Doxorubicin-loaded iron oxide nanoparticles for glioblastoma therapy: a combinational approach for enhanced delivery of nanoparticles

The DOX loading efficiency on the EDT-IONPs was calculated to be 5 ± 0.05%. The DOX-EDT-IONPs had a ζ of 0.0 ± 0.02 mV compared to − 20.05 ± 2.7 mV for EDT-IONPs. The change in surface charge of the nanoparticles upon drug loading can be attributed to the electrostatic interactions between the amine groups of DOX and carboxylic acid groups of EDT coating. In addition, the D_H of the EDT-IONPs increased from 51.8 ± 1.3 nm (polydispersity index: PDI 0.14) to 75.5 ± 3.2 nm (PDI 0.27) upon DOX loading. The release profile of DOX from the nanoparticles is depicted in Fig. 1e. The nanoparticles demonstrated a burst release of 42 ± 5% within the initial 3 hours, while the remaining coated DOX gradually released within a 4-day period. Moreover, upon release of the loaded-DOX from nanoparticles within 4 days, the surface charge of the nanoparticles became negative again and returned to − 25.69±2.8 mV. Release studies performed at pH 4.5 also showed an accelerated initial release of DOX from the nanoparticles with up to 64 ± 4% within the initial hours. The enhanced release at pH 4.5 was due to the reduced electrostatic interactions between DOX and IONPs¹⁰. The increased release of DOX observed under these acidic conditions is similar to previous reports with DOX-IONP in the acidic tumor microenvironment or acidic cellular compartments such as endosomes^32,33.

The release of DOX from the IONPs observed in the present study was similar to previous reports with polymer-based nanoparticles. Poly-L-arginine/chitosan-coated iron oxide nanoparticles exhibited 40% and 65% release of DOX within 2 h at pHs 7 and 5, respectively³⁴. Although covalent bonding of DOX to the surface of the nanoparticles can result in increased loading and reduced initial burst release, these advantages are countered by potential reductions in the total release of the drug from the nanoparticles. For instance, when DOX was covalently conjugated to iron oxide nanoparticles via a pH-sensitive hydrazone linkage, there was a 29% burst release within 2 h. However, only 4% of the loaded DOX was further released within 24 h and the cumulative release was only around 35% under acidic pH conditions³⁵. The release rate observed in the present study is well-suited for the proposed delivery approach involving transient opening of the BBB. Previous in vivo studies using the cadherin peptides for transient opening of the BBB indicated a therapeutic delivery window of approximately 60 minutes following treatment³⁶. Thus, the DOX-EDT-IONPs would be expected to enter the brain within an hour, while carrying over 60% of the initial concentration of the loaded DOX. Moreover, the rapid release of DOX (within an hour) from the DOX-EDT-IONPs that magnetically has been drawn to the target site, can increase the chance of DOX entering the brain through the transiently open tight junctions of the BBB to provide a higher concentration of the drug within the brain.

In order to deliver an effective concentration of both DOX and EDT-IONPs to GBM cells in our studies, the concentration of 20 µg/mL of EDT-IONPs was selected for use in the remaining studies. This concentration, which was well tolerated in various cell lines, enables delivering enough DOX to observe cytotoxicity on the tumor cells. We previously reported biocompatibility of IONPs on endothelial, astrocyte and neuron cells at a concentration up to 100 µg/mL³⁸. Moreover, iron oxide nanoparticles clinically demonstrate acceptable biocompatibility and they are captured by the reticuloendothelial system (RES), by which the iron is incorporated into the body’s iron cycle⁵. In practice, iron oxide nanoparticles are coated with biocompatible and hydrophilic materials, diminishing the non-specific protein adsorption on the nanoparticle surface and decreasing their recognition and clearance by the RES, thereby their circulation time, as well as accumulation in the brain tumor can be augmented³⁹.

In terms of drug accumulation in U251 GBM cells, treatment with DOX-EDT-IONPs, was more effective than DOX alone. In the present study, the DOX-EDT-IONP resulted in approximately 2-fold greater uptake compared to an equal concentration of DOX in solution (Fig. 3d). In addition, application of an external magnetic field further enhanced the DOX accumulation in the U251 cells (2.8 ± 0.5-fold, Fig. 3c). In practice, the efficacy of chemotherapy with DOX is limited by the multiple drug resistance (MDR) mechanisms due to the overexpression of ATP-binding cassette (ABC) and P-gp efflux transporter in cancer cells. The expression of P-gp in U251 has been previously reported⁴³. In this regard, Wang et al.,⁴³ reported that co-administration of β-asarone and TMZ could decrease P-gp and MDR1 expression in U251, thus promoting TMZ’s entry into the GBM cells. Therefore, in this study, the DOX loaded on the nanoparticles could bypass the P-gp efflux system, leading to higher DOX’s uptake in the GBM cells^5,35. Similarly, higher uptake of DOX upon treatment of C6 glioma cells with DOX-loaded-polysorbate 80-SPIONs in comparison to that of free DOX was reported through endocytosis of the nanoparticles¹⁰.

The studies of cell morphology indicated that in addition to a significant reduction in the cell population, both DOX and DOX-EDT-IONPs treatments induced notable morphological changes from a cuboidal morphology of normal U251 to a shrunken and spindle-like structure of actin cytoskeleton and a disrupted nucleus (Fig. 8). The effect of DOX in induction of remodeling in actin cytoskeleton and disruption of central stress fibers leading to impaired cell adhesion and increased cell detachment has been reported previously⁴⁹. Moreover, phosphorylated H2AX (γ-H2AX), mediating DNA double-strand break, is an early and sensitive biomarker in DNA double‐strand break response⁵⁰. In Fig. 8, γ-H2AX can be visualized as foci by immunofluorescence in U251 treated with either DOX or DOX-EDT-IONPs. Such findings indicate DNA damage following DOX treatment in GBM cells. This is in accordance with previous findings of DOX-induced DNA damage and appearance of γ-H2AX in breast⁵⁰ and lung⁵¹ cancer cells.

Caspases are essential mediators of programmed cell death and they are triggered sequentially, in which activation of Caspase 12 leads to the activation of Caspase 9 and the subsequent ‘effector’ Caspase 3⁵⁹. Both DOX and DOX-EDT-IONPs treatments upregulated the Caspase 3 gene expression, which is consistent with its upregulation in C6 glioma¹⁰, leukemia HL-60⁶⁰, and MCF-7 breast cancer⁶¹ cells upon DOX treatments. p53 is a tumor suppressor protein whose mutation is the most prevalent genetic alteration in human cancers⁶². In fact, the p53 protein can inhibit DNA synthesis and regulates cell apoptosis through competition with the DNA repair mechanisms²¹. The U251 cells treated with DOX and DOX-EDT-IONPs exhibited an upregulated expression of p53.

Maternally Expressed Gene 3 (MEG3) is an imprinted non-coding RNA that acts as a tumor suppressor through both p53-dependent and p53-independent pathways⁶³. Furthermore, it has been found that MEG3 expression markedly is diminished in glioma tumors, whereas whose expression can inhibit cell proliferation and promoted cell apoptosis in U251 and U87 GBM cell lines⁶⁴. lncRNA-growth arrest-specific 5 (Gas5) is another tumor-suppressor gene that is downregulated in glioma cells⁶⁵. Suppressing the GBM tumor malignancy has been observed through introduction of Gas 5 and consequently downregulation of miR-222⁶⁶. Here, both DOX and DOX-EDT-IONPs treatments were found to be effective in upregulation of both tumor suppressors, i.e. MEG3 and Gas5, which potentially leads to GBM cell apoptosis.

MiR-155 is an important oncogenic microRNA that is overexpressed in various malignant tumors including GBM, whose mechanism of action is associated with a blockade of Caspase-3 activity and regulation of multiple genes involved in cancer cell proliferation, and invasiveness⁶⁷. The expression of MiR-155 in U251 was downregulated upon treatment with either DOX (0.457 ± 0.24 fold) or more significantly with DOX-EDT-IONPs (0.28 ± 0.03-fold, p < 0.05). It also has been reported that downregulation of MiR-155 can enhance the chemosensitivity of U251 cells to Taxol by interrupting the activity of EAG1 pathways and inducing apoptosis⁶⁷.

In addition, the Wnt signaling pathway plays an important role in malignant transformation and tumor progression in gliomas⁶⁸, and the capacity of intracranial tumor formation has been found to be reduced upon Wnt silencing, in vivo⁶⁹. Here, U251 demonstrated a significant downregulation of Wnt1 upon the treatments with either DOX (0.21 ± 0.04 fold) or DOX-EDT-IONPs (0.17 ± 0.03 fold).

Due to the inability of DOX to cross the BBB and penetrate into the tumor site, it demonstrates little effectiveness in treating GBM when administered systemically⁷⁰. Having considered that, development of an efficient drug delivery system enabling penetration of DOX across the BBB and enhancing its bioavailability is a matter of significant importance in GBM chemotherapy. In selecting the most appropriate cell culture model of the BBB to evaluate the DOX-EDT-IONP delivery approach, both the bEnd.3 brain endothelial cell line and the MDCK-MDR1 cell line were considered. Our previous studies with bEnd.3 indicated this particular model was well suited for examining nanoparticle permeability⁷¹. However, the available brain endothelial cell culture models do not form a restrictive paracellular barrier required for screening the passage of small molecules⁷² and because of this often overestimate the BBB penetration. The MDCK-MDR1 cells overexpress P-gp and have reduced paracellular diffusion of solutes due to the complex tight junction proteins. For this reason, the MDCK-MDR1 cell line is often used to assess the BBB permeability and P-gp liabilities of drugs for central nervous system indications⁷³. Since the penetration of DOX across the BBB is mainly restricted by the P-gp expression under normal physiological conditions⁷⁴, the MDCK-MDR1 cells overexpressing P-gp were used in the present study to provide a more representative barrier cell for the BBB-GBM co-culture model.

Transient disruption of the BBB with hyperosmotic solutions like mannitol has been reported to enhance the delivery of therapeutic molecules as well as IONPs into the brain^75,76. In this regard, mannitol has extensively been used in combination with anti-tumor agents in clinical trials of glioma therapy over the last three decades⁷⁷. Similarly, Sun et al.,⁷⁸ reported a significant increase in permeability of both EDT-IONPs and aminosilane-coated (AmS)-IONPs across brain endothelial cell monolayers when tight junctions were disrupted using mannitol. However, the extensive opening of the BBB by mannitol and the long recovery time for re-establishment of the BBB integrity can cause a substantial and uncontrolled influx of low and high molecular weight substances from the blood into the brain that can result in neurological toxicity, jeopardizing patient safety⁷⁹. Previous studies using cadherin binding peptides to modulate BBB permeability suggested a more controlled opening of the BBB in terms of both magnitude and duration of opening was possible⁸⁰. In the present study, the cadherin binding peptide, ADTC5 was used to transiently modulate permeability in the BBB-GBM co-culture model. While ADTC5 was able to increase the permeability of the DOX-ECT-IONPs, especially when combined with an external magnetic field (Fig. 11), wholesale disruption of monolayer integrity was much less than observed with hyperosmotic mannitol, as demonstrated by the permeability to the 35 kDa molecular weight IRDye [2.7 ± 0.4% and 3.1 ± 0.3% without and with ADTC5, respectively, compared to 15.6 ± 0.6% with mannitol (Fig. 5S)]. Moreover, as mentioned previously, the recovery time for re-establishment of the BBB integrity was reported to be within 60 min post-injection of the cadherin binding peptide in vivo³⁶. This means that using the cadherin peptide allowed the MDCK-MDR1 monolayers to maintain barrier properties to large IRDye macromolecule marker, while allowing enhanced penetration to the IONPs, especially in the presence of an external magnetic field. We have named this approach magnetic enhanced convective diffusion (MECD) as the IONPs diffuse across the transiently disrupted cell barrier in a bulk flow manner that is accelerated by the presence of an external magnetic field.

By transiently opening the MDCK-MDR1 monolayer tight junctions using ADTC5 and in combination with an external magnetic field, the GBM cell viability significantly decreased upon treatment with DOX-EDT-IONPs compared to the GBM cells treated with free DOX (cell viability 66 ± 3.3% and 45 ± 3.7% for GBM cells treated with free DOX and DOX-EDT-IONPs, respectively) (Fig. 11b). This result was consistent with the higher DOX-EDT-IONP permeability through the MDCK-MDR1 monolayer when both ADTC5 and external magnetic fields were applied.

The permeability of DOX-EDT-IONPs through BBB-GBM co-culture model was also examined. Under normal conditions, DOX-EDT-IONPs showed 5.2 ± 0.4% penetration across the MDCK-MDR1 monolayer over 4 hours. The diffusion of DOX-EDT-IONPs could be increased by either enhancing the MDCK-MDR1 monolayer permeability with ADTC5 (6.2 ± 0.45%) or by application of an external magnetic field (7.4 ± 0.5%) (Fig. 11c). However, combining both ADTC5 treatment with an external magnetic field significantly augment DOX-EDT-IONP penetration by 8.5 ± 0.36%. To the best of our knowledge, this is the first report on the combinational effect of cadherin binding peptide and external magnetic field as an effective approach to enhance the permeability of drug delivery systems across the BBB.

As mentioned earlier, IONPs uniquely provide a site-specific magnetic targeting utilizing an external magnetic field to draw the nanoparticles to the site of action and enhancing their bioavailability⁸¹. For instance, by applying an external magnetic field, overall tumor exposure to magnetic nanoparticles was enhanced by 5-fold compared to non-targeted tumors⁸². Moreover, ADTC5 has shown an enhanced delivery of various marker molecules (e.g., ¹⁴C -mannitol, Gd-DTPA) across the MDCK monolayer in vitro, and the BBB in vivo through binding to the EC1 domain of E-cadherin, blocking the cadherin–cadherin interactions and thus enhancing the delivery of molecules into the brain via the paracellular pathway of the BBB⁸⁰.

Therefore, the developed DOX-EDT-IONPs in combination with the magnetic enhanced convective diffusion and the cadherin binding peptide for transiently opening the BBB tight junctions were found effective to enhance DOX’s bioavailability and anti-cancer effect in GBM cells by virtue of overcoming the MDR and enhancing the permeability of DOX through a BBB model in vitro.

This combinational approach can potentially be an efficacious alternative for the passive targeting through the enhanced permeability and retention (EPR) effect, or ligand-based active targeting of the IONPs in clinical practice. In fact, the EPR effect in humans has been found not as prominent as in animal models⁸³. Moreover, during early stages of brain tumor development, the EPR effect cannot play an important role inasmuch as the BBB is still intact, and leakiness is observed at the stages when tumor volume is high and difficult to treat⁸⁴. In addition, the infiltrating tumor cells are mostly associated with the intact BBB that would impede passive targeting of nanoparticles⁸⁵. On the other hand, the clinical outcomes of active targeting of nanoparticles to brain tumors have not yielded the results anticipated due to altered expression of target receptors in some types of tumors, tumor heterogeneity and the interpatient variability⁸⁶. BIND-014 and MM-302 are two examples of active targeting nanomedicines that failed in clinical studies⁸⁷. In light of these clinical studies, this novel combinational approach of using cadherin binding peptide for transiently opening the BBB tight junctions in juxtaposition with magnetic enhanced convective diffusion can be an alternative and effective approach for the passive targeting and ligand-based active targeting of drug-loaded IONPs in clinical practice. This combinational approach can provide a site-specific magnetic targeting to reduce systemic distribution of the drug-loaded IONPs, a transiently opening of the BBB tight junctions using a cadherin binding peptide, and an enhanced convective diffusion of the magnetic nanoparticles into the brain. These together can reduce the systemic toxicity of chemotherapy, enhance the permeability of the drug-loaded nanoparticles into the brain and ameliorate the efficacy of GBM chemotherapy by providing a therapeutic concentration of the effective anti-cancer drugs like DOX that are intrinsically impermeable to the BBB.

The chemical reagents were acquired from Sigma Aldrich (St. Louis, MO), and the cell culture and biochemical reagents were purchased from Thermo Fisher Scientific Inc, USA, unless otherwise specified.

Iron oxide nanoparticles were fabricated under mild conditions at room temperature as previously described³¹. Briefly, Fe(acac)₃ (2.83 g, 8 mmol) was dissolved in ethanol/DI water (6:4) and purged with nitrogen for 1 h, followed by adding NaBH₄ (3.03 g, 80.0 mmol) in deoxygenated DI water under stirring (1000 rpm). When the color of the reaction mixture changed from red to black, it indicates the formation of IONPs (approximately 20 min). For coating, (Trimethoxysilylpropyl)-ethylenediamine triacetic acid (EDT, 16 ml) was added, and the reaction mixture was stirred overnight at room temperature. The blackish brown solution was filtered, and the solvent was evaporated at 50 °C under low pressure. The obtained viscous mixture was dissolved in 200 ml of cold ethanol and left until excess NaBH₄ became crystallized, which was removed by filtration. Finally, ethanol was completely removed, the product was dissolved in 50 ml DI water and dialyzed against DI water to remove the unreacted EDT, followed by centrifugation at 4000 rpm for 30 min³¹. The dark reddish-brown supernatant was collected and stored for further use.

The size distribution of EDT-IONP in DI water was measured by dynamic light scattering (DLS) using a Photocor Complex system. The FTIR spectrum was taken using a Thermo Nicolet iS10 FTIR spectrometer. Transmission electron microscopy (TEM) images of the EDT-IONPs were obtained using a Philips CM 10 electron microscope (FEI, Hillsboro, USA).

To load DOX on the EDT-IONPs, EDT-IONPs (20 µg) and DOX (20 µg) in 200 µL phosphate-buffered saline (PBS, pH 6) was combined and incubated overnight under ambient conditions. Afterwards, the mixture was centrifuged at 12,000 rpm for 10 min and the solution was completely withdrawn. Then, the nanoparticles were washed with PBS (pH 7.4) twice to remove free DOX and the nanoparticles were centrifuged again to collect the DOX-loaded EDT-IONPs (DOX-EDT-IONPs).

To assess the biocompatibility of the synthesized EDT-IONPs, a mouse brain-derived microvessel endothelial cell line, bEnd.3 (American type tissue culture collection, Manassas, VA) was employed as a cell culture model for the BBB. The Madin–Darby canine kidney (MDCK) transfected with multi-drug resistant protein 1 (MDR) was also used. MDCK is an epithelial cell line originally derived from the normal dog kidney and transfected with MDR, expressing P-gp and tight junction proteins. Therefore, the MDCK-MDR1 has been reported as a model for the BBB permeability⁷³. Furthermore, an authenticated human U251 GBM cell line was used for biocompatibility evaluation of EDT-IONPs. The bEnd.3, MDCK-MDR1 and U251 cells (cells at passage number 20–30) were cultured at a density of 2 × 10⁴ (bEnd.3, MDCK-MDR) and 1 × 10⁴ (U251) cell/cm² in 96-well plates, and incubated overnight at 37 °C allowing the cells to attach. Next day, the cells were treated with EDT-IONPs (0.25 to 50 µg/mL) suspended in the cell culture medium for 48 h. Thereafter, the culture medium was removed, and the cells were washed with PBS followed by incubation with fresh medium containing 0.5 mg/mL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT, 0.5 mg/mL) reagent at 37 °C. After 3 h, the medium was withdrawn, and blue crystals were dissolved in pure DMSO^88–90. The absorbance of the solutions was measured using a Synergy HT plate reader (BioTek, Winooski, VT) at the wavelength of 570 nm and the relative cell viability was calculated as [OD]_test/[OD]_control, upon five measurements.

The release of DOX from the EDT-IONPs was measured in PBS (pH 7.4 mimicking physiological pH, and 4.5 mimicking pH of acidic intracellular compartments such as endosomes) at 37 °C. For this purpose, the DOX-EDT-IONPs were suspended in 1 mL PBS in Eppendorf tubes and at various time points, the tubes were centrifuged at 12,000 rpm for 10 min to pellet the nanoparticles and the solution was completely collected followed by re-suspension of the nanoparticles in 1 mL of fresh PBS. The concentration of the released DOX in the solution was determined by fluorescence measurement (excitation and emission wavelengths of 485 nm and 590 nm, respectively) using a Synergy HT plate reader. The concentration of the released DOX from DOX-EDT-IONPs was calculated using a serial dilution of a DOX standard solution.

To study the cellular uptake of DOX-EDT-IONPs; bEnd.3, MDCK-MDR, and U251 cells were grown in 24-well culture plates to reach a confluent monolayer and then they were treated with cell culture medium containing either EDT-IONPs or DOX-EDT-IONPs (10 and 20 µg/mL) for 4 h at 37 °C with and without a static magnetic field (rare-earth circular magnets, diameter: 20 mm, Lee Valley, Winnipeg, CA). Then the cells were washed with cold PBS to remove non-adhered nanoparticles, and lysed with 0.1% triton solution in PBS overnight at − 20 °C. The content of IONPs was determined based upon the Ferrozine assay as previously reported³⁸. Briefly, HCl (500 µL of 12 M) was added to wells, and were incubated at room temperature for 1 h with gentle shaking to digest the IONPs, followed by neutralization with NaOH (500 µL of 12 M). Then, hydroxylamine hydrochloride (120 µL of 2.8 M) in 4 M HCl was added, and the samples were incubated for 1 h at room temperature with gentle shaking. Afterwards, ammonium acetate solution (50 µL of 10 M, pH 9.5) and ferrozine (300 µL of 10 mM) in 0.1 M ammonium acetate solution were added sequentially to each well, and the absorbance of the solutions was determined at 562 nm by a Synergy HT plate reader. The concentration of EDT-IONPs was quantified based upon an iron chloride standard solution. The protein content of the lysed cells was also measured using a BCA protein assay kit.

The localization of EDT-IONPs in the cell organelles was also studied using TEM as previously described^29,42. For this purpose, U251 cells were treated with either EDT-IONPs or DOX-EDT-IONPs in accordance with the uptake study, and after washing with PBS, the cells were disassociated using a 0.25% trypsin EDTA solution (Hyclone, Logan, UT). After centrifugation of the collected cells (5 mins at 1500 g), the cell pellet was resuspended in 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.3) for 3 hours at room temperature. Then the samples were fixed for 2 h at room temperature in 1% osmium tetroxide in 0.1 M phosphate buffer, dehydrated in ascending concentrations of ethanol and embedded in Epon resin. Thin sections were stained with uranyl acetate and lead citrate, and photographed by TEM.

To measure the cellular uptake of DOX, U251 cells were grown in 6-well plates as described earlier and treated with cell culture media supplemented with an equal drug concentration of either DOX or DOX-EDT-IONPs to initiate the cellular drug accumulation. After 2 h, the cells were washed with cold PBS three times and lysed with 0.1% triton solution in PBS as described somewhere else⁹¹. The concentration of DOX in the cell lysates was measured as delineated in section “Drug release from EDT-IONPs” and normalized with the protein content of the lysed cells.

The cytotoxicity of DOX-EDT-IONPs against U251 cells was studied using MTT and flow cytometry analyses. For MTT assay, the cells were cultured as described in section “”. Next day, the medium was changed with fresh medium (negative control), medium containing free DOX with equivalent concentrations corresponding to DOX released from EDT-IONPs at the same period of time (positive control), EDT-IONPs and DOX-EDT-IONPs. After a 48-h treatment, viability of the cells was determined by MTT assay as described in section “”. Cytotoxicity of DOX-EDT-IONPs on cancer cell Biocompatibility assessment of EDT-IONPs

Moreover, cell apoptosis/necrosis was investigated using Annexin V-FITC/PI apoptosis Kit. For this study, the cells were treated with either EDT-IONPs, free DOX or DOX-EDT-IONPs over a 48-h period, followed by incubation in fresh cell culture media without any treatment for 24 h. Afterwards, the cells were stained with Annexin V-FITC and PI in accordance with the manufacturer’s protocol, and consequently were analyzed using flow cytometry (BD FACSCanto II Flow Cytometer instrument (BD Bioscience)). In addition, to study the effect of various treatments on cell proliferation, U251 cells were stained with a fluorescent carboxyfluorescein succinimidyl ester dye (CFSE, 50 mM), for 20 min at 37 °C. Thereafter, the medium was removed, and the cells were washed and treated with either free DOX, EDT-IONPs, or DOX-EDT-IONPs for 48 h followed by changing the media and leaving the cells without further treatment for 24 h. Then, the fluorescence intensity of the cells was determined using flow cytometry. In fact, during each cell division, the cellular content of CFSE decreases that results in a sequential halving of the cellular fluorescent intensity with each mitotic event⁹².

To observe any changes in morphology, the U251 cells were treated for 48 h as mentioned above, followed by washing with PBS, fixating with paraformaldehyde (4% v/v) for 20 min at room temperature and permeabilization with Triton X-100 (0.2% v/v) for 10 min. The specimens were then blocked with BSA solution (3% w/v) for 1 h at room temperature, washed with PBS, and the cells incubated with primary phosho-H2AX antibody solution (1:500 in 3% BSA, 0.3% Triton X-100 in PBS) at 4°C overnight. Afterwards, the primary antibody was withdrawn, and a goat anti-rabbit secondary antibody labeled with Alexa 488 dye (1:500 in the same buffer as the primary antibody) was added to each well and incubated at room temperature for 1 h. Then, the cells were washed with PBS and the actin cytoskeleton was stained with ActinRed for 30 min followed by the nucleus staining with DAPI solution (100 nM) for 5 min at 37 °C. Finally, the samples were washed with PBS and visualized by a fluorescence microscope (Zeiss Axio observer Z1, Germany).

The extent to which the various treatments resulted in ROS generation in the U251 cells was evaluated via the peroxide-dependent oxidation of the non-fluorescent 2′,7′-dichlorofuorescein diacetate (DCFDA). In this cell-based assay, DCFDA freely diffuses into the cells. Once inside, the DCFDA is transformed to the highly fluorescent and cell impermeable 2′,7′-dichlorofluorescein (DCF) through ROS mediated metabolism⁹³. For this study, the cells were cultured in black 96 well plates at a density of 5000 cell/cm². Next day, the cells were washed with PBS and exposed to 50 μM DCFHDA in PBS for 45 min at 37 °C. Afterwards, the DCFHDA solution was removed, and the cells were washed and treated with either EDT-IONPs, DOX or DOX-EDT-IONPs in cell culture media over 72 h. At various time points, cellular accumulation of ROS in response to the treatments was calculated by measuring the oxidation of DCFDA to the fluorescent DCF using a Synergy HT fluorescent plate reader at Ex/Em 485/590 nm.

Nanoparticles as a drug carrier for brain tumor therapy need to first overcome the limited permeability of the BBB as well as the efflux transporters such as P-gp expressed on the brain endothelial cells, which are responsible for low drug permeation into the brain. The Madin–Darby canine kidney epithelial cell line stably transfected with human multi-drug resistant protein 1 (MDCK-MDR) cells overexpress P-gp, and have reduced paracellular diffusion due to the complex tight junction proteins. Together these properties make MDCK-MDR1 cells a reproducible and accurate in vitro cell culture model for examining and predicting the penetration of drugs and solutes across the BBB⁷³. In the present study, MDCK-MDR1 cells (passage number 20–30, cell density 100,000 cell/cm² were plated on the apical side of a porous polycarbonate membrane inserts (4.6 cm² pore size: 3.0 μm, Corning Inc., USA). Once a confluent MDCK-MDR1 monolayer was obtained (typically in 6 days), U251 cells were cultured in the basolateral side of the well plates. Free DOX (1 µg/mL) or DOX-EDT-IONPs was added to the apical media compartment of the insert along with an IRdye 800CW PEG as a permeability marker. In addition, a cyclic ADTC5 peptide (Cyclo(1,7)Ac-CDTPPVC-NH₂), which was synthesized as previously reported⁹⁴ was added to the apical media compartment of the insert to block the cadherin–cadherin interactions and thus enhancing drug delivery through the MDCK-MDR1 monolayer. The cells were then incubated at 37 °C for 4 h in both the presence and absence of a static magnetic field (rare-earth circular magnet, diameter: 30 mm, Lee Valley, Winnipeg, CA). Afterwards, the apical media and the inserts were removed and the GBM cells with the basolateral cell culture media were incubated for an additional 48-h after which the basolateral media was collected to determine IONP (Ferrozine assay) and IR dye permeability as well as the cell viability (MTT assay).

The studies were conducted in triplicate and the results were reported as the mean ± standard deviation (SD). Statistical analysis was conducted using analysis of variance (ANOVA) and p < 0.05 was considered as the criterion of significance, as previously reported.

Supplementary information