Eriodictyol-cisplatin coated nanomedicine synergistically promote osteosarcoma cells ferroptosis and chemosensitivity

Abstract

The ever-increasing chemoresistance of osteosarcoma (OS) has been observed in the recent decades, impeding OS therapeutic improvement and posing an urgency to exploit to the alternative and/or supplementary therapies for the optimization of OS chemotherapeutic regimen. Ferroptosis, a regulated cell death, has been identified as a natural anticancer mechanism as well as a synergist for chemotherapeutics in various cancers. Herein, we affirmed the tumor-suppressing properties of eriodictyol and illustrated that its antitumor effects might ascribe to the ferroptosis-inducing activity, in which eriodictyol could bind with BACH1 to repress the transcription and translation of GPX4 and eventually result in the GPX4-related ferroptosis. Further investigation found that eriodictyol could exhibit a synergistic effect with cisplatin, facilitating the antitumor effects of cisplatin. Lastly, through utilizing hollow mesoporous prussian blue nanocubes loaded with eriodictyol and cisplatin, we formed the ferroptosis-synergistic nanocomplexes to facilitate OS cells ferroptosis and cisplatin sensitivity. Through direct catalytic oxidation of unsaturated lipids, exogenous iron delivery, GSH exhaustion, and GPX4 transcriptional inhibition, this ferroptosis-synergistic nanocomplex could excellently enhance OS cells ferroptosis in both vitro and vivo, with no obvious organ injury observed. Therefore, our ferroptosis-synergistic nanocomplex may represent a promising alternative therapeutic strategy for OS patients.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-025-03206-3.

Introduction

The standardized osteosarcoma (OS) treatment combines surgical excision, neoadjuvant chemotherapy and adjuvant chemotherapy, which has excitingly promoted the OS therapeutic effect over the decades [1–3]. However, the increasingly austere chemoresistance has hindered the advancement of OS treatment, suggesting that enhancing chemotherapy sensitivity or reversing chemoresistance is essential for OS treatment [1, 4–6]. Recent studies have found that eriodictyol, a natural flavonoid compound, exhibits anti-cancer potential and regulates multiple vital tumor-associated pathways in various cancers [7–12]. However, the role of eriodictyol in OS occurrence and development remains unexplored. In this study, we validated that eriodictyol exerted the anti-tumor property against OS via proliferative inhibition and apoptotic promotion. Mechanically, eriodictyol could stabilize BTB Domain and CNC Homolog 1 (BACH1) proteins, a ferroptosis-promoting transcriptional factor, to reduce the transcription of glutathione peroxidase 4 (GPX4) and trigger OS cells ferroptosis. Intriguingly, recent researches have revealed a heightened susceptibility to ferroptosis in cancer cells exhibiting multidrug resistance, particularly those in a mesenchymal state with a propensity for metastasis [13–20]. Since the single-targeting treatments was insufficient for OS treatments, the synergistic application of ferroptosis inducers and conventional chemotherapeutic medication may offer promising avenues for the advancement of novel therapeutic strategies in OS treatments.

Though ferroptosis is regarded as an extremely promising strategy for cancer treatment, the practice of the past 10 years has witnessed that its tumor therapeutic effect is always limited by insufficient intracellular oxidation level, limited iron accumulation, and less favorable pH levels [21–23]. As we known, in the context of ferroptosis, GSH, the potent reducing agent, functions as a cofactor for GPX4, facilitating the intracellular reduction of phospholipid hydroperoxides (PLOOHs) to their corresponding alcohols (PLOHs) while GSH depletion will trigger the fulminant iron-dependent oxide accumulation and concomitant ferroptosis [13, 14]. Based on this phenomenon, increasing the intracellular oxidation level by using chemotherapeutics or GSH exhaustion, endogenous iron release and exogenous iron delivery can serve as feasible strategies to improve ferroptosis effect [24–26].

In this work, our drug synergistic experiments demonstrated that eriodictyol could enhance the sensitivity of OS cells to cisplatin. Notably, recent studies found that apart from binding to DNA and interfering with DNA replication and transcription, the anti-tumor mechanism of cisplatin could derived from intracellular cisplatin in conjugation with GSH to result in the formation of the Pt-GS complex and GSH depletion [27–29]. Therefore, the combination of cisplatin and ferroptosis inducers may be a potential strategy to enhance OS platinum-related chemotherapeutic effects and ferroptosis sensitivity. Prussian blue (PB) was reported to trigger Fenton reaction-independent ferroptosis for its direct catalytic oxidation of unsaturated lipids by its mixed valence transition metal (Fe²⁺/Fe³⁺) [30]. Herein, we chose hollow mesoporous prussian blue (HMPB) nanocubes, which exhibits an outstanding specific surface area along with electron transfer site, as a shell material [31, 32], and through utilizing HMPB nanocubes loaded with eriodictyol and cisplatin, we formed the ferroptosis-synergistic nanocomplexs (NCs) to facilitate OS cells ferroptosis and cisplatin sensitivity. Different from the application of chemotherapeutic drugs or ferroptosis alone, in this ferroptosis-synergistic nanocomplex, HMPB NCs, as a shell material, could not only provide Fenton reaction-independent ferroptosis and deliver iron ion, but also could diver eriodictyol and/or cisplatin. Meanwhile, the synergistic effect of eriodictyol and cisplatin in our ferroptosis-synergistic NCs could excellently enhance OS cells ferroptosis. Therefore, our ferroptosis-synergistic nanocomplex could facilitate OS cells ferroptosis through direct catalytic oxidation of unsaturated lipids, exogenous iron delivery, GSH exhaustion, and GPX4 transcriptional inhibition, providing a novel strategy to improve the effect of tumor treatment.

Methods and materials

Tissue specimens and patient information

The paraffin-embedded OS samples and their paired adjacent normal tissue samples were collected from patients undergoing tumor resection at the Department of Orthopedics, Xiangya Hospital, Central South University. This study was implemented in line with Declaration of Helsinki and approved by the Ethics Committee of Xiangya Hospital, Central South University (Approval number: 2024030282). The details of experiments involving humans sample were made known to the patients or their legal guardians and informed consents were obtained from them.

Cells, reagents and transfection

The human OS cell lines, U2OS and MG63 cells, and 293T cells were obtained from Pricella Company (Wu Han, China). U2OS, MG63, and 293T cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (Biological Industries, Israel) added with 10% fetal bovine serum and antibiotics (Gibco, USA) and maintained at the circumstance of 5% CO2 and 37℃.

The antibodies used in this study were as follows: anti-BACH1 (14018-1-AP, proteintech, China), anti-Ki67 (ab15580), anti-GPX4 (30388-1-AP, proteintech, China), anti-β-tubulin (10094-1-AP, proteintech, China). Eriodictyol (552–58-9, MCE), Autophinib (1644443-47-9, MCE), NLRP3/AIM2-IN-3 (1787787-60-3, MCE), Necrostatin-1 (4311-88-0, MCE), Ferrostatin-1 (347174-05-4, MCE), Deferoxamine mesylate (138-14-7, MCE), and Cuprizone (370-81-0, MCE) were purchased from MedChemExpress company. Tunel kit (C1089, Beyotime, China), GSH and GSSG test kit (S0053, Beyotime, China), EDU Kit (C0075L, Beyotime, China), DAPI solution (C1002, Beyotime, China), Lipid oxidation (MDA) test kit (S0131M, Beyotime, China), Glutathione peroxidase test kit (S0056, Beyotime, China), Cell Counting Kit-8 (CCK-8) (C0038, Beyotime, China), Actin-Tracker Green-488 (C2201S, Beyotime, China) and dihydroethidium (S0063, Beyotime, China) were obtained from Beyotime biocompany. Liperfluo probe (L248, Dojindo, Japan), FerroOrange probe (F374, Dojindo, Japan), and JC-1 MitoMP Detection Kit (MT09, Dojindo, Japan) were purchased from Dojindo company.

BACH1, GPX4 and and their corresponding negative control plasmids used for dual-luciferase, and siBACH RNA were synthesized and purchased from General Biosystems Company (Nanjing, China). Cy5.5 (TD0091) was purchased from TargetMOI company. Recombinant human BACH1 proteins (R2409) were purchased from FineTest biocompany.

Hematoxylin–eosin (H&E), immunohistochemistry (IHC), immunocytochemistry (ICC), and tissue immunofluorescence staining

H&E, IHC, and tissue immunofluorescence staining were performed for formalin-fixed, paraffin-embedded tissue sections using BACH1 and Ki67 antibodies, hematoxylin, eosin, and tunnel solution, and OS cell slides incubated with BACH1 and GPX4 antibodies were utilized to the application of ICC. The details of these experiments were implemented as described in our previous studies [33].

Western blotting

Total proteins were extracted from cell lysates, dispersed by 10% SDS polyacrylamide gel electrophoresis, and then transferred onto the polyvinylidene membrane. After incubated in 5% defatted milk, the membranes were incubated overnight with primary antibodies at 4 °C and were then incubated with secondary antibodies for 1 h. Eventually, blot signals of the membranes were visualized using an enhanced ECL kit (BL520A, Biosharp, China).

Cell proliferation and colony formation assays

For the CCK-8 and EDU assays, cells with the density of nearly 3000 cells/well and 10,000 cells/well were seeded in the 96-well plate, respectively. For the colony formation assay, cells with the density of about 1000 cells/well were cultured in the 6-well plate. After cells attached to the well, the corresponding treatments were performed to the treated groups and dimethyl sulfoxide or saline with the corresponding concentration was added to the negative control groups. The details of these experiments were performed as our previous studies described [33].

Cell death assays

In term of the tunnel assay and the cell apoptosis assay, cells with the density of about 3000 cells/well and 30,000 cells/well were seeded in the 12-well plate and the 6-well plate, respectively. Subsequently, the corresponding treatments were conducted for the corresponding treated groups after cells attached to the well. After 24-h treatment, the tunnel kit and the cell apoptosis kit were utilized for the detection of cells death in line with the manufacturer’s instructions.

The detection of the intracellular Fe(II) concentration

The tumor cells were cultured in the 12-well plate and the corresponding treatments were applied after cells were attached to the well. After treating 24 h, Fe(II) probe was utilized to investigate the intracellular Fe(II) concentration in accordance with the manufacturer's instructions.

The detection of the intracellular oxidative species

The cancer cells were seeded in the 6-well plate and then cells were exposure to the corresponding treatments 24 h. Subsequently, MDA kit, DHE and liperfluo probe were used to explore the level of the oxidative species in line with the manufacture’s instruction.

Detection of the relative glutathione peroxidase, GSH and GSH/GSSG

The cancer cells were seeded in the 6-well plate and then cells were exposure to the corresponding treatments 24 h. Subsequently, glutathione peroxidase test kit and GSH and GSSG test kit were used to explore the level of the relative glutathione peroxidase, GSH and GSH/GSSG in line with the manufacture's instruction.

Detection of mitochondrial membrane potential (MMP)

JC-1 kit was applied to detect MMP of the treated cells, in which cells exhibited poly JC-1 represented the cells with less damaged mitochondrion and cells exhibited mono JC-1 represented reversed situation. In brief, the tumor cells were seeded in the 12-well plate, treated with the corresponding treatments 24 h, and imaged using fluorescence microscope.

Transmission electron microscopy (TEM)

Cells were grown in a 10 cm culture dish, fixed with 2.5% glutaraldehyde for 5 min, detached using a cell scraper, and centrifuged for 2 min. The resulting pellet was then fixed in the dark at room temperature for an additional 30 min. Subsequently, the cells were treated with 1% aqueous osmium tetraoxide. Following dehydration in a series of ethanol (30–100%) and acetone, the pellet underwent gradual infiltration with anhydrous acetone and epoxy resin overnight, followed by embedding in resin and polymerization at 60 °C for 48 h. The embedded samples were sectioned into thick slices using an ultra-microtome (Leica EM UC7). Ultrathin sections were mounted on copper grids, stained with uranium acetate and lead citrate, and examined using 120 keV transmission electron microscopy (Talos L120C G2).

Transcriptome sequencing

The transcriptome sequencing was conducted by OE Biotech Co., Ltd. (Shanghai, China). In brief, the total RNAs were extracted from the IC50-eriodictyol-treated cells and the corresponding negative control cells using TRIzol reagent (AG, Changsha, China) in line with the manufacturer's instruction. Subsequently, a series of processes including quantification, qualification, library preparation, sequencing, and bioinformation analysis were performed for the extracted RNA samples. Differential expressed analysis (DEA) was implemented using the method of DESeq2, and Q value < 0.05 and foldchange > 2 or foldchange < 0.5 were considered as the selective threshold of the statistically significant differential expression genes (DEGs). Gene set enrichment analysis (GSEA) was performed for the investigation of biological function alteration derived from eriodictyol treatment.

Chromatin immunoprecipitation (ChIP) and Dual-luciferase assays

ChIP kit (Bersinbio, Guangzhou, China) was utilized for ChIP assays. In brief, a standardized operation including crosslinking, lysis, chromatin extraction, immunoprecipitation, anticrosslinking, and DNA purification were implemented for cells of each treatment group. Subsequently, the purified DNA segment were utilized for qPCR. The sequence of primers were as follows: BACH1-forward primer: CAGGAGAATCACTTGAACCCG, BACH1-reverse primer: CACTTCGTGGCTTGGACAT.

Luc-PairTM Duo Luciferase HS Assay Kit (GeneCopoeia, Guangzhou, China) was utilized for the dual-luciferase assay. In brief, 239T cells were seeded in the 12-well plates, infected with the luciferase reporter plasmid and treated with eriodictyol. After treated 48 h, cells of each treated group were cleaved and utilized for fluorescence detection according to the manufacturer's instructions.

Molecular docking and surface plasmon resonance (SRP)

The molecular docking was performed to investigate the docking stiuation between BACH1 proteins and eriodictyol. In brief, the 2-dimensional structure of eriodictyol was depicted via Chemdraw software, energy-minimized in the MM2 manner using Chemdraw 3D, and ultimately the optimal stucture with the minimal energy saved as mol2 file. Subsequently, the molecular structure of human BACH1 protein, PDBID of which called 2ihc, was downloaded from PDB database, visualized using pymol, and undergone dewatering, hydrogenation, charge calculation and non-polar hydrogen merging using Mgtools 1.5.6. Then, the molecular structures of BACH1 protein and eriodictyol were saved as pdbqt files and docked using Autodock vina 1.1.2 with the parameters (center_x = 16, center_y = -2, center_z = 55, size_x = 20.0, size_y = 20.0, size_z = 20.0, exhaustiveness = 8, num_modes = 9, energy_range = 3). Ultimately, the optimum construction with the higher score was visualized using pymol and Discovery studio software. SPR assay was conducted by Gene Universal company to assess the affinity constant between recombinant human BACH1 and eriodictyol.

The optimum ratio of the drug combination

The optimum ratio of drug combination was determined using the method provide by SynergyFinder website [34]. Briefly, U2OS or MG63 cells were seeded to a 96-well plate (1 × 10⁵ cells/well), then treated with eriodictyol and cisplatin in a series of concentration for 48 h. Then, the cell inhibitory rate were measured using CCK8 solution, and optimum ratio of drug combination were calculated using SynergyFinder R package.

Synthesis and characterization of HMPB@ERY, HMPB@PT, and HMPB@ERY@PT

First, we synthesize HMPB for subsequent nanomedicine synthesis. In brief, 6 g of polyvinylpyrrolidone (PVP) and 1 g of piscine gelatin were dissolved in 60 ml of 0.01M HCl. After complete dissolution, 300 mg of potassium ferrocyanide were added at room temperature, and the mixture was stirred for 30 min. Next, the mixture was placed in an oil bath and stirred at 80 °C for 24 h. After completion, centrifugation and washing were performed, and the product was freeze-dried and weighed. Then, 30 mg of the synthesized sample were re-dispersed in 30 ml of 1M hydrochloric acid. After thorough dispersion, 120 mg of PVP were added, and the mixture was stirred at room temperature for 4 h. Subsequently, the mixture was placed in a high-pressure reactor and maintained at 140 °C for 4 h, followed by centrifugation and washing.

For synthesis of HMPB@ERY, 4 ml HMPB at a concentration of 1 mg/ml was centrifuged, and the supernatant was discarded before freeze-drying for 24 h. Next, 4 mg eriodictyol were weighed and dissolved in 4 ml DMSO. The eriodictyol solution was then slowly added dropwise to the HMPB powder while simultaneously subjecting the mixture to ultrasonic treatment. The mixture was allowed to shake overnight at room temperature. Finally, centrifugation and washing were performed, and the supernatant was collected to calculate the drug loading capacity.

For synthesis of HMPB@PT, 4 ml HMPB at a concentration of 1 mg/ml was centrifuged, and the supernatant was discarded before freeze-drying for 24 h. Next, 4 mg cisplatin was weighed and dissolved in 4 ml water. The cisplatin solution was then slowly added dropwise to the HMPB powder while simultaneously subjecting the mixture to ultrasonic treatment. The mixture was allowed to shake overnight at room temperature. Finally, centrifugation and washing were performed, and the supernatant was collected to calculate the drug loading capacity.

For synthesis of HMPB@ERY@PT, 4 ml HMPB at a concentration of 1 mg/ml was centrifuged, and the supernatant was discarded before freeze-drying for 24 h. Next, 4 mg cisplatin was weighed and dissolved in 4 ml water. The cisplatin solution was then slowly added dropwise to the HMPB powder while simultaneously subjecting the mixture to ultrasonic treatment. The mixture was allowed to shake overnight at room temperature. Afterward, centrifugation and washing were performed, and the supernatant was collected to calculate the drug loading capacity. The resulting centrifuged precipitate was then freeze-dried. Subsequently, 4 mg eriodictyol was weighed and dissolved in 4 ml DMSO. The eriodictyol solution was slowly added dropwise to the freeze-dried HMPB-cisplatin powder while simultaneously subjecting the mixture to ultrasonic treatment. The mixture was again allowed to shake overnight at room temperature. Finally, centrifugation and washing were performed, and the supernatant was collected to calculate the drug loading capacity. For synthesis of CY5.5-HMPB@ERY@PT, 1 mg/mL HMPB@ERY@PT were dispersed in PBS (0.1 M, pH 8.5). After 1.0 mg/mL Cy5.5 was added, the reaction mixture was stirred at 4 °C for 24 h. Finally, the nanoparticles were centrifuged repeatedly to remove excess Cy5.5.

The particle size, zeta potential, and morphology of HMPB@ERY, HMPB@PT, and HMPB@ERY@PT were evaluated by dynamic light scattering (Nano-ZS90, Malvern, UK) and transmission electron microscope (TEM, JEM-2100, Japan). In-vitro drug release behavior was evaluated in PBS (pH 7.4) medium containing 1.0% Tween80 (w/v) using a dialysis membrane (MWCO 10–12 kDa) method. Eriodictyol quantification was performed by HPLC (1260 Infinity, Agilent Technologies, USA) and the ultraviolet detection for cisplatin quantification.

The endocytosis of NPs

For the investigation of the endocytosis of NPs, U2OS and MG63 OS cell slides incubated with CY5.5-HMPB@ERY@PT NPs 12 h for confocal fluorescence. Additionally, fluorescence signal intensities at different time points were captured using flow cytometry.

Mouse models

The animal experiments in this study obtained approval from the Animal Care and Ethics Committee of Xiangya Hospital of Central South University (Changsha, China). All operation in the animal experiments were conducted in accordance with the animal ethical standards of Xiangya Hospital (Approval number: 2024030282). In brief, 4-week-old female nude mice were injected with U2OS cells subcutaneously. After the subcutaneous tumor formation, mice were divided into different group containing five mice, and the corresponding treatments were applied for mice via tail vein injection (q.o.d., 8 times). Subsequently, the tumor volume of mice was measured every 5 days, and tumors were harvested after 30 days, with their weights measured. For the in-vivo fluorescence molecular imaging (FMI), U2OS tumor-bearing mice were randomly separated into two groups (n = 3 per group) and injected intravenously with 100 μL of 5mg/mL Cy5.5-HMPB@ERY@PT or Cy5.5. Fluorescence signal intensities at different time points were captured using an AniView Spectrum System (BLT Photon Technology, Guangzhou, China). After 48 h postinjection, mice were euthanized, with their tumors and major organs (hearts, livers, spleens, lungs, kidneys, and intestines) dissected and examined by in-vitro imaging.

Statistical analysis

GraphPad Prism 8 were used for data processing and statistical analysis of all the experimental results. The quantitative data were expressed as the mean ± SD and tested by student’s t-test was while the qualitative data expressed in frequency were analysed by Chi-square test or Fisher’s exact test. The indicated P value (* P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001) was considered statistically significant.

Results

The application of eriodictyol could suppress the proliferation but facilitate cell apoptosis of OS cells

The exploration of IC50 of 48 h eriodictyol treatments found that the IC50s of U2OS and MG63 cells against eriodictyol were 27.39 μM and 26.11 μM, respectively (Fig. 1A). Subsequently, the concentration of 25 μM was utilized to exploration the inhibition of eriodictyol on OS cells proliferation. In the CCK-8 experiment, we found that the growth of OS cells between eriodictyol-treated and DMSO-treated groups became more significantly different with the continuous increase of eriodictyol-treated time [U2OS (P_Day2 < 0.01, P_Day3 < 0.01, P_Day4 < 0.001); MG63 (P_Day2 < 0.05, P_Day3 < 0.01, P_Day4 < 0.001)] (Fig. 1B). The EDU assays shown that eriodictyol could reduce the proportion of proliferative OS cells and the proliferative OS cells between the eriodictyol-treated and the corresponding negative control groups were distinctly different (P < 0.001) (Fig. 1C). Additionally, colony formation assays also exhibited that eriodictyol could exert a significant inhibitory effect on OS cells proliferation (P < 0.0001) (Fig. 1D). Furthermore, the tunnel assay and flow cytometry were performed to investigate the effects of eriodictyol on OS cells apoptosis. After treating with 0, 6.25, 12.5, 25 μM eriodictyol, OS cells were undergone the tunnel and flow cytometry. The results of tunnel assays illustrated that the tunnel positive OS cells gradually increased with increasing concentration of eriodictyol, indicating that eriodictyol could facilitate OS cells apoptosis (Fig. 1E). Similar with the tunnel assays, the flow cytometry also demonstrated that eriodictyol could promote OS cells apoptosis, especially the late apoptosis (Fig. 1F). Therefore, eriodictyol may serve as a promising medication for OS treatment or adjuvant therapy.

Fig. 1.

Eriodictyol could effectively inhibit proliferation but promote apoptosis in OS cells. A The investigation of 48 h IC50 of eriodictyol against U2OS and MG63 cells. B CCK-8 assays of eriodictyol against U2OS and MG63 cells. C EDU assays of eriodictyol against U2OS and MG63 cells. D Colony formation of eriodictyol against U2OS and MG63 cells. E Tunnel assays of eriodictyol against U2OS and MG63 cells. (F) Flow apoptosis cytometry. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

The tumor-suppressing effects of eriodictyol may derived from its ferroptosis-inducing property

Since previous studies have indicated that eriodictyol could trigger tumor cells ferroptosis in various cancers, we utilized several cell death inhibitors, including ferroptosis, necroptosis, autophagy, pyroptosis, cuproptosis inhibitor to investigate the potential cell death type induced by eriodictyol. After extensive literature reading and review of manufacturer's instructions, 10 nM Ferrostatin-1, 20 nM Deferoxamine mesylate, 10 nM Necrostatin-1, 10 nM Autophinib, 10 nM NLRP3/AIM2-IN-3, and 10 nM Cuprizone were performed for the respective eriodictyol-treated groups, respectively. From the results, we found that the application of two classical ferroptosis inhibitors could significantly reverse eriodictyol-inducing OS cells death, especially in the content of high concentration of eriodictyol application (Fig. 2A, B). For further confirming this finding, we performed transcriptome sequencing and actually found ferroptosis related pathway was significantly enriched after the eriodictyol application, validating the occurrence of OS cells ferroptosis during the eriodictyol application (Fig. 2C, D). Subsequently, we performed a variety of ferroptosis-related assays to detect the occurrence of ferroptosis following by eriodictyol application. An increasing intracellular Fe(II) concentration was found after the concentration of eriodictyol was elevated (Fig. 2E). Additionally, the application of eriodictyol seems to decrease the reduced mediums in a dose-dependent manner (Fig. 2F–K). Conversely, the oxidative products, detected through DHE and liperfluo probes and MDA assay, were conspicuously increased (Fig. 2L–N). The occurrence of ferroptosis always accompanied with the damage of mitochondrion, which could be identified through MMP and TME detections. After treated with a gradually increasing concentration of eriodictyol and incubated with JC-1 probes, MMP was detected via a fluorescence microscope, exhibiting eriodictyol could damage OS cells mitochondrion in a dose-dependent manner (Fig. 2O). Additionally, the eriodictyol-induced microstructural damage of OS cells mitochondrion was identified through TEM, which unveiled that eriodictyol application could result in mitochondrial ridge fracture, shrinkage, and membrane rupture (Fig. 2P). Lastly, we validated the effects of eriodictyol on OS cells ferroptosis in in-vivo and the results illustrated that eriodictyol (5 mg/kg) exerted a wonderful inhibition on tumor growth. Meantime, ferroptosis inhibitors could effectively reverse the eriodictyol-induced ferroptosis, in which DFO exhibited a superior effect than Fer-1 (Fig. 2Q–S). Given the conspicuous ferroptosis-related phenotypes in both vitro and vivo assays, ferroptosis-inducing effect of eriodictyol may contribute to its tumor suppressing effects.

Fig. 2.

Eriodictyol exerted a wonderful ferroptosis inducing effects against OS cells. A, B Ferroptosis inhibitors reversed the cell viability inhibited by eriodictyol. C, D Ferroptosis-related pathway was enriched after the treatment of eriodictyol. E Eriodictyol could elevate the intracellular Fe(II) concentration. F, G Eriodictyol could inhibit the glutathione peroxidase activity in OS cells. H–K Eriodictyol conspicuously decreased the GSH and GSH/GSSG in OS cells. L–N Eriodictyol increased the intracellular oxidative level of OS cells. O, P Eriodictyol could result in OS cells mitochondrial damage. Q–S Eriodictyol could suppress OS growth in vivo and ferroptosis inhibitors could reverse the tumor-suppressing effects of eriodictyol. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

Eriodictyol could facilitate structural stability of BACH1, increasing its expression and eventually mediating the BACH1/GPX4 signaling pathway

Intriguingly, the DNA-binding transcription repressor activity was identified to enrich in the eriodictyol-treated groups, indicating eriodictyol application might activated the transcription-repressing activities (Fig. 3A, B). Since prior studies illustrated that NRF2/BACH1 served as the activator/repressor of ferroptosis-related genes, we assumed that eriodictyol could affect BACH1 expression to mediate OS cells ferroptosis. Then, we performed the molecular docking to investigate the bond between BACH1 protein and eriodictyol, and the result demonstrated that eriodictyol could well-bind with BACH1 proteins, with the binding energy of -6.4 kcal/mol (Fig. 3C–E). Subsequently, for the validation of molecular docking results, we further conducted SPR assay and the results exhibited that the affinity of recombinant human BACH1 with eriodictyol was 3.53 μM, which provided the solid experimental evidence that eriodictyol could well-bind with BACH1 proteins (Fig. 3F). Meantime, we found that BACH1 seem to exhibit a conspicuously decreased expression in the OS tissues compared with the adjacent normal tissues, indicating BACH1 might serve as a tumor suppressor in OS patients (Fig. 3G). To illustrate the effects of eriodictyol binding with BACH1 proteins on the BACH1 expression, we utilized western blotting to investigate the expressional alteration of BACH1 after treating with a gradually increasing concentration of eriodictyol and found eriodictyol could increase BACH1 expression in the dose-dependent manner (Fig. 3H). As a regulated cell death, ferroptosis can be precisely mediated and interacted with a series of ferroptosis related genes including GPX4. In our transcriptomic sequencing, GPX4 was identified as a downregulated gene in the transcriptional level (Fig. 4A), posing a hypothesis that BACH1 might serve as an upstream transcriptional factor of BACH1 to mediate its transcription and expression. For validating this assume, ChIP and dual-luciferase assays were performed and the results of ChIP assay demonstrated that BACH1 in the eriodictyol-treated groups exhibited higher bind with GPX4 genes than that of negative control groups (Fig. 4B, C). In additionally, dual-luciferase assays illustrated that BACH1 could inhibit the transcription of GPX4 gene and the eriodictyol application could facilitate this process, triggering a transcriptional-inhibitory effect on GPX4 expressional process (Fig. 4D–F). Ultimately, through western blotting, cell and tissue immunofluorescence staining, we identified eriodictyol could increase BACH1 expression but repress GPX4 expression (Fig. 4H–J).

Fig. 3.

Eriodictyol could bind with BACH1 proteins and elevate its expression. A, B RNA sequencing indicated that after the eriodictyol treatment transcriptional activities, especially DNA-binding transcription repressor activity, were activated. C–E Molecular docking exhibited that eriodictyol could excellently bind with BACH1 proteins. F SPR assay provided the experimental evidence regarding eriodictyol binding with BACH1 proteins. G Compared with the adjacent normal tissues, the decreased BACH1 expression was detected in OS tissues. H Eriodictyol could increase BACH1 expression in the dose-dependent manner

Fig. 4.

Eriodictyol-resulted BACH1 downregulation could lead to the decreasing expression of GPX4. A RNA sequencing found that eriodictyol could decrease GPX4 expression in transcriptional level. B, C ChIP-qPCR exhibited that BACH1 could bind with GPX4 gene. D–F Dual-luciferase assay indicated that eriodictyol could repress the transcription of GPX4 gene. G, H Western bloting and cell immunofluorescence staining diaplayed eriodictyol could increase BACH1 expression but inhibit GPX4 expression. I, J Tissue immunofluorescence staining illustrated eriodictyol could increase BACH1 expression but inhibit GPX4 expression. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

To further validate that the ferroptosis-inducing effects of eriodictyol were conducted via the increased BACH1 expression, we knocked down BACH1 expression in eriodictyol-treated OS cells. From the results, we found GPX4 expression was restored after BACH1 knockdown in eriodictyol-treated OS cells (Figure S1A, B). Meatime, BACH1 knockdown rescued the weaken OS cells viability resulted from eriodictyol treatment, with the reduced ROS level and the prothetic MMP, further confirming the key role of BACH1 in the ferroptosis-inducing effects of eriodictyol (Figure S1C-G). Therefore, eriodictyol might act as a structural stabilizer of BACH1 proteins, mediating the BACH1/GPX4 signaling pathway and eventually harnessing cells ferroptosis.

Eriodictyol could cooperate with cisplatin to exert a powerfully synergetic tumor-suppressing effect

Previous studies illustrated that ferroptosis could facilitate the sensitivity of cancer cells to conventional chemotherapeutic agents, especially cisplatin. To confirm the synergistic tumor suppression derived from the combination of eriodictyol and cisplatin, we conducted drug-synergistic assays for OS cells. First, IC50 of U2OS and MG63 cells against cisplatin were identified as 11.29 μM and 3.48 μM, respectively (Supplementary Figure S2A-B). Then, a series of drug combination containing a concentration gradient of eriodictyol and cisplatin were utilized for the investigation of the optimal combined concentration. As a result, 10 μM eriodictyol combining with 8 μM cisplatin was identified as the optimal combined concentration for U2OS cells while 15 μM eriodictyol combining with 3 μM cisplatin was considered as that of MG63 cells (Fig. 5A–F). Then, in the subsequent experiments, a series of groupings including DMSO, a concentration of 10 μM eriodictyol, a concentration of 8 μM cisplatin, and a combination of 10 μM eriodictyol and 8 μM cisplatin were utilized for U2OS. Meantime, a series of groupings including DMSO, a concentration of 15 μM eriodictyol, a concentration of 3 μM cisplatin, and a combination of 15 μM eriodictyol and 3 μM cisplatin were applied for MG63 cells. The cell proliferative assays exhibited that the single eriodictyol/cisplatin application and their combination could exert a time-dependently increasing inhibition on OS cells proliferation, in which the combination shown a superior than the single medication (Fig. 5G, H). Since our studies found that eriodictyol exhibited oxidation-inducing and reduction-exhausted ability in OS cells and prior studies highlighted that cisplatin could deplete the intracellular GSH to influence the cellular redox state, we investigated the alteration of the oxidative and reduced situation under the medication intervention. The detection of ROS illustrated that both of two agents possessed the oxidation-inducing and reduction-exhausted ability in OS cells and their combination obtained stronger oxidation–reduction regulatory ability (Fig. 5I–N). Furthermore, we explored MMP alteration after the single or combinative intervention and similar trend was observed, in which mono JC-1 symbolizing the damaged mitochondrion increased after the eriodictyol/cisplatin application and were further augmented under the combinative application (Fig. 5O). Taken altogether, based on the synergetic tumor-suppressing and redox regulated effect of the combination of eriodictyol and cisplatin, the application of eriodictyol might serve as a promising supplement in term of cisplatin-related OS treatments.

Fig. 5.

Eriodictyol exerted a synergistic promotion with cisplatin to inhibit OS cells growth. A–F Synergistic effects between eriodictyol and cisplatin. G, H The detection of the proliferative inhibition. I, J The investigation of the oxidative level. K–N The detection of GSH and GSH/GSSG level. O The investigation of mitochondrial membrane potential. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

Eriodictyol-cisplatin coated nanomedicine exerted a distinguished anti-tumor and ferroptosis-inducing effect

For effectively delivering medications as well as promoting therapeutic effects, we utilized HMPB NCs to load eriodictyol and cisplatin and formed a ferroptosis-synergistic nanocomplex to facilitate OS cells ferroptosis and cisplatin sensitivity. First, we prepared the HMPB particles and tested their structural characteristics. The results of TEM analysis shown that HMPB, HMPB@Ery, HMPB@PT, and HMPB@Ery@PT possessed the particle size of about 150–170 nm, with the wonderful dispersion and the conspicuous hollow and mesoporous structures (Fig. 6A). Meantime, DLS as well as Zeta assays were performed to assess whether the drugs were loaded successfully. From the test results, the hydrated particle size of naked HMPB was 200 nm, and the particle size increased after loading the drugs, which could prove the successful loading of the drugs (Fig. 6B). In addition, the naked HMPB potential was -10.8 mV, and the D potential changes after loading the drugs, which also confirmed the successful loading of the drugs (Fig. 6C). The drug loading capacity of eriodictyol and/or cisplatin were shown in Table S1 and Supplementary Figure S3–4 [HMPB@Ery@PT: HMPB/Ery/PT = 4 mg/2.37 mg/2.08 mg; HMPB@Ery: HMPB/Ery = 4 mg/1.92 mg; HMPB@PT: HMPB/PT = 4 mg/2.2 mg]. Both eriodictyol and cisplatin showed a sustained release pattern from the NCs within 48 h (Fig. 6D, E).

Fig. 6.

The NCs could conspicuously inhibit the proliferation but facilitate apoptosis of OS cells. A–C Characteristics of the NCs. D, E Drug release of the NCs. F The detection of the proliferative inhibition of the NCs. G, H The detection of the apoptotic promotion of the NCs. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

To explore the phagocytosis efficiency of NPs in OS cells, we performed confocal fluorescence and flow cytometry for U2OS and MG63 cells treated with CY5.5-HMPB@ERY@PT. The image of confocal fluorescence exhibited that the excellent phagocytosis of NPs in both U2OS and MG63 cells and the results of flow cytometry showed that a time-dependent phagocytic increase of NPs in OS cells, with a phagocytic peak in 7-h treatment (Supplementary Figure S5A, B). Therefore, our NPs could be well gobbled up by tumor cells. For exploring the biological effects of these NCs, we investigated the 48-h IC50 of HMPB@Ery@PT and found that the 48-h IC50 of HMPB@Ery@PT against U2OS and MG63 were 7.35 μg/ml and 1.45 μg/ml, respectively (Supplementary Figure S2C-D). Therefore, 7.35 μg/ml HMPB@Ery@PT for U2OS and 1.45 μg/ml for MG63 were used in the subsequent assays. Meantime, according to the drug loading capacity and release situation, 6.26 μg/ml HMPB@Ery, 5.07 μg/ml HMPB@PT, 5.83 μM cisplatin, 6.73 μM eriodictyol, 3.48 μg/ml HMPB were utilized for U2OS cells in the corresponding groups. Additionally, 1.24 μg/ml HMPB@Ery, 1.0 μg/ml HMPB@PT, 1.15 μM cisplatin, 1.33 μM eriodictyol, 0.69 μg/ml HMPB were treated for MG63 cells in the corresponding groups. Subsequently, we utilized EDU and flow apoptosis assays to explore the effects of different treatments on OS cells proliferation and apoptosis. Both the EDU and flow apoptosis assays showed that compared with the NC groups, HMPB-treated, eriodictyol-treated, and cisplatin-treated groups exhibited a significant inhibition on OS cells proliferation and apoptosis. Additionally, after loaded in the HMPB NCs, eriodictyol and cisplatin possessed a more excellent proliferative inhibition and apoptotic promotion. More notably, HMPB@ERY@PT displayed the most conspicuous proliferative inhibition and apoptotic promotion among all treated groups, suggesting its strong tumor suppressing effects (Fig. 6F–H). Taken altogether, the NCs could exert superior tumor-suppressing effects against OS cells, providing an alternative or/and supplementary method for OS treatments.

Since eriodictyol, cisplatin, and HMPB could facilitate ferroptosis, we detected that ferroptosis-related phenotype after the application of various treatment. The Fe(II) probe indicated that compared with the NC groups, the intracellular Fe(II) concentration of other groups were increased, and the HMPB, HMPB@ERY, HMPB@PT, and HMPB@ERY@PT increased more conspicuous than other group (Fig. 7A). Subsequently, we investigated that the cellular ROS level and found that all treatment effectively elevated the intracellular oxidative level and the oxidative facilitation of HMPB@ERY@PT was most remarkable among all groups (Fig. 7B). Contrary with the trend of ROS level, the GSH and GSH/GSSG exhibited a decreasing level, and the NCs showed more distinct reduction-exhaustion (Fig. 7C–F). In the MMP detection, poly JC-1 conspicuously decreased while mono JC-1 displayed an opposite trend after treating with the various agents and the NCs, indicating the structural damage of OS cells (Fig. 7G). Lastly, we focused on the structural alteration of mitochondrion between ERY, HMPB@ERY, and HMPB@ERY@PT groups. The results of TEM illustrated the mild mitochondrial damage in ERY groups, the moderate mitochondrial damage in HMPB@ERY groups, and the severe mitochondrial damage in HMPB@ERY@PT groups (Fig. 7H, I), indicating the success and therapeutic potential of both drug synergism and the NCs.

Fig. 7.

The NCs exerted a wonderful ferroptosis inducing effects against OS cells. A The NCs could obviously elevate the intracellular Fe(II) concentration. B The NCs increased the intracellular oxidative level of OS cells. C–F The NCs conspicuously decreased the GSH and GSH/GSSG in OS cells. G The NCs could result in OS cells mitochondrial damage. H The transmission electron microscopic detection of eriodictyol-treated, HMPB@ERY-treated, and HMPB@ERY@PT-treated OS cells. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

Eriodictyol-cisplatin coated nanomedicine exhibited a promising in-vivo tumor-suppressing effect as well as safety

First, to explore the in-vivo biodistribution and tumor targeting effect of NPs, mice were injected with our CY5.5-HMPB@ERY@PT or CY5.5 through a rat tail vein, with fluorescence signal intensities captured at different time points. From the results, we found that compared with CY5.5 molecules, our NPs possessed a more excellent in-vivo biodistribution and tumor targeting effect, which exhibited a better tumor accumulation and persistence (Supplementary Figure S5C). Notably, our NPs showed a strong accumulation in tumor site but little distribution in other major organs (Supplementary Figure S5D). These evidence supported the viability of our NPs in term of the in-vivo biodistribution and tumor targeting effect of NPs. For validating the effectiveness and safety of the NCs in vivo, we constructed the subcutaneous tumor model and performed the treatments in the corresponding group through the tail vein injection. Similar with the in-vitro experiments, according to the drug loading capacity and release situation, 5mg/kg HMPB@ERY@PT, 4.26mg/kg HMPB@ERY, 3.45mg/kg HMPB@PT, 1.20mg/kg PT, 1.33mg/kg eriodictyol, and 2.37mg/kg HMPB were applied for the corresponding groups, with the application of saline as the negative control group. The tumor growth were monitored throughout the study and the results illustrated that compared with the NC group, the tumor volume in other groups were obviously decreased (Fig. 8A–C). Here, we focused on the eriodictyol, HMPB@ERY, and HMPB@ERY@PT, and found the conspicuous difference in both tumor volume and weight, in which HMPB@ERY@PT exhibited an outstanding tumor inhibition (Fig. 8A–C). Meantime, we utilized immunofluorescent staining to investigate the proliferative and apoptotic situation in various groups, and a significant difference was observed between NC group and other groups, with the most proliferative inhibition and apoptotic promotion in the HMPB@ERY@PT group (Fig. 8D). Lastly, we also investigated the in-vivo safety of these agents and NCs in the current concentration. Blood chemistry was also performed and the results of blood routine examination and blood chemistry showed that there was no difference among all groups under the current concentration in the conventional index, which confirmed the safety of our NPs (Supplementary Figures S6–7). Meantime, through performing the H&E staining on the kidney, heart, liver, spleen, and lung, we did not discover the obvious injury of these organs, indicating the NCs could exert an exciting tumor-suppressing effect within the safe dose (Fig. 8E). Therefore, our NCs might serve as a viable therapeutic strategy for OS treatments in the future.

Fig. 8.

The NCs showed a superior tumor-suppressing effect in vivo. A–C Compared with the NC group, the tumor volume in other groups were obviously decreased, in which HMPB@ERY@PT exhibited an outstanding tumor inhibition. D Tissue section exhibited that the proliferative and apoptotic difference between NC group and other groups, with the most proliferative inhibition and apoptotic promotion in the HMPB@ERY@PT group. E H&E staining of various important organs did not discover the obvious injury of these organs. The data represent the mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 indicates a significant difference between the indicated groups

Discussion

Since the standardized surgical treatments are widespread adopted, the survival improvements of OS patients mostly depend on the sensitivity of their malignant cells towards the chemotherapeutic medications [1, 3]. However, increasing insensitivity or resistance of OS cells against the conventional chemotherapies have been observed in the recent decades, impeding the therapeutic improvement of OS treatment [1, 4, 35]. What is worse, large-dose application of multi-chemotherapeutics undoubtedly results in several adverse effects for OS patients, such as myelosuppression, immunologic derangement and even death related events [4, 36]. Therefore, the unsatisfactory therapeutic outcome and severe adverse effects pose an urgency to exploit to the alternative and/or supplementary therapies for the optimization of OS chemotherapeutic regimen [35, 37]. Herein, we affirmed the tumor-suppressing properties of eriodictyol and illustrated that its antitumor effects might ascribe to the ferroptosis-inducing activity, in which eriodictyol could bind with BACH1 to repress the transcription and translation of GPX4 and result in the GPX4-related ferroptosis. Further investigation found that eriodictyol could exhibit a synergistic effect with cisplatin, facilitating the antitumor effects of cisplatin. Ultimately, a ferroptosis-synergistic HMPB nanocomplex, enveloping eriodictyol and cisplatin, was formed to facilitate OS cells ferroptosis and cisplatin sensitivity, and the in-vitro and in-vivo experiments confirmed its excellent tumor-suppressing activities and in-vivo safety.

In recent years, considerable investigations have identified bioactive compounds extracted from natural herbal plants as the antitumor medication, of which mechanisms involve in the classical oncogenic pathways and various kinds of cell death [38–43]. Similarly, accumulating evidence have found numerous natural herbal extracts could exhibit inspiring anti-OS effects via the intervention of canonical signaling pathways including p53 pathway and the induced cancer cell regulated death such as ferroptosis, posing the prospect of natural compound for OS treatments [44–47]. In the present study, we confirmed the antitumor effects of eriodictyol against OS and the RNA sequencing exhibited that the ferroptosis related pathway was enriched after the application of eriodictyol. Notably, previous studies have found that eriodictyol played an important role in the mediation of ferroptosis [48–50]. For example, Wang et al. found that eriodictyol could regulate ovarian cancer cells ferroptosis, mitochondrial dysfunction, and cell viability to suppress the in-vitro and in-vivo tumor progression through Nrf2/HO-1/NQO1 signaling pathway [48]. However, several researches identified eriodictyol and its analogue as a kind of ferroptosis inhibitors and found that eriodictyol and its analogue might activate NRF2 and possess a powerful anti-inflammation to relieve the inflammatory diseases [49, 50]. For clarifying the role of eriodictyol in OS ferroptosis, a variety of in-vitro and in-vivo assays was conducted and the results confirmed that the treatment of eriodictyol could result in OS cells ferroptosis. Long-term clinical and experimental practice has proved that ferroptosis, as a natural anticancer mechanism, plays multiple roles in the occurrence and development of cancer, such as regulating tumor cell proliferation, migration, chemotherapy sensitivity and affecting tumor microenvironment immune response, and targeting cancer cells ferroptosis can serve as a potential target for cancer treatment [13, 51, 52]. Similarly, increasing studies have demonstrated that ferroptosis plays a critical role in OS occurrence and progression, and the breakthrough of OS treatments, particularly the treatment of drug-resistant OS patients, may depend on the intervention of ferroptosis [53–55]. Therefore, eriodictyol may be a promising alternative or supplementary medication for OS treatments.

Since ferroptosis was referred in 2012, the deepening mechanical exploration has been constantly conducted and a series of ferroptosis mediators have been identified, facilitating the understanding and investigation of this regulated cell death [13, 51]. GPX4, as a glutathione peroxidase that transform the peroxy bonds of lipid peroxides into hydroxyl groups during ferroptosis, has been utilized as a target for cancer ferroptosis [16, 56]. Notably, the decreased transcriptional and translational expressions of GPX4 were detected after the treatment of eriodictyol, indicating eriodictyol might affect the transcription of GPX4 to cause the GPX4-related ferroptosis in OS cells. Prior studies have illustrated that ferroptosis is precisely mediated by two oxidative stress-reactive transcription factors, NRF2 and BACH1, in which NRF2 inhibits ferroptosis while BACH1 plays an opposite role [52, 57, 58]. Meantime, considerable studies have demonstrated that BACH1 could regulate the transcription and translation of ferroptosis related elements including GPX4 to mediate the occurrence of ferroptosis [59]. Intriguingly, through the RNA sequencing, SPR, and molecular docking, the present study found that eriodictyol could bind with BACH1 proteins, and the subsequent assays using ChIP, dual-luciferase and western blotting assays illustrated that eriodictyol elevated BACH1 expression could repress the transcription and translation of GPX4 to trigger OS cells ferroptosis, complementing the underlying mechanism of the anti-OS of eriodictyol. Existing evidence has suggesting that GPX4 plays an critical role in OS cells ferroptosis and targeting GPX4 exhibits an excellent ferroptosis-inducing as well as tumor-suppressing effects in OS cells [16, 56], hinting the therapeutic potential of eriodictyol for OS treatments.

Notably, increasing researches have demonstrated that ferroptosis could facilitate the sensitivity of cancer cells, especially stem-like tumor cells with the drug-resistance, towards the conventional therapeutic regimen, especially platinum chemotherapeutics [16, 56]. Similarly, increasing studies have also exhibited that triggering cancer cells ferroptosis could reverse the chemoresistance of OS cells towards the conventional chemotherapeutic agents [60–64]. Intriguingly, the recent reports have indicated that the formation of the Pt-GS complex may be a novel anti-cancer mechanism of cisplatin, which results in GSH depletion and facilitate the occurrence of ferroptosis [27–29]. The studies conducted by He et al. also suggested that ferroptosis could deplete the intracellular GSH to overcome OS cells chemoresistance [65]. Therefore, the combination of cisplatin and ferroptosis inducers may be a potential strategy to enhance OS platinum-related chemotherapeutic effects. For validating this hypothesis, we performed drug combination experiment and found that eriodictyol could perform a synergistic effect with cisplatin, in which the combination of these two agents exhibited a more significant suppressive effects on OS cells proliferation, elevated the intracellular oxidative levels, and consumed more intracellular GSH. Ferroptosis, as a cell death characterized by the excessive production of lipid peroxide, can be reversed through the reduced agents, principally GSH, the processes of which was executed by GPX4 proteins [16, 29]. As was shown in our study, eriodictyol could inhibit the expression of GPX4 via GPX4 to intervene the reduction of lipid peroxide, and the combination of eriodictyol of eriodictyol and cisplatin exhibited a more complete induction on OS cells ferroptosis, offering a novel synergetic strategy for OS chemotherapeutic regime.

Recent decades witnessed the rapid development and tremendous successes in nanopharmacy in the researches and treatments in various cancers, including OS [54, 62, 63, 66]. PB nanomaterial is a safe, FDA-approved antidote for thallium poisoning and has been widely used for drug carriers due to its metal–organic framework that can be used for drug loading and fluorescent dye conjugation [67, 68]. Recent studies have shown that PB molecules can induce ferroptosis for its ability to deliver iron ions into OS cells and its direct catalytic oxidation of unsaturated lipids by its mixed valence transition metal (Fe2+/Fe3+) [30]. For effectively delivering medications as well as promoting therapeutic effects, we utilized HMPB NCs to load eriodictyol and cisplatin and formed a ferroptosis-synergistic nanocomplex to facilitate OS cells ferroptosis and cisplatin sensitivity. In this ferroptosis-synergistic nanocomplex, HMPB shell could initiate OS cells ferroptosis via direct catalytic oxidation of unsaturated lipids as well as exogenous iron delivery, and both eriodictyol and cisplatin could weaken the protection of the reduced mediums through GSH exhaustion and GPX4 transcriptional inhibition, synergistically promoting the occurrence and progression of ferroptosis. Indeed, we observed the excellent tumor-suppressing effects in OS cells in both vitro and vivo. The in-vitro experiment exhibited that the nanocomplex exerted a superior ferroptosis-inducing and tumor-suppressing effects than other treatments, in which the OS cells mitochondrion were obviously destroyed accompanied with the conspicuous elevation of intracellular Fe(II) concentration as well as oxidative level. Meantime, the in-vivo assays illustrated that the nanocomplex provided a wonderful inhibition in term of tumor growth and the conspicuous apoptosis of OS cells was captured through the tunnel assays. More importantly, no obvious injury was detected in the organs of mice after the treatments of this ferroptosis-synergistic nanocomplex, indicating its in-vivo safety to some extent. Therefore, our ferroptosis-synergistic nanocomplex may represent a promising alternative therapeutic strategy for OS patients.

Author contributions

Wei Luo, Yusheng Li, and Zili Lin selected the topic and revised the manuscript, Zili Lin, Ziyi Wu, Qing Liu, and Xiangya Li conducted the experiments, wrote the manuscript and made figure as well as table. All authors reviewed the manuscript.

Funding

This work was supported by Hunan Provincial Health High-Level Talent Scientific Research Project (No. R2023175), the Provincial Science and Technology Innovation Plan Project of Hunan (No. 2023SK2023), the Provincial Science Foundation of Hunan (No. 2022JJ30075).

Availability of data and materials

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Xiangya Hospital of Central South University.

Competing interests

The authors declare no competing interests.