Fe₃O₄@DMSA@GOx/DOX@RCM@FA: A ferroptosis-mediated multifunctional biomimetic nanosystem for precision theranostics in ovarian cancer

2.1. Materials

Iron acetylacetone, dibenzyl ether, oleylamine, oleic acid, DOX hydrochloride, and dimethylsulfoxide (DMSO) were purchased from Sigma (Merck, USA). Dimercaptosuccinic acid (DMSA), fatty acid methyl ester sulfonate (MES), 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide (EDC), 1,2-Distearoyl-sn-glycero-3-phosphoethanolamine-polyethylene glycol (DSPE-PEG), N-Hydroxysuccinimide (NHS), C11 BODIPY 581/591 probe, and JC-1 were purchased from MCE (USA). GOX was obtained from Yeyuan Biological Technology Co., Ltd. (Shanghai, China). 4’, 6-Diamino-2-phenylindole dihydrochloride (DAPI) and Cell-Counting-Kit-8(CCK8) were purchased from BeyotimeBiotechnology. ROS assay kit and Phospho-Histone H2A. X(γ-H2AX) were bought from Cell Biolabs, Inc. Indocyanine green (ICG) was purchased from Xi’an kaixin Biological Technology Co., Ltd. (Shanxi, China). Calcein-AM (CA) and propidium iodide (PI) were provided by Shanghai so lab Biotechnology Co., Ltd. Dulbecco’s modified Eagle’s Medium (DMEM), fetal bovine serum (FBS), and 0.25% ethylene diamine tetraacetic acid (EDTA) solution were purchased from Invitrogen-Gibco (Waltham, MA, USA).

2.2. Cell culture

Human ovarian carcinoma (SKOV3 and ID8) cells were obtained from the Key Laboratory of Obstetrics and Gynecology, Qilu Hospital, Shandong University. These cells are cultured in Rosewell park memorial institute (RPMI) 1640 medium with FBS (10%) at 37°C in a 5% CO₂ atmosphere.

2.3. Animals

Female Sprague-Dawley, C57, and nude-mice were purchased from JicuiYaokang Biotechnology Co., Ltd. (Nanjing, China). All the animal studies strictly followed the animal protocols approved by the Animal Care and Use Committee of Qilu Hospital (DWLL-2019-005).

2.4. Preparation of Fe₃O₄@DMSA@GOX/DOX@RCM@FA

2.6. Magneto-thermal experiment

To estimate the magneto-thermal properties of the Fe₃O₄@DMSA nano-system, an infrared thermal imager (UTi85A) was used to record the temperature changes in different concentrations of Fe₃O₄@DMSA nanoparticles under an AMF. Specifically, Fe₃O₄@DMSA nanoparticles were dissolved in 1 mL ultrapure water in a glass cuvette at different concentrations (100, 200, 500, and 1000 μg/mL) and irradiated under an AMF for 10 min (1.0 W/cm²). Ultrapure water (1 mL) without nanoparticles was used as a control. The temperature in each cuvette was measured every 30 s using an IR thermal imager. Additionally, a subcutaneous tumor-bearing mouse model was constructed, and after IV injection of Fe₃O₄@DMSA, the in vivo magnetic thermal conversion effect of Fe₃O₄@DMSA was tested by the above method.

2.7. T2-weighted magnetic resonance imaging

The Fe₃O₄ relaxation rate of T2-MRI was determined at different concentrations of Fe₃O₄@DMSA, Fe₃O₄, and glucosamine solution. 8-channel head coil, TR=3,000 ms, and echo times (TE) of 10-100 ms. were used to measure the corresponding T2 values, and the slope of the curve is the relaxation rate (r2). The subcutaneous tumor-bearing mouse model was established. The MRI signal was obtained using the above method.

2.8. Trajectory detection

SKOV3 cells (1×10⁴ cells/well) were seeded in Petri dishes. Fe₃O₄-DMSA (10 µg/mL) solution was added to the same. The movement of the nanoparticles was tracked under the microscope.

2.9. Release study

The release rate of DOX from nanoparticles was studied using the dialysis method. Briefly, DOX-NPs were dissolved in PBS (1 mL) (the concentration of DOX was 1 mg/mL) (with or without AMF), transferred to dialysis bags (3500 kDa), and immersed in 50 mL of DMEM. It was subjected to a shaking water bath (100 rpm) for 160 h. Then, 5 mL of each saline sample was collected for predetermined time intervals. Then, the DOX concentrations were analyzed by ultraviolet spectrophotometry.

2.10. Pharmacokinetic data analysis in rats

DOX (5 mg/kg) or DOX-NPs were injected into the tail vein of female Sprague-Dawley rats (n = 6/group). Blood samples were collected (0.17-60 h). MeOH (200 μL) and ethyl ether (2 mL) were added into blood samples. The mixture was vortexed (5 min) and centrifuged at 10,000 rpm (10 min). Finally, the DOX concentrations were analyzed using high-performance liquid chromatography (HPLC).

2.11. Cellular uptake of the nanoparticles

Solutions with the same DOX concentration were incubated with SKOV3, fixed with 4% formaldehyde for 30 min, and washed with PBS. PBS (5% Triton-X-100) was allowed to permeate for 15 min, and then PBS (10% NGS) was blocked for 1 h. The specific primary antibody was incubated at 37°C for 30 min and rinsed with PBS (0.1% Tween-20) thrice. The second antibody was incubated for 30 min and washed with PBS three times. Nuclei were stained with DAPI for 15 min, and then rinsed with PBS (0.1% Tween-20) thrice for 10 min. Fluorescence microscopy (Olympus, Tokyo, Japan) was used to analyze the cell uptake characteristics of different groups.

2.12. Tissue-distribution study in vivo

SKOV3 cells were used to construct a subcutaneous tumor model of female Bagg albino nude mouse (BALB/c) nude mice, and DOX standard curves of blood and major tissue samples were constructed. Subcutaneous tumor-bearing mice were randomly divided into four groups, which received an IV injection of DOX or DOX nanosolution (DOX concentration was 5 mg/kg). The mice were euthanized at different times, and plasma and tissue samples were quickly collected for HPLC analysis to determine the distribution characteristics of DOX tissues. ICG nanomaterials were injected into the tail vein of tumour-bearing mice, and the distribution of the nanomaterials in mice was dynamically monitored by mouse imaging to further explore the characteristics of tissue distribution.

2.13. Cell culture and cytotoxicity measurements

CCK8 was used to determine the effects of different drug-carrying nanocarriers with or without AMF, different drug concentrations (DOX concentration 0.5-4 μg/mL,) and different action times on the survival rate of tumor cells. Specifically, SKOV3 cells were inoculated and different concentrations of nanomaterials were added for different incubation times, 10 μL CCK8 was added to each pore plate and incubated for 2 h. Enzyme-labeled apparatus (PerkinElmer, Boston,MA, USA) detects absorbance at 450nm wavelength.

2.14. In vitro functional tests

2.15. Measurement of intracellular ROS concentration using 2′,7′-Dichlorofluorescin diacetate (DCFH-DA) probe

SKOV3 cells (5×10⁵ cells/well) were seeded into 6-well plates and incubated for 24 h, and then Fe₃O₄-NPs with different concentrations were added with or without the AMF conditions for 24 h. The cells were collected and incubated with a DCFH-DA probe (10 µM) for 45 min. The production of ROS was detected by fluorescence microscope, and the fluorescence intensity was quantified by Image J.

2.16. Measurement of Lipid peroxide using C11 BODIPY 581/591 probe

SKOV3 cells (5 ×10⁵) were instituted in 6-well plates and incubated with different groups of nanomaterials for 24 h. C11 BODIPY 581/591 was added to make the final concentration 50 μmol/mL and incubated for 1 h. The cells were washed with PBS more than twice and then stimulated by 488 nm and 565 nm lasers with a fluorescence microscope. The relevant fluorescence signal images were collected.

2.17. Measurement of the point changes of intracellular mitochondria by JC-1 probe

SKOV3 cell tablets (5×10⁴/well) were prepared and divided into five groups (control group, DOX, Fe₃O₄/DOX@RBM@FA, Fe₃O₄@GOX/DOX@RBM@FA, and Fe₃O₄@GOX/DOX@RBM@FA(AMF)). When the cells grew to 70-80%, PBS and drugs were added, and the cells were incubated for 12 h. Fe₃O₄@GOX/DOX@RBM@FA (AMF) was subjected to a magnetic field (390 kHz, 2.58 kA m⁻¹) for 15 min. The medium of each group was removed, and they were washed thrice with PBS. Fresh 1 mL medium and 1 mL JC-1 working liquid were added and thoroughly mixed, and incubated in a cell incubator at 37°C for 30 min. After that, the medium containing the probe was removed, the JC-1 buffer was applied thrice, and fluorescence images were obtained by confocal fluorescence microscopy.

2.18. Measurement of the nuclear damage was determined by γ-H2AX probe

SKOV3 cell tablets (5×10⁴/well) were prepared and divided into five groups (control group, DOX, Fe₃O₄/DOX@RBM@FA, Fe₃O₄@GOX/DOX@RBM@FA, and Fe₃O₄@GOX/DOX@RBM@FA(AMF)), and incubated at 37°C for 12 h. The magnetic field (390 kHz, 2.58 kA m⁻¹) was applied to Fe₃O₄@ GOX/DOX@RBM@FA (AMF) for 15 min. The medium was removed, and the field was washed with PBS. The fixing solution was added and left for 10 min. It was then removed, and the field was washed with washing solution thrice. The immunostaining sealing solution was added for 20 min. The immunostaining blocking solution was removed, γ-H2AX murine monoclonal antibody was added, and incubated for 1 h. The liquid was washed thrice, 10 min each time. Anti-mouse 488 was added and incubated at room temperature for 1h. Wash the liquid twice, 5-10 min each time. Nuclear staining solution (DAPI) was added and allowed to stain for 5 min. The nuclear staining solution was removed, and the field was washed with washing solution three times for 5 min each time. Finally, fluorescence images were obtained by confocal fluorescence microscopy.

2.19. Cell internal structure were detected by biological electron microscopy

SKOV3 cells (5×10⁴/well) were divided into five groups (control group, DOX, Fe₃O₄/DOX@RBM@FA, Fe₃O₄@GOX/DOX@RBM@FA, and Fe₃O₄@GOX/DOX@RBM@FA(AMF)), which were seeded into 24-well plates for 24 h. The magnetic field (390 kHz, 2.58 kA m⁻¹) was applied to Fe₃O₄@GOX/DOX@RBM@FA for 15 min. The medium was discarded and 2.5% glutaraldehyde fixing solution was added. The field was fixed at room temperature for 5 min and the cells were gently scraped down in one direction to avoid cell breakage. The cell fluid was sucked into the centrifuge tube with a pasteurized pipette and centrifuged (2500 rpm, 2 min). The cell masses were gently picked up and placed in a new electron microscope, fixating solution without the fixating fluid. The cells were then fixed at room temperature for 30 min away from light, and then transferred to 4°C for storage. The prepared samples were photographed by biological TEM.

2.20. Anti-tumor efficiency

SKOV3 (LUC-transfected cells) mouse subcutaneous tumor-bearing model was established, and different treatments were given when the tumor volume reached 100 mm³. The body weight and tumor volume were measured every 3 days. A week after the final injection, the mice were imaged. The tumor-bearing mice were euthanized, and histological staining and sequencing were performed. The tumor volume was measured with a caliper, and the formula was as follows:

$\text{Tumor\ volume\ }\left({\text{cm}}^{\text{3}}\right)=\text{1}/\text{2}\left(\text{tumor\ length\ }\times {\text{\ tumor\ width}}^{\text{2}}\right)$

An abdominal tumor-bearing mouse model was established and treated according to the design method, and the survival curve was drawn.

2.21. Histological analysis

Tumor tissues in different treatment groups were detected by immunohistochemical staining (GPX4, KI67, CD8, and TUNEL). The distribution of iron oxide was observed by Prussian blue staining. The image was obtained at 400X magnification.

2.22. Sequencing

Tumor tissues of different treatment groups were sequenced in the second generation. Through statistical analysis, differentially expressed genes and enrichment pathways were found.

2.23. Biosafety and toxicity of nanoparticles

After the various treatments for 7 days, the blood was collected for serum biochemistry and blood chemistry analyses. The major organs were harvested for Hematoxylin & Eosin (H&E) staining.

2.24. Statistical analysis

To ensure the highest level of experimental reliability, all animal studies in this work were designed in collaboration with biostatisticians to determine appropriate sample sizes through power analysis prior to experimentation. This a priori approach guarantees adequate statistical power in animal research. Furthermore, every experimental dataset presented represents consistent findings from three independent biological replicates, demonstrating the reproducibility of our results. Statistical Package for the Social Sciences (SPSS) version 11.0 software (SPSS Inc., Chicago, IL, USA) was used for statistical analyses. The data were presented as the mean ± standard deviation (SD) and analyzed by one-way ANOVA followed by the Students’ t-test. Statistical significance was set at p < 0.001 (very significant), p < 0.01 (highly significant), and p < 0.05 (significant).

3.1. Erythrocyte membrane and drug-carrying nanoparticles preparation and characterization

As illustrated in the central schematic Figure 1, we developed a biomimetic drug delivery platform through a multi-step preparation process. Initially, Fe₃O₄@DMSA@GOX nanoparticles were synthesized using iron(III) acetylacetonate and GOX as primary components, while simultaneously isolating high-purity erythrocyte membranes via hypotonic lysis. These elements were subsequently integrated to construct the complete bionic drug delivery system (Fe₃O₄@DMSA@GOX/DOX@RCM@FA). Comprehensive characterization by TEM confirmed the successful assembly of these nanostructures, revealing well-defined spherical morphology with an average diameter of 121.9 nm (Figure 1a). The TEM images clearly demonstrate both the integrity of the iron oxide core and the uniform coating of the erythrocyte membrane, with an approximate 35 nm shell thickness consistent with natural cell membrane dimensions. The element mapping (Figure 1b) confirmed that FDGDFM were composed of Fe, O, C, N, and P elements, which indicated the successful doping of Fe into the RBM capsulation. The results of the laser particle size analyzer (dynamic light scattering-DLS) show that the particle size of Fe₃O₄@DMSA is 60.7 nm, and 128.5 nm with the addition of GOX. Excessive RCM is added to Fe₃O₄@DMSA@GOX nanoparticles. Fe₃O₄@DMSA@GOX/DOX@RCM@FA-NPs with a particle size of 162.7nm were prepared by extrusion method (Figure 1c). Compared with Fe₃O₄@DMSA@GOX nanoparticles, the size of cell-coated nanoparticles uniformly increased by about 35 nm, which is similar to TEM and previous studies. In vitro long-term stability was evaluated by DLS assay, as well as the size of the synthesized nanoparticles. There was little change at 4 weeks in PBS (Figure S1), indicating high stability of drug-carrying nanoparticles. In addition, the potential on the surface of Fe₃O₄@DMSA@GOX nanoparticles is only −9.6±0.44 mV. After RBM coating, the surface potential of the FDGDFM nanoparticle is -16.9±1.36 mV (Figure 1d). The negative structure of FDGDFM nanoparticle can avoid the interaction with negatively charged such as serum components, and be retained in blood for a long time [34]. The results of the magnetic hysteresis loop show that the hysteresis loops of the magnetic nanostructures all pass through the zero point and there is no remanence, which are typical superparamagnetic characteristics (Figure 1e). Moreover, the X-ray diff raction (XRD) demonstrating successful assembly of FDGDFM nanoplatforms due to electrostatic and chemical adsorption (Figure 1f), compared with standard cards, it is confirmed that the synthesized nanomaterial is Fe₃O₄. In addition, the successful functionalization of FDGDFM nanoparticles was further confirmed by the FT-IR spectra (Figure 1g). After coupling GOX with Fe₃O₄@DMSA, changes of 1650 cm^-1 amide bond were detected. The results showed that compared with Fe₃O₄@DMSA group, Fe₃O₄@DMSA@GOX, Fe₃O₄@DMSA@GOX@DOX, and Fe₃O₄@DMSA@GOX@DOX@RBM@FA groups, there was an obvious peak shape at 1650 cm^-1, which confirmed that GOX was successfully modified on the surface of each group. As can be seen from the UV-vis spectra, an obvious absorption characteristic peak belonging to DOX was observed at 495 nm and 530 nm [35], which indicated that the DOX was successfully loaded into the RBM (Figure 1h). The DOX encapsulation was further confirmed by UV absorption spectra. The drug loading efficacy and encapsulation rate of DOX in FDGDFM nanoparticles were 32.4% and 89.4%, respectively. In this study, we used erythrocyte membranes to load DOX, which can increase the solubility of the DOX payload. As the original drug is insoluble in water, DOX is currently used in the form of DOX hydrochloride. Embedding DOX in the erythrocyte membrane can completely solve its solubility problem and eff ectively reduce the toxic side effects of adjuvants.

Close

Figure 1.

Characterization of Fe₃O₄@DMSA@GOX/DOX@RCM@FA nanoparticle. (a) Transmission electron microscopy (TEM) images of Fe₃O₄@DMSA@GOX/DOX@RCM@FA. (b) The EDS image of the FDGDFM nanoparticles. (c) The size distribution of the FDGDFM nanoparticles. (d) The zeta potential of Fe₃O₄@DMSA, Fe₃O₄@DMSA@GOX, Fe₃O₄@DMSA@GOX/DOX@ RBC@FA respectively. (e) Hysteresis regression image of Fe₃O₄@DMSA@GOX/DOX@RBC@FA. (f) XRD pattern of Fe₃O₄@DMSA@GOX/DOX@RBC@FA. (g) Infrared spectrum of Fe₃O₄@DMSA, Fe₃O₄@DMSA@GOX, Fe₃O₄@DMSA@ GOX/DOX and Fe₃O₄@DMSA@GOX/DOX@RBM@FA nanoparticles respectively. (h) UV-absorption spectra of Fe₃O₄@DMSA@GOX/DOX@RBM@FA nanoparticles. (i) The enzymatic reaction at different concentrations of FDGDFM nanoparticles.

Export to PPT

In summary, the results of the above characterization fully verified that the RBC membrane to achieve high efficiency encapsulation of Fe₃O₄ and DOX, and the nanoparticles have the advantages of stable properties and good biocompatibility (Figure S2).

3.2. MRI and magnetothermal imaging of Fe₃O₄

To assess the magneto-thermal transform efficacy, the Fe₃O₄ nanoparticles were subjected to AMF, and the temperature changes were recorded. As shown in Figure 2(a), the temperature in the Fe₃O₄-DMSA nanoparticles was significantly higher than the control group. Specifically, the temperature diff erence between the tumor and surrounding tissue in the Fe₃O₄-DMSA nanoparticles at the concentration of 1,000 μg/mL after 10 min AMF exposure increased by 31.3 ± 0.37°C, while the temperature diff erence was only 5.37±0.31°C in the control group (P<0.001), exhibiting the potent magneto-thermal transform efficacy of Fe₃O₄ nanoparticles. Furthermore, we measured the magneto-thermal transform eff ectiveness of Fe₃O₄ nanoparticles in tumor tissues after establishing a subcutaneous SKOV3 tumor murine model. After being given diff erent drug formulations, the tumor-bearing mice were exposed to AMF and imaged via an IR thermal camera. As shown in Figure 2(b), the temperature in tumor tissues was significantly higher in the Fe₃O₄-DMSA group compared with that in the blank PBS group, further confirming the potent magneto-thermal transform efficacy of Fe₃O₄ nanoparticles in vivo.

Close

Figure 2.

MRI and magnetothermal imaging of Fe₃O₄. (a) Photothermal image and statistical results of magnetothermal transition at different concentrations of Fe₃O₄ (100 ∼ 1000 μg/mL) under AMF. (b) Infrared thermal imaging images and temperature change curves of SKOV3 tumor-bearing mice under the action of AMF (AMF). (c) MRI results of different materials under different TE and concentration conditions in vitro. (d) The relaxation rate of different materials under TE =10 ms conditions. (e) MRI images of tumor-bearing mice with different materials. (f) Time-lapse images depicting the movement of the nanoparticles toward SKOV3 cells,and the mean square displacement diagrams and plane trajectories of nanoparticles under the attraction of an external magnet.

Export to PPT

Diagnosing and monitoring cancer is a key factor in improving prognosis and overall survival. As a commonly used clinical imaging detection means, MRI has improved the detection efficiency to a certain extent, but due to the lack of tumor targeting properties of commonly used clinical contrast agents, its sensitivity and specificity for tumorrecognition are still limited [36]. Fe₃O₄ nanoparticles can be used as contrast agents to enhance MRI imaging in addition to magnetothermal conversion efficiency and magnetic targeting eff ect [37]. We first measured the changes in T2 signal intensity after Fe₃O₄, Fe₃O₄-DMSA, and lopromide as positive controls. As shown in Figure 2(c), the T2 signal intensity of the Fe₃O₄ nanoparticle group and Fe3O4-DMSA group decreased gradually, while the T2 signal intensity of iodopropylamine group increased, and the signal was almost invisible after 50 s. The results showed that the comparison eff ect of Fe₃O₄ nanoparticles was better than that of iodopropylamine, and the loading of DMSA did not aff ect the T2 signal intensity. In addition, under 10 conditions, the relaxation rate of Fe₃O₄-DMSA was 118.39 mM^-1s^-1, which was significantly higher than that of Fe₃O₄ and iodopropyl groups, which were 24.25 mM^-1s^-1 and 1.79 mM^-1s^-1, respectively (Figure 2d). The T2 relaxation rate significantly increased after modification of DMSA, and these results demonstrate that Fe₃O₄-DMSA is an eff ective contrast agent. In addition, to evaluate the imaging eff ect of Fe₃O₄-DMSA in vivo, a subcutaneous tumor-bearing mouse model was constructed, and the mice were given MRI imaging after administration. As shown in Figure 2(e) and Figure S3, the contrast imaging effect of MRI in Fe₃O₄-DMSA group was significantly stronger than that in the control group, indicating the eff ectiveness of Fe₃O₄-DMSA for MRI enhancement imaging. The coated Fe₃O₄ nanoparticles can change the nuclear spin relaxation of water particles and darken the target region. Additionally, the Fe₃O₄ nanoparticle had a magnetic targeting eff ect, which could endow Fe₃O₄-DMSAwith a targeting eff ect under a magnetic field. As shown in Figure 2(f), the Fe₃O₄-DMSA could move toward the direction of the external magnet. Moreover, the mean square displacement diagrams and plane trajectories of Fe₃O₄-DMSA under an external magnetic field confirmed that the nanoparticle had the magnetic targeting eff ect and could be propelled to a targeted location under magnetic field conditions. In this study, we demonstrated that the Fe₃O₄-DMSA nanosystem can eff ectively improve the T2 relaxation rate of MRI and show high imaging contrast eff ect and sensitivity. As mentioned above, the prepared Fe₃O₄-DMSA showed excellent tumor targeting in the magnetic field direction, which made up for the lack of tumor targeting in existing MRI imaging contrast agents.

3.3. Sustained-release and tumor targeting in vitro and in vivo

To explore the release of nanoparticles in vivo and in vitro, the former environment was first simulated, the nanomaterials were placed in PBS, and the cumulative release curve of DOX was drawn. As shown in Figure 3(a), FDGDFM nanoparticles were released rapidly within 5 min, and the release tended to be stable with the passage of time. The cumulative release amount of FDGDFM nanoparticles was significantly higher than that of FDGDFM without AMF due to the increase in temperature, which proved that the nanoparticles had a slow-release effect, and the release amount was positively correlated with temperature. Subsequently, to verify its release effect in vivo, SD rats were injected intravenously with different drugs at different time points to take blood, and DOX content in blood was detected to draw pharmacokinetic curves. The DOX content in the blood of erythrocyte membrane coated nanoparticles peaked at about 20 h, which were significantly higher than that of free DOX (3 h), and the AUC of DOX in FDGDFM and FDGDFM (AMF) groups was 133 μg/mL/h and 160 μg/mL/h, respectively, about thrice that of free DOX group (Figure 3b). These favorable pharmacokinetic characteristics indicate that nanoparticles supported by the erythrocyte cell membrane can significantly prolong the in vivo cycle time of DOX. This long cycle feature benefits from the natural erythrocyte cell membrane, which can help the nanoparticle drug carrier system enhance the EPR effect during anti-tumor therapy, and then prolong the drug action time [38], thereby improving the therapeutic effect of tumor therapy. The increase in temperature did not prolong the release time but increased the drug content in vivo, which was consistent with the release in vitro.

Close

Figure 3.

Sustained-release and tumor targeting of nanoparticles. (a) The cumulative release of DOX from the FDGDFM nanoparticles with or without AMF. (b) Pharmacokinetics of SKOV3 tumor-bearing mice after intravenous injection of DOX, Fe₃O₄@DOX@RBM@FA, and Fe₃O₄@DOX@RBM@FA (AMF). (c) The uptake of DOX and DOX-NPs groups at 1 h and 2 h were observed by fluorescence microscopy. DOX showed green fluorescence after fluorescence excitation, and DAPI nuclear staining showed blue fluorescence (scale =20 μm), and the fluorescence intensities in the diff erent DOX groups analyzed by Image J software. (d) The distribution of DOX groups at 1 h and 2 h were observed by HPLC. (e) Drug tissue distribution of SKOV3 tumor-bearing mice in DOX and DOX-NPs groups at 1 h and 2 h. the IVIS imaging system. In vivo fluorescence images showing tumor retention of ICG, and ICG-NPs over a span of 72 h after intravenous administration via the tail vein in SKOV3 tumor-bearing mice. Note: *P<0.05, **P<0.01, ***P<0.001.

Export to PPT

The tumor-targeting ability of synthesized FDGDFM nanoparticles was detected in vitro. DOX and DOX-nanoparticles were incubated with SKOV3 cells for 1, 2, 4, and 8 h, respectively, and DOX fluorescence imaging was used to explore the uptake capacity of nanoparticle tumor cells. The cell uptake results have been shown in Figure 3(c) and Figure S4. After incubation for 1 h, cell uptake in the erythrocyte membrane loading group was 1.6 times that of the DOX group. DOX was promoted to enter the cell through endocytosis after the fusion of the erythrocyte membrane with the tumor cell membrane. In addition, the uptake of cells in the group loaded with Fe₃O₄ nanoparticles under the action of magnets was 2.5 times and 1.6 times higher than the DOX group and the group without magnets, respectively. The results showed that the magnetic targeting of Fe₃O₄ promoted the internalization of drugs. Moreover, after the nanoparticle coupling folic acid (FA), cell uptake was the highest, mainly because the folic acid molecule has an active tumor-targeting property, promoting more drugs to enter the tumor cells to enhance the tumor-killing effect.

To further explore the tumor-targeting specificity of FDGDFM in vivo, we constructed a tumor-bearing mouse model of SKOV3 and injected it intravenously with nanoparticles carrying or not carrying FA. Euthanized mice at different times were collected from tumor tissues and major organs, and DOX content in the tissues was detected by HPLC [38] to explore the targeting characteristics of tumor tissues based on FA (Figure 3d and Figure S5). The DOX content in 2 h tumor tissue of DOX@RBM@FA group was 111.70±0.29 μg/mL, which was significantly higher than that in DOX group alone (52.63±0.17 μg/mL), and the results at other time points were consistent with the 2 h results. These results once again demonstrate the tumor-targeting properties of FA. In addation, after 2 h of drug administration, the DOX concentration of Fe₃O₄@DOX@RBM (magnetic field) treatment in tumor tissues was 190.40±0.65 μg/mL, which was significantly higher than that of theFe₃O₄@DOX@RBM (59.65±0.27 μg/mL), suggesting the potent tumor-targeting capability of Fe₃O₄@DOX@RBM (magnetic field) due to the magnetic targeting effect of Fe₃O₄ nanoparticles (Figure S6).

Finally, we further verified the sustained-release and targeting ability of the synthesized FDGDFM nanoparticles in tumor-bearing mice. These SKOV3 tumor-bearing mice were injected intravenously with ICG, Fe₃O₄@DOX@RBM@ICGand Fe₃O₄@DOX@RBM@ICG@FA (ICG concentration was 1 mg/kg), and ICG signals were detected at 11 time points by the IVIS imaging system (Figure 3e). The results showed that the fluorescence signal in tumor tissues of RBM-NPs group was still very strong at each time point after 12 h, while ICG fluorescence signal was almost invisible in tumor tissues of ICG group, mainly due to the long circulation of erythrocyte membrane in vivo, which promoted the circulation of nanocarriers in the body and increased the distribution of tumor tissues. We further evaluated the tumor targeting properties of FA. The results showed that FA-nanoparticles mainly concentrated fluorescence in tumor sites, with less distribution in other tissues, while the other two groups were mainly concentrated in other organs such as the liver and kidney. The FA-nanoparticle group achieved the highest fluorescence intensity at the tumor site at 72 h, while the fluorescence intensity of other groups was lower at the tumor site, which proved the tumor-targeting property of FA (Figure S7).

3.4. Toxicity and efficiency therapy on OC cells.

The evaluation of the cytotoxicity of nanoparticles is very important for their biological application. First, co-cultured DOX and DOX-NPs with SKOV3 cells for 24 h with DOX (0.1-10 μg/mL). We found that the IC50 value of Fe₃O₄@GOX@DOX@RBM@FA (AMF) group for 24 h was 0.78 μg/mL. The IC50 values of DOX, DOX (AMF), Fe₃O₄@GOX@DOX@RBM, Fe₃O₄@GOX@DOX@RBM (AMF), and Fe₃O₄@GOX@DOX@RBM@FA groups were 4.89, 1.85, 2.56, 1.33, and 1.76 μg/mL, respectively. Due to the tumor-targeting property of FA and the magnetothermal property of Fe₃O₄ under AMF, Fe₃O₄@GOX@DOX @RBM@FA had better anti-tumor effects under the action of AMF, as shown in Figure 4, subpart Figure 4(a), and S6-S8. In addition, DOX and DOX-nanoparticles were co-cultured with SKOV3 cells for 8, 24, 48, and 72 h, and cell proliferation was detected by CCK8. We found that with the increase of DOX concentration and the extension of culture time, the tumor inhibition rate increased significantly, but the tumor inhibition rate in the DOX group did not increase significantly with the action of AMF, which proved that AMF had no toxic effect on OC cells (Figure S8). However, the cytotoxicity of Fe₃O₄-NP groups was significantly enhanced with the action of AMF. For example, when DOX concentration was 2 μg/mL, the SKOV3 cells’ survival rate of Fe₃O₄-NP nanoparticles without AMF effect was 64.9% after incubation for 24 h. However, SKOV3 cell survival decreased to 41.1% after increased AMF irradiation (Figure 4a). These results prove that AMF, as an effective treatment method for Fe₃O₄, can significantly improve the killing of tumor cells. However, when the nanomaterial was coupled with FA molecules, the survival rate of 24 h SKOV3 cells with and without AMF was 23.2% and 44.1%, respectively. This was due to the tumor-targeting properties of FA molecules, which promote more nanomaterials to enter the tumor interior, thus establishing a more effective tumor-killing effect.

Close

Figure 4.

Toxicity and efficiency therapy on OC cells. (a) CCK8 was used to detect tumor cell killing ability in DOX and DOX-NP groups of SKOV3 at diff erent time points. (b, c) The migration ability of tumor cells in diff erent groups was detected by a scratch test after 24 h of incubation. (d, e) Apoptosis detection via annexin V-FITC/PI assay in SKOV3 cells induced by PBS, DOX@RBM@FA, Fe₃O₄@DOX@RBM, Fe₃O₄@GOX@DOX@RBM@FA, and Fe₃O₄@GOX@DOX @RBM@FA (AMF) after 24 h of incubation. (f, g) Transwell assays were used to detect the invasion ability of tumor cells in different groups. Note: *P<0.05, **P<0.01, ***P<0.001.

Export to PPT

In addition, to vividly express the therapeutic effect of nanomaterials on tumor cells, we evaluated the survival of cells under the synergistic effects by migration, cloning experiments, and apoptosis. First, we measured the anti-migration effects of DOX and DOX-NPs in tumor cells, and we found that the AMF-exposed group had the lowest mobility of FDGDFM of all treatment groups. The mobility of all nanomaterials was significantly reduced compared to free DOX (Figure 4b and c). AMF-sensitizing Fe₃O₄ nanoparticles can significantly reduce migration and enhance the therapeutic effect of nanoparticles. We conducted a transwell assay to further examine the anti-invasion ability of FDGDFM on tumor cells. As shown in Figure 4(f) and 4(g), the anti-invasion effect of FDGDFM on tumor cells in the AMF group was significantly higher than in other treatment groups. It is worth noting that the anti-invasion ability of FDGDFM without AMF treatment was significantly stronger than that of FDGDFM without Fe₃O₄ nanoparticles. This result is consistent with the above scratch test results, which fully proved that Fe₃O₄ nanoparticles could reduce the invasion and migration ability of tumor cells under AMF irradiation. The coupling of FA promotes more DOX to enter the tumor cells and kill them. The two have a synergistic effect in inhibiting the invasion of tumor cells. At the same time, the effect of FDGDFM on apoptosis of tumor cells was significantly enhanced, as shown in Figure 4(d,e). The apoptosis rate of FDGDFM (AMF) group was as high as 51.4%, while those of control group, DOX group, FDGDM group, and FDGDFM group were 7.1%, 36.3%, 33.1%, and 36.4%, respectively. The results indicated that FAGDFM (AMF) had the most prominent effect on apoptosis of tumor cells. In summary, the tumor targeting properties based on FA and the enhanced tumor killing effect under the action of Fe₃O₄ nanoparticles AMF can greatly reduce the distant metastasis of tumor cells and improve cell apoptosis, thus playing an anti-tumor role.

3.5. Efficiency and mechanism of DNA damage and ferroptosis therapy on cancer cells.

Due to the tumor targeting ability and excellent tumor killing effect of synthetic FDGDFM nanoparticles in vitro and in vivo, we carefully studied the anticancer properties of FDGDFM nanoparticles by directly or indirectly destroying the DNA of tumor cells and inducing ferroptosis. DNA damage was detected by phosphorylated histone H2A. X (γ-H2AX, phosphorylated at serine [39,40]); fluorescence images have been shown in Figure 5(a). It is generally believed that the change of fluorescence intensity of γ-H2AX antibody in the five groups of mice is consistent with the results of CCK8 and cloning. Compared with the DOX group, the fluorescence point density of γ-H2AX in the nucleus of the FA-nanoparticle group was higher, indicating that FA, as a tumor targeting molecule, can significantly improve the targeting of nanoparticles to cancer cells. In addition, when coated with GOX nanomaterials, its fluorescence intensity was significantly higher than DOX, because GOX, as a key enzyme of glucose oxidation, can catalyze glucose to produce a large amount of H₂O₂ to cause DNA damage [41]. Importantly, the largest DNA damage generation was observed in the AMF+ FDGDFM nanocomposite group, suggesting that the coordinated kill of FDGDFM nanoparticles produces more ¹O₂, thus accelerating deep damage to SKOV3 cells.

Close

Figure 5.

Efficiency and mechanism of DNA damage and ferroptosis therapy on cancer cells. (a) Representative γ-H2AX immunofluorescence images of SKOV3 cells incubated with PBS, DOX@RBM@FA, Fe₃O₄@DOX@RBM, Fe₃O₄@GOX@DOX@RBM@FA, and Fe₃O₄@GOX@DOX@RBM@FA (AMF)(scale bar=20 μm). (b) Mitochondrial transmembrane potential of SKOV3 cells after different treatments tested using JC-1 fluorescent probes. Green fluorescence represents JC-1 monomers (λex = 514 nm, λem = 529 nm), and red fluorescence denotes JC-1 aggregates (λex = 585 nm, λem = 590 nm) (scale bar=20 μm). (c) The ability of different groups to produce ROS with or without AFM. Green fluorescence represents ROS (scale bar=20 μm). (d) The ability of different groups to producelipid peroxide with or without AMF. Blue fluorescence represents DAPI-stained nucleus, green and red fluorescence represents lipid peroxide (scale bar=20 μm). (e) Morphological assessment of mitochondria by bio-TME after treatment of SKOV3 cells with various nanoparticles.

Export to PPT

Abnormal mitochondrial membrane potential, ROS production, and LPO/LP are the signs of ferroptosis [41,42]. The abnormal change of mitochondrial membrane potential is not only an important marker of mitochondrial damage, but also an early warning of ferroptosis. We used JC-1 fluorescent probe to detect the changes of mitochondrial membrane potential and investigate the occurrence of ferroptosis. The green fluorescence of FDGDFM+ AMF group was significantly enhanced (Figure 5b), and the mitochondrial membrane potential was significantly reduced. The results showed that the proportion of SKOV3 cells increased in mitochondrial depolarization, suggesting that FDGDFM nanoparticles promoteferroptosis in tumor cells under the action of AMF.

Subsequently, to verify intracellular ROS production, ROS fluorescence probes were used to detect ROS levels in SKOV3 cells of different treatment groups by fluorescence imaging. Compared with the PBS group, the FDGDFM group could observe a stronger ROS fluorescence signal. After AMF treatment, the fluorescence signal of cells treated with FDGDFM nanoparticles was most significantly enhanced (Figure 5c). Finally, the C11 BODIPY fluorescent probe was used to detect cell LPO. According to the CLSM image (Figure 5d), only when FDGDFM was exposed to AMF was a significant enhancement of the green LPO fluorescence of SKOV3 cells observed. Quantitative FCM analysis further showed that the LPO/LP value of FDGDFM + Light group was about 2.9 times higher than that of the FDGDFM group. To observe the occurrence of ferroptosis, the morphology of SKOV3 cells was observed by TEM. Biological TEM showed that DOX-treated cells only showed significant mitochondrial swelling, while the FDGDFM(AMF) group showed massive mitochondrial atrophy, increased mitochondrial membrane density, and decreased (or disappeared) mitochondrial ridge, which were typical morphological features of ferroptosis (Figure 5e), which was consistent with previous reports. Combined with the above results, we systematically demonstrated that FDGDFM nanoparticles under the action of AMF can induce lethal ferroptosis in cancer cells.

3.6. Anticancer performance of FDGDFM nanoparticles in vivo

Encouraged by the excellent anti-tumor properties and immunomodulatory effects of the nanoparticles synthesized in this study, we also conducted anti-tumor experiments in tumor models derived from human cell lines. BALB/c mice implanted with SKOV3 tumors with a tumor size of 100 mm³ were randomly divided into five groups: (1) PBS, (2) DOX, (3) Fe₃O₄@GOX@DOX@FA@RBM nanoparticles, (4) AMF+Fe₃O₄@GOX@DOX@FA@RBM, and (5) Fe₃O₄@DOX@FA@RBM. The treatment plan has been detailed in Figure 6, subpart Figure 6(a). Tumor size and weight were recorded within 32 days, and tumor weight was measured at 32 days. The tumor of PBS group grew quickly, while the DOX, Fe₃O₄@GOX@DOX@FA@RBM nanoparticles, and Fe₃O₄@DOX@FA@RBM nanoparticles treatments showed moderate inhibitions of tumor growth. Notably, the AMF+Fe₃O₄@GOX@DOX@FA@RBM group treatment significantly inhibited the growth of tumors (Figure 6b, c, and f). Compared to the positive control, the final tumor volume in the Fe₃O₄@GOX@DOX@FA@RBM (AMF) group reduced by 62.69%, while the reductions in tumor volume in DOX, Fe₃O₄@GOX@DOX@FA@RBMnanoparticles, and Fe₃O₄@DOX@FA@RBM were 22.84%, 56.54%, and 43.62%, respectively. After the mice were sacrificed, the tumors were separated and weighed. As shown in Figure 6(d), the tumor weight in the Fe₃O₄@GOX@DOX@FA@RBM(AMF) group was the smallest compared to the other groups, displaying a 2.05 fold (P<0.001), 1.37 fold (P<0.001), 0.98 fold (P<0.001), and 0.341 fold (P<0.05) lower than PBS, DOX, Fe₃O₄@GOX@DOX@FA@RBMnanoparticles, and Fe₃O₄@DOX@FA@RBM nanoparticles groups, respectively, demonstrating that the Fe₃O₄ (AMF) treatment could significantly inhibit the growth of tumor in mice. Interestingly, the change in average body weight between treatments was negligible, indicating little toxic side effects caused by the constructed nanoparticles (Figure 6e). To further verify the distribution of FDGDFM nanoparticles in vivo, we stained the vital organs with Prussian blue after FDGDFM nanoparticles were treated. As shown in Figure 6(g), ferric oxide was more distributed in tumor tissues, which was consistent with the results of cell-uptake and tissue distribution, again proving the tumor targeting properties of this nanoparticle.

Close

Figure 6.

In vivo antitumor effects of different treatment groups. (a) Diagram of treatment plans in subcutaneous tumor-bearing mouse models with different treatments. (b, c) Fluorescence imaging and photographs map of tumor tissue in tumor-bearing mice. (d) Weight of the primary tumor from the mouse after treatment. (e) Relative mouse body weight during treatment. (f) Growth curve of the subcutaneous tumor-bearing mouse model with indicated treatments. Data are shown as mean ± SD (n = 5). (g) Prussia blue staining of tumors from mice treated with Fe₃O₄@GOX@DOX@RBM@FA (scale bar=200 μm). Note: *P<0.05, **P<0.01, ***P<0.001.

Export to PPT

According to the above results, FDGDFM nanoparticles under the action of AMF have remarkable anti-cancer properties, and the results in vivo and in vitro are consistent. By catalyzing glucose to form H₂O₂, GOX provides catalytic raw materials for Fenton reaction and promotes the generation of more ROS. The coupling of FA molecules can also significantly improve the tumor killing ability, mainly because FA has tumor targeting properties and can promote more FA nanoparticles to bind specifically to tumor cells. Erythrocyte membrane, as the carrier of DOX, not only has the characteristics of long circulation in vivo, but also can promote its fusion with tumor cell membrane and promote the intracellular release of Fe₃O₄ and DOX. Fe₃O₄ has a significant anti-cancer effect under the action of AMF, mainly because trioxide produces a large amount of magnetic hyperthermia under the action of AMF and can synergically promote chemotherapy drug sensitization and induce ferroptosis.

In addition, we also evaluated the anti-metastasis ability of FDGDFM nanoparticles. In Figure 7, first, an abdominal-cavity tumor-bearing mouse model was constructed. After randomly grouping the mice into different groups, the mice were euthanized, and the tumor-bearing mice were dissected to observe the tumor distribution, as shown in Figure 7(b). The results show that compared with other treatments, Fe₃O₄@GOX@DOX@FA@RBM (AMF) treatment could significantly reduce tumor metastasis, but no tumor metastasis was observed in kidney, abdominal wall and uterine organs, and the tumor metastasis rate in liver, spleen and stomach organs was only 20%, and intestinal tumor metastasis rate was only 40%, which was significantly better than that in other groups. These results indicate that Fe₃O₄@GOX@DOX@FA@RBM(AMF) has a strong anti-metastasis effect in mice. In vivo metastasis was consistent with in vitro scratch and invasion and migration tests. Finally, in order to evaluate the effect of Fe₃O₄@GOX@DOX@FA@RBM nanomaterials on the survival rate of mice and the treatment scheme has been detailed in Figure 7(a). The survival of mice was observed up to 120 days, and the survival curve was drawn (Figure 7c). Median survival time results were PBS < DOX= Fe₃O₄@DOX@FA@RBM<Fe₃O₄@GOX@DOX@FA@RBM<Fe₃O₄@GOX@DOX@FA@RBM (AMF). The results of survival trend in each group were consistent with the results of tumor cell killing in vitro and anti-cancer effect in vivo. In conclusion, Fe₃O₄@GOX@DOX@FA@RBM(AMF) effectively reduced the metastasis of tumor cells and significantly increased the survival time of mice. The above results may be due to the synergistic effects of FA tumor targeting, RBM nanocore long cycling, Fe₃O₄ magnetothermal generation, ferroptosis induction and chemotherapy sensitization, which jointly improve tumor killing.

Close

Figure 7.

In vivo survival verification with indicated treatments. (a) Diagram of treatment plans in abdominal tumor-bearing mouse models with different treatments. (b) Heat map of tumor metastasis frequency in tissues and organs. (c) Kaplan–Meier survival graph of abdominal tumor-bearing mice after treatment. (d) H&E analysis of different tissues/organs with different treatments (scale bar=200 μm). (e,f) The serum biochemistry and blood chemistry between the control group.

Export to PPT

Moreover, the RNA sequencing results of tumor tissues in each group were collected (Figure 8a). The results showed that compared with the control group, 20 differentially expressed genes were screened in Fe₃O₄@GOX@DOX@FA@RBM (AMF) group, more than half of which were important genes in the apoptosis pathway.

Close

Figure 8.

Explore the mechanism of anti-tumor action. (a) KEGG pathway databases of different groups sequencing. (b) The TCGA database analysis showed that Gpx4 was highly expressed in OC tissues. (c) Gpx4 expression level was negatively correlated with the prognosis of OC patients. (d-e) The relationship between Gpx4 expression and OC staging and lymph node metastasis simultaneously. (f) GPX4, CD8, Ki67, and TUNEL staining of tumor tissue with different treatments (scale bar=200 μm).

Export to PPT

3.7. KEGG enrichment analysis for ferroptosis-related pathway (TFRC/FTL/ACSL3/some/SLC7A11/GPX4/SLC3A2/ACSL5)

GPX4 protects cells from LPO through its enzyme activity, thereby maintaining cell survival. TCGA database analysis showed that the high expression of Gpx4 in OC tissues (Figure 8b) was negatively correlated with the prognosis of OC patients (Figure 8c). In addition, our research group also analyzed the relationship between Gpx4 expression with OC staging and lymph node metastasis (Figure 8d-e). These results suggest that this nanoparticle increases the iron content in tumor cells by down-regulating the expression of Gpx4 and subsequently induces ferroptosis in tumor cells. To verify that the nanoparticle can regulate the expression of key molecules of ferroptosis, we collected tumor tissues from different treatment groups and performed immunohistochemical staining of the GPX4 molecule. As a key inhibitory protein of ferroptosis, GPX4 plays a crucial role in preventing ferroptosis [43]. Results As shown in Figure 8(f), Gpx4 expression in tumor tissues of Fe₃O₄@GOX@DOX@FA@RBM(AMF) group was significantly reduced, suggesting that the nanomaterials might induce ferroptosis by inhibiting Gpx4 expression. The results of sequencing and immunohistochemistry provide support for the tumor-killing of ferroptosis in vitro and in vivo and the detection of important indicators of ferroptosis.

Subsequently, tumor tissues from different treatment groups were collected and analyzed by immunohistochemical staining. Proteins such as Ki67 and TUNEL play a crucial role in cell growth and differentiation [44-46]. The results have been shown in Figure 8(f). Fe₃O₄@GOX@DOX@FA@RBM(AMF) showed the best anti-proliferation effect, with only 37% of Ki67+ cells (Figure 8f). Overall, Fe₃O₄@GOX@DOX@FA@RBM nanoparticles under AMF conditions showed perfect anti-tumor effects in vivo.

Finally, we used ID8 cells to construct a C57 tumor-bearing mouse model and explored the immune regulation of nanoparticles. Interestingly, we found that CD8⁺ T lymphocytes in tumor tissues of Fe₃O₄@GOX@DOX@FA@RBM (AMF) group was significantly increased after treatment, which was consistent with previous reports that ferroptosis could promote an immune-response. The results provided a way for further investigation of immunity induced by ferroptosis.

3.8. Biosafety and toxicity of FDGDFM nanoparticles.

We carefully investigated the potential toxic effects of FDGDFM nanoparticles in vitro and in vivo. The cytotoxicity of blank nanoparticles in SKOV3 cells was using a CCK8 assay. As shown in Figure S2, when Fe₃O₄ concentration was as high as 500 μg/mL, the cell survival rate remained above 95% at 8 h. In addition, after the SKOV3 tumor-bearing model was treated for 7 days, the blood of mice in each group was systematically collected for blood chemistry and serum biochemical analysis. There were no significant differences in serum biochemistry and blood chemistry between the control and treatment groups (Figure 7e-f). Finally, major organs, including vital organs such as the heart, liver, spleen, lungs, and kidneys, were collected, and H&E staining was performed. No significant pathological abnormalities were found in the sections of major organs (Figure 7d). The average body weight of mice in each group was negligible (Figure 6e). According to the above results, the FDGDFMnanoparticles synthesized by us have good biocompatibility.