents [15–17].
On the other hand, graphene, a one-atom-thick hexagonal lattice of sp2 carbon atoms, has gotten much attention mainly due to its superior physicochemical properties (e.g., excellent mechanical and electronic transport characteristics) [18–20] as well as its importance in basic scientific research and potential applications including, (bio)chemical sensors [21], nanoelectronic devices [22], transparent conductors [23], solar cells [24], and biomedical sciences [25,26]. Among the graphene derivatives, the GO has a lot of functional groups including epoxide, hydroxyl and carboxylic groups on its surface which allow the further modification or attachment of bioactive molecules. In addition, GO has higher dispersability in water than graphene due to its strong hydrophilicity that arises from surface reactive oxygen functional groups [27,28]. In comparison with other nanoparticles-based drug delivery systems, the graphene and its derivatives have higher potential and advantages for this purpose according to the following reasons.
Production in large scales with low cost [29]
High drug loading capacity due to their ultra-high surface area [30]
Easy surface modification through both covalent and non-covalent approaches to improve the biocompatibility [31]
Applications in photothermal therapy (PTT) due to their high near-infrared (NIR) absorbance [32]
Furthermore, graphene and its derivatives have high potential for development of theranostic nanomedicines through the anchoring of inorganic nanoparticles (e.g., magnetic and gold) onto the surface of nanosheets [33,34], as well as labeling of these nanomaterials with radionuclides (e.g., 64Cu, 66Ga, 125I, and 131I) [35,36]. It is well established that the graphene and its derivatives with well-engineered surface coating and small sizes are non-toxic in a reasonable dose range and could be gradually removed from the body via urine and feces [37,38].
Since the antitumor activity approval of mitoxantrone (MTX) by FDA, this synthetic anticancer drug has been used for treating breast cancer, prostate cancer, acute myeloid leukemia and non-Hodgkin’s lymphoma. However, this anticancer drug has some side effects including cardiotoxicity, hepatotoxicity, mucositis, and hair loss [39,40]. Considering these facts, smart and targeted delivery of MTX to the required area is critical in cancer chemotherapy to minimize its side effects. For this aim, the used of stimuli responsive and targetable carriers is an immensely important for enhanced delivery of this drug.
In this investigation, GO-PEG/Fe3O4 nanomaterial was synthesized and its performance for delivery of MTX as an anticancer drug was evaluated. For this purpose, the GO was synthesized by oxidizing pristine graphite powder, and then carboxylated through a substitution nucleophilic reaction in the presence of sodium hydroxid (NaOH), and chloroacetic acid. Afterward, carboxylated GO (GO-COOH) was reacted with amine end-caped PEG in the presence of
Materials
Graphite, potassium permanganate (KMnO4), sodium nitrate (NaNO3), hydrochloric acid (HCl), sulfuric acid (H2SO4), N-(3-dimethylamino propyl-N-ethylcar-bodiimide) hydrochloride (EDC · HCl), N-hydroxysuccinimide (NHS), chloroacetic acid, oleylamine, benzyl ether, methoxypolyethylene glycol amine (mPEG-NH2;
Synthesis of Fe3O4 NPs
A 250 mL three-necked round-bottom glass reactor was charged with iron(III) acetylacetonate (1.06 g, 3 mmol), oleylamine (15 mL), and benzyl ether (15 mL). The reactor was stirred at room temperature for about 30 minutes, and then places in an oil bath at 120 ± 3°C under a flow of nitrogen gas for about 1 hour, in order to remove the humidity. Afterward, the temperature was rapidly increased to 300°C, and the reaction was continued for another 1 hour at this temperature. At the end of this time, the reaction mixture was cooled to room temperature, added ethanol (3 × 40 mL), centrifuged at 8000 rpm for 10 minutes, and the black NPs were dried under reduced pressure at room temperature.
Synthesis of GO from graphite
The GO was synthesized through the oxidation of graphite powder as described in literature [41,42]. For this, a 500 mL three-necked round-bottom glass reactor was charged with NaNO3 (3.0 g, 35 mmol), graphite powder (6.0 g), and sulfuric acid (95–98%; 360 mL). The reaction mixture was stirred for about 30 minutes at room temperature, and then KMnO4 (18.0 g, 114 mmol) was added slowly to the reaction mixture in an ice/water bath at 4°C. At the end of this time, the reaction mixture was warmed to 35°C, and stirred at this temperature for about 8 hours. Afterward, another portion of KMnO4 (18.0 g, 114 mmol) was added into the flask, and stirred for about 12 hours at mentioned temperature. The content of the reactor was diluted using deionized water (1000 mL), and then H2O2 (60 mL, 30 wt.%) was added into the diluted product in order to reduce the unreacted KMnO4. The crude product was centrifuged (8000 rpm, 10 minutes), and washed with hydrochloric acid solution (1 molL−1) several times. The synthesized graphite oxide was washed with distilled water until its pH reached about 7. Finally, the graphite oxide was exfoliated by sonication for about 1 hour. The suspension was centrifuged, and the GO powder was obtained after drying in vacuum at 60°C.
Synthesis of GO-PEG
The synthesized GO was further stabilized through the PEGylation. First, the GO was carboxylated using the following method. An aqueous suspension of GO (10 mL, ∼4 mgmL−1) was sonicated for about 1 hour to give a homogenous solution. Then, NaOH (1.0 g, 25 mmol) and chloroacetic acid (0.95 g, 10 mmol) were added to the GO suspension and stirred for about 24 hours at 45 ± 3°C. At the end of this time, the solution was neutralized, and purified by repeated rinsing and filtrations. The GO-COOH suspension was transferred into a 25 mL round-bottom flask, EDC (40 mg, 0.2 mmol) and NHS (24 mg, 0.2 mmol) were added to the flask, and the mixture was stirred for about 3 hours. Afterward, the mPEG-NH2 (200 mg, 0.10 mmol) was added to the flask, and stirred for 24 hours at 40 ± 3°C. The final product (GO-PEG) was filtered using ultra-filtration method through centrifugal devices (Amicon® Ultra-4 100 kDa, Millipore, Temecula, CA, USA) and washed with double distilled water for several times.
Synthesis of GO-PEG/Fe3O4 nanocomposite
The synthesized magnetite NPs (20 mg) and GO-PEG (30 mg) were dispersed in
Preparations of Fe3O4-MTX and GO-PEG/Fe3O4-MTX
A 50 mL three-neck round-bottom flask was charged with Fe3O4 NPs (50 mg), and dimethyl sulfoxide (DMSO, 10 mL) under argon atmosphere. The content of flask was stirred for about 20 minutes. In a separate 50 mL three-neck round-bottom flask, the drug (MTX; 10 mg, 0.023 mmol) was dissolved in DMSO (10 mL) and stirred for 10 minutes. At the end of this time, the content of this flask was added to the above mentioned flask, and the mixture was stirred overnight at room temperature. Afterward, the crude product was centrifuged at 8000 rpm (10 minutes). The product was further purified by external magnetic field and decantation process for several times using DMSO as the solvent, in order to remove un-loaded MTX molecules. The MTX-loaded GO-PEG/Fe3O4 was prepared by the same method.
Cells cultures
Mesenchymal stem cells (MSCs) were purchased from Iranian National Cell Bank (Pasteur Institute, Tehran, Iran). The MSCs were cultivated in RPMI1640 (Roswell Park Memorial Institute medium) possessing 100 IU penicillin per 100 μg streptomycin, enriched with 10% (v/v) of fetal bovine serum (FBS; Invitrogen, Carlsbad, CA, USA). The cells were cultured into flasks and kept in a humidified incubator with 5% CO2 at 37°C and the media refreshed twice weekly.
In vitro cytotoxicity assay
In this study, we utilized colorimetric test based on tetrazolium dye, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), bio-reduction for evaluation of the cytotoxic effects caused by the synthesized nanomaterials and their corresponding MTX-loaded nanosystems. Briefly, the MSCs cells was trypsinized, harvested and seeded in 96 well plates. After 24 hours post-seeding and at 40–50% confluency, the cultured cells were treated with different concentrations (1, 2, and 3 μgmL−1) of free MTX, Fe3O4 NPs, Fe3O4-MTX, GO-PEG/Fe3O4, and GO-PEG/Fe3O4-MTX for 48 hours. After 48 hours, the media containing drugs was removed and 50 μL of MTT solution was added to each well followed by 150 μL cultivation medium, then placed in a cell culture incubator for another 4 hours. Finally, the remaining MTT solution aspirated, the formed formazan crystals were dissolved in DMSO (200 μL) containing Sorenson buffer (25 μL), and the absorbance of converted dye was measured at 570 nm using a spectrophotometric plate reader, ELx 800 (Biotek, San Francisco, CA, USA) [43].
Characterization
Fourier transform infrared (FTIR) spectra of the samples were taken in a Shimadzu 8101M FTIR (Shimadzu, Kyoto, Japan) between the wavenumber range of 4000 to 400 cm−1. The samples were prepared by grinding the dry powders with potassium bromide (KBr), and compressing the mixture into disks. The spectra were recorded at room temperature. Ultraviolet-visible (UV-vis) spectroscopy was taken in a Shimadzu 1650 PC UV-vis spectrophotometer (Shimadzu, Kyoto, Japan). The thermal properties of the synthesized samples were investigated by means of thermogravimetric analyzer (TGA-PL STA 1640 equipment (Polymer Laboratories, Shropshire, UK)). The TGA experiments were conducted under a nitrogen atmosphere from 25 to 700°C with heating rate of 10°C min−1. The morphologies of the synthesized samples were studied using scanning electron microscope (SEM) type 1430 VP (LEO Electron Microscopy Ltd, Cambridge, UK).
Results and discussion
The increasing demands for enhanced DDSs have been stimulated considerable interest during last decade. In this context, the magnetic NPs and graphene-based nanosystems are particular of attention, in part due to superior physicochemical as well as biological characteristics of both nanomaterials. As shown in Scheme 1, the aim of this study is the design of magnetic GO-based drug delivery system for targeted cancer therapy using chemo-photothermal therapy. MTX can be loaded efficiently onto/into the GO-PEG/Fe3O4 nanocomposite through physical interactions (e.g., hydrogen bonding). Considering above mentioned facts, the association of SPIONPs and GO (through the anchoring of magnetic NPs onto the surface of GO) may be open new opportunities for design and development of DDSs in part due to superior physicochemical and biological features of both nanomaterials [44,45].
The overall strategy for the synthesis of GO-PEG/Fe3O4 nanocomposite.
The FTIR spectra of graphite, GO, and GO-COOH is shown in Fig. 1. The graphite shows a weak bands at 1485 and 3370 cm−1 related to the stretching vibration of aromatic C=C, and hydroxyl groups, respectively. After oxidizing of graphite into GO some new bands including the stretching vibration of carbonyl group at 1728 cm−1, the stretching vibrations of C–O groups of the epoxy and carboxylic acid groups at 1205 and 1197 cm−1, the C–O–C stretching vibration at 1045 cm−1, and the hydroxyl stretching vibration as a broad and strong band centered at 3365 cm−1. After carboxylation of GO using a nucleophilic substitution reaction the most significant changes in the FTIR spectrum of the GO-COOH are the appearance of aliphatic C–H stretching vibrations at 2960 and 2865 cm−1, increasing the intensity of carbonyl stretching vibration at 1730 cm−1, and significantly decreasing hydroxyl stretching vibration intensity at 3430 cm−1.
FTIR spectra of graphite, GO, and GO-COOH.
The FTIR spectra of the GO-PEG, Fe3O4 NPs, and GO-PEG/Fe3O4 nanocomposite is shown in Fig. 2. The most important bands in the FTIR spectrum of the GO-PEG can be listed as the stretching vibration of C–O at 1118 cm−1, the aliphatic C–H stretching vibrations at 2950–2800 cm−1 region, and the –CH2 bending vibration at 1505 cm−1. These characteristics bands confirm the successful incorporation of PEG chains into the GO. It is worth to note that the PEGylation of organic as well as inorganic nanomaterials has been widely used in biomedical sciences. Some advantages of PEGylation in drug delivery systems can be listed as decreased immunogenicity and frequency of administration, increased bioavailability, increased in vivo circulation half-life of the drug by preventing enzymatic degradation and renal clearance, and optimized pharmacokinetics [3].
The FTIR spectrum of the Fe3O4 NPs shows the stretching vibration of oxygen-metal at 567 cm−1, the surface hydroxyl groups at 3330 cm−1, as well as the characteristic bands related to oleylamine including the stretching vibration of C=C at 1645 cm−1, C–H bending vibrations at 1500–1350 cm−1 region, and the weak C–H stretching vibrations at 2950–2800 cm−1 region.
As seen in Fig. 2, the FTIR spectrum of the GO-PEG/Fe3O4 nanocomposite shows the combination bands related to both GO-PEG and Fe3O4 nanomaterials that confirm the anchoring of Fe3O4 NPs onto/into the GO-PEG sheets. The most important bands are labeled in the FTIR spectrum of the GO-PEG/Fe3O4 nanocomposite.
FTIR spectra of GO-PEG, Fe3O4 NPs, and GO-PEG/Fe3O4 nanocomposite.
The thermal properties of the graphite, GO, and GO-PEG were studied by means of thermogravimetric analysis (TGA). It should be pointed out that the mass loss of all samples around 100°C is related to the loss of moisture and adsorbed water molecules, especially in the case of GO, and GO-PEG samples due to their strong hydrophilic characteristics. As seen in Fig. 3, the pristine graphite is almost stable up to 700°C, and the residue at 700°C for pristine graphite is about 96 wt.%.
It is well documented that the reactive oxygen functional groups (e.g., epoxide, hydroxyl, and carboxyl groups) in the GO structure has resulted to strongly hydrophilic property, thus water molecules may be adsorbed in π-stacked structure of this compound. As seen in TGA curve of the GO, two distinct weight loss processes are observed; the first step (120 to 230°C) corresponds to the CO, CO2, and steam release from the most labile functional groups, whereas the second step (230 to 310°C) is associated with the degradation of more stable oxygen functionalities [42]. After which the loss rate slows down, and the residue at 700°C is 37 wt.%. The weight loss between 310 to 700°C is related to the pyrolysis of the carbon skeleton.
In contrast, the TGA curve of the GO-PEG sample shows the major weight loss around 330 to 480°C, which related to the degradation of PEG chains and pyrolysis of the carbon skeleton in GO. After which the loss rate slows down, and the residue at 700°C for this sample is 14 wt.%. In addition, the weight loss around 130 to 240°C, is attributed to the CO, CO2, and steam release from the most labile functional groups, whereas the weight loss around 240 to 330°C is associated with the degradation of more stable oxygen functionalities in the GO.
TGA curves of graphite, GO, and GO-PEG.
The surface morphologies of the graphite, GO, GO-PEG, and GO-PEG/Fe3O4 nanocomposite were studied using SEM as shown in Fig. 4. As seen in this figure, graphite (Fig. 4(a)) looks like thin “petal” flake with a typical lamella structure in which graphene layers are not distinguishable. However, in GO (Fig. 4(b)), the lamella structure is crumpled and wrinkled due to oxidation process, while homogeneous graphene nanosheets exist. These nanosheets are folded and it is possible to distinguish the edges of individual sheets, including kinked and wrinkled areas.
As seen in Fig. 4(c), the PEG chains grew from GO nanosheets, and covered the some surface areas of the nanosheets. Finally, the incorporation of Fe3O4 NPs resulted to relative disappearance of folded structure and re-stacking of nanosheets. However, the edges of individual sheets and presence of Fe3O4 NPs in nanocomposite are distinguishable (Fig. 4(d)).
SEM images of graphite (a), GO (b), GO-PEG (c), and GO-PEG/Fe3O4 nanocomposite (d).
As known MTX is a cationic drug and can be easily loaded into the developed negatively charged nanocomposite through the physical interactions. In detail, the functional groups of drug including amine, hydroxyl and carbonyl interacts with functional groups of developed GO-PEG/Fe3O4 nanocomposite through the formation of firm hydrogen bonds. The MTX-encapsulation efficiency (EE) and MTX-loading efficiency (LE) were calculated using the following equations through the analyzing of supernatant solutions for determination of unloaded drug concentration using UV-vis spectroscopy at 302 nm, and the results obtained are summarized in Table 1.
In these equations
The MTX-encapsulation efficiency (EE) and MTX-loading efficiency (LE) results of Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite
The MTX-encapsulation efficiency (EE) and MTX-loading efficiency (LE) results of Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite
In vitro release profiles of MTX from drug-loaded Fe3O4 NPs, and GO-PEG/Fe3O4 nanocomposite at various pH values (T = 37°C).
The in vitro pH-dependent drug release behaviors of the the MTX-loaded Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite were investigated using dialysis method. The release rates of MTX from drug-loaded nanomaterials were calculated using the following equation:
As seen in this figure, in both pH values (5.3 and 7.4) the MTX-loaded GO-PEG/Fe3O4 nanocomposite has higher drug release values than those of the MTX-loaded Fe3O4 NPs. This may be resulted from higher drug loading capacity of the GO-PEG/Fe3O4 nanocomposite (see Table 1).
Furthermore, as seen in Fig. 5, both MTX-loaded Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite at pH 5.3 have higher drug release values than those of the physiological condition (pH = 7.4 and T = 37°C). This phenomenon may be resulted from the repulsion forces between the protonated amine groups at both oleylamine and MTX at acidic pH (5.3).
At the end of this section, it should be pointed out that the external pH of cancerous tissue tends to be lower in comparison with the surrounding normal tissue, which is originated from abnormal metabolism of cancer tissues [46]. According to this fact and the results obtained from in vitro drug release study, the developed GO-PEG/Fe3O4 nanocomposite can be considered as a smart drug delivery nanosystem that can control precise release of MTX to the required area (cancerous tissue) in response to pH, and magnetic triggers.
In vitro cytotoxicity effects
It should be pointed out that the biocompatibility of any materials is the primary concern for biomedical application (e.g., drug delivery). Thus, the cytotoxicity effects of the Fe3O4 NPs, GO-PEG/Fe3O4, and MTX-loaded Fe3O4 NPs and GO-PEG/Fe3O4 against MSCs were evaluated using MTT assay in time period of 48 hours, and the results obtained were compared with the cytotoxicity effect of the free MTX as the reference.
As seen in Fig. 6, Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite were not able to induce significant cytotoxicity, especially at lower concentrations. In contrast, MTX-loaded Fe3O4 NPs and GO-PEG/Fe3O4 nanocomposite as well as free MTX exhibited cytotoxicity effects against MSCs. In comparison with free MTX the MTX-loaded GO-PEG/Fe3O4 nanocomposite showed higher cytotoxicity effects in all concentrations. In addition, the Fe3O4-MTX and free MTX showed approximately the same value of cytotoxicity. The higher cytotoxicity effect of the GO-PEG/Fe3O4-MTX may be originated from higher drug loading capacity and slow releasing of the loaded anticancer drug.
The cytotoxicity assay results of Fe3O4 NPs, GO-PEG/Fe3O4 nanocomposite, free MTX, and MTX-loaded Fe3O4 NPs and GO-PEG/Fe3O4 with different concentrations (1.0, 2.0, and 3.0 μg mL−1) in time period of 48 hours against Mesenchymal stem cells (MSCs).
Taking together, this study has shown a simple and powerful strategy for development of a drug delivery system using PEGylated GO and magnetite nanoparticles. The chemical structures of all samples as representatives were characterized by means of FTIR spectroscopy. The surface morphology of the developed GO-PEG/Fe3O4 nanocomposite was studied using SEM, and the growth of PEG chains from GO nanosheets as well as anchoring of Fe3O4 NPs onto the surface of GO-PEG during the sonication were confirmed. The developed GO-PEG/Fe3O4 nanocomposite showed higher drug loading capacity as well as higher cytotoxicity effect than those of the Fe3O4 NPs.
As the biocompatibility, pH responsive drug release behavior results, as well as superior physicochemical and biological characteristics of graphene and Fe3O4 NPs the developed GO-PEG/Fe3O4 nanocomposite may be applied as de novo drug delivery system (DDS) for cancer therapy using both chemo- and photothermal therapy approaches.
Footnotes
Acknowledgement
We express our gratitude to the Sahand University of Technology for supporting this project.
Conflict of interest
There is no conflict of interest between the Authors.