Manganese Biomineralized Ferritin Nanoplatforms: A New Approach for Ovarian Cancer

Introduction Ovarian cancer ranks as the third most common malignancy of the female reproductive system but has the highest mortality rate among these tumors.1 Due to its deep pelvic location, early symptoms such as abdominal distension and indigestion are easily confused with other conditions, and currently, there is a lack of highly sensitive and specific screening methods.2 Consequently, over 70% of cases are diagnosed at an advanced stage (III/IV), with a five-year survival rate of only 30–40%.3 Although developed countries achieve slightly higher survival rates (~48%), overall prognosis remains poor.4 Standard treatment involves surgical resection combined with platinum-based chemotherapy.5 However, the development of drug resistance leads to recurrence in approximately 70% of cases within three years.6,7 Systemic intravenous administration remains the primary chemotherapy delivery method, often resulting in severe adverse reactions including immune responses and organ toxicity. Therefore, developing targeted delivery systems tailored to ovarian cancer characteristics to enhance tumor-specific drug accumulation while minimizing systemic toxicity is crucial.8,9 Such approaches, combined with multimodal therapy, are essential to improve treatment outcomes and address current therapeutic limitations. Ovarian cancer exhibits high metabolic activity to meet its demands for nutrients and energy required for malignant proliferation. Cancer cells overexpress transferrin receptors (TfR1, CD71) to enhance iron uptake, while the expression of iron export proteins such as ferroportin is reduced, leading to intracellular iron overload.10,11 This accumulation results in excessive reactive oxygen species (ROS) generation, inducing DNA damage and epigenetic alterations that promote tumor initiation and progression.12,13 Ferritin (Fn), a natural iron-binding protein, targets the CD71 receptor on ovarian cancer cells.14 As a nanocarrier, it utilizes the enhanced permeability and retention (EPR) effect to enhance tumor-specific drug delivery.15 However, its targeting specificity is limited as normal cells (eg, in the liver) also express transferrin receptors, causing non-specific accumulation.16,17 To overcome this, biomimetic-mineralized Fn has been developed to shield off-target binding and enhance systemic stability. The “shield” then dissolves within the tumor microenvironment characterized by mild acidity and redox imbalance promoting more efficient uptake by tumor cells.18,19 Manganese ions (Mn2+) possess unique coordination properties that enable the formation of stable complexes with proteins, polymers, or liposomes. This characteristic has been exploited to develop multifunctional nanocarriers through biomimetic mineralization.20,21 These manganese-based systems can respond to tumor microenvironmental stimuli—such as acidic pH and elevated glutathione levels—for intelligent drug release, facilitating multimodal therapy approaches including chemotherapy, photothermal therapy, and immunotherapy. In this study, we mimic natural mineralization processes by utilizing Mn2+ to biomineralize Fn nanocages, resulting in manganese-mineralized ferritin (MFn). MFn disassembles under weakly acidic conditions, releasing Mn2+ and the encapsulated drugs. Additionally, within the tumor microenvironment, Fenton-like reactions generate hydroxyl radicals, which synergistically enhance anti-tumor effects.22,23 The outer shell of MFn acts as a barrier to reduce off-target accumulation—particularly in organs like the liver—and enhances tumor-specific accumulation and delivery efficiency.24 This multimodal, synergistic therapeutic strategy markedly improves anti-tumor efficacy. Dihydroartemisinin (DHA), the primary active metabolite of artemisinin, can form peroxide bridges with intracellular Fe2+ in tumor cells, generating ROS that directly disrupt cell membranes, induce lipid peroxidation, and trigger ferroptosis.25 Additionally, DHA downregulates the expression of GPX4 and SLC7A11, decreasing GSH levels and weakening cancer cell resistance to lipid peroxidation. It also disturbs mitochondrial iron homeostasis, amplifying Fenton reactions, increasing ROS production, and accelerating ferroptosis.26,27 Clinically, DHA has been combined with sorafenib to optimize therapeutic efficacy of liver cancer treatment.28 Recent studies indicate that DHA combined with cisplatin reduces chemotherapeutic dosage and toxicity while enhancing efficacy through dual mechanisms of apoptosis and ferroptosis.29–31 Ferroptosis-based combination therapy, leveraging multiple mechanisms and precise delivery techniques, positions DHA as a promising anticancer agent due to its unique ferroptosis-inducing capacity, low toxicity, and broad-spectrum activity.32 It has the potential to become a key component of combination therapies for ovarian cancer, demonstrating a favorable clinical outlook. Building on the analysis of clinical ovarian cancer samples and the conceptual framework outlined above, we developed a biomimetic mineralization strategy to encapsulate Fn within a “temporary protective shell” composed of biocompatible MFn, loaded with DHA (resulting in DHA@MFn), as illustrated in Scheme 1. This design effectively shields Fn from nonspecific uptake by the liver, thereby prolonging circulation time and significantly enhancing tumor-specific accumulation. Within the tumor microenvironment, DHA@MFn promotes photothermal ablation of tumor cells while simultaneously releasing Mn2+ ions that catalyze Fenton-like reactions to produce hydroxyl radicals, amplifying oxidative stress.33,34 Concurrently, DHA consumption depletes glutathione (GSH), inducing ferroptosis and synergistically enhancing chemodynamic therapy (CDT). This multifaceted approach results in robust antitumor effects through combined oxidative damage, ferroptosis induction, and thermal ablation.35 This innovative nanoplatform offers a promising solution to the longstanding challenge of off-target organ accumulation, particularly in the liver, while achieving enhanced tumor targeting via biomimetic design. Its capacity to integrate multiple therapeutic modalities within a single construct makes it particularly effective against tumor heterogeneity and resistance, offering a comprehensive anticancer strategy. Currently, most ferritin-based nanomedicines remain in the preclinical research stage. Ferritin’s inherent tumor-targeting properties, excellent biocompatibility, and flexible drug-loading capacity hold significant potential for clinical translation. Material and Methods Materials DHA and MnCl2 were purchased from Aladdin Biochemical Technology Co., Ltd (Shanghai, China). Mccoy’s 5A and trypsin were obtained from HyClone (Logan, Utah, USA). Phosphate buffered saline (PBS) and fetal bovine serum (FBS) and were acquired from Gezhe Biotechnology Co.,Ltd (Anhui, China). Reactive oxygen species (ROS) detection kit, Annexin V-FITC/PI Apoptosis assay kit, Mitochondrial membrane potential assay kit with JC-1, Lipid Peroxidation Assay Kit with BODIPY 581/591 C11 were purchased from Biyuntian Biotechnology Co., Ltd (Shanghai, China). SLC7A11 antibody was purchased from Huabio (Hangzhou, China). GPX4 antibody was purchased from Servicebio (Wuhan, China). All other chemical reagents were obtained from Sinopharm Chemical Reagent Corporation (Shanghai, China). Cell Lines and Animals Ovarian cancer cell lines SKOV3 (RRID: CVCL_0032) was acquired from MeisenCTCC Cell Technology Co., Ltd (Zhejiang, China) and was cultured in McCoy’s 5A supplemented with penicillin/streptomycin (1%, v/v) and 10% non-heat-inactivated FBS. All cells were cultured in a humidified atmosphere containing 5% CO2 at a temperature of 37°C. BALB/c mice (female, 4–6 w, 18–22 g) were acquired from SLAC Laboratory Animal Co., Ltd (Shanghai, China) and housed under specific pathogen-free (SPF) conditions in a controlled environment. All animals were kept in standard laboratory cages with a 12 h light/dark cycle and maintained at a constant temperature (22 ± 3°C) and humidity (50 ± 10%). They had access to sterilized rodent chow and filtered water ad libitum. Regular monitoring ensured their health and well-being throughout the study. All animal experiments were performed in compliance with the Guidelines for the Care and Use Ethics Committee of Fujian Medical University (Certificate number: IACUC FJMU 2024–0317). Human Samples Ovarian tumor tissue samples were obtained from Fujian Medical University Union Hospital. Twenty ovarian cancer tissue specimens were obtained from patients undergoing gynecological surgery at the Department of Obstetrics and Gynecology, Union Hospital Affiliated to Fujian Medical University. Following pathological examination of the resected tumor tissue, approximately 300 cubic millimeters of fresh cancer tissue was isolated and reserved for modeling in this experiment. After tissue fixation, immunofluorescence staining for TfR was performed. All specimens were collected with written informed consent and in accordance with a protocol approved by the Institutional Review Board of Fujian Medical University Union Hospital (Certificate number: 2023WSJK006). This study was conducted in compliance with the principles of the Declaration of Helsinki. Preparation and Characterization of DHA@MFn Blank Fn, composed exclusively of heavy-chain subunits, was expressed in Escherichia coli and purified following established recombinant protein purification protocols.36 A total of 4 mg of Fn was dissolved in 4 mL of PBS buffer, and the pH was gradually adjusted to 12.0 under gentle stirring at room temperature over 2 h to reach equilibrium. Subsequently, 400 μL of DHA solution (5 mg/mL) and 25 μL of MnCl2 solution (0.1 M) were added to the mixture, which was stirred at room temperature overnight to facilitate biomineralization and drug loading. After the biomineralization and drug loading processes, the DHA@MFn was isolated through ultrafiltration using a 30 kDa molecular weight cutoff centrifugal filter at 4000 rpm for 20 minutes. The resulting product was then washed three times with deionized water to remove residual impurities, yielding a purified DHA@MFn suitable for subsequent analyses. The morphology of blank Fn and DHA@MFn was observed using transmission electron microscopy (TEM, Thermo Fisher, USA), and their elemental compositions were confirmed via energy-dispersive X-ray spectroscopy (EDS) coupled with TEM. Particle size, polydispersity index (PDI), and zeta potential of both blank Fn and DHA@MFn were analyzed using dynamic light scattering (DLS, Anton Paar, USA). The encapsulation efficiency, drug loading capacity, and release behavior of DHA@MFn under different pH conditions (pH 5.5 and pH 7.4) were evaluated using UV-Vis spectroscopy to quantify DHA content. To assess the stability of the biomineralized DHA@MFn, samples stored in PBS at pH 7.4 at 4°C were monitored for changes in particle size and DHA loading over one week (n=3). To elucidate the drug release profiles of DHA@MFn at pH 5.5 and pH 7.4, 10 mg of DHA@MFn were dispersed in 20 mL of PBS at the respective pH. The suspension was stirred at 37°C, and at predetermined time points (0.5, 1, 2, 4, 8, 12, 24, 48, and 72 h), 1 mL samples were extracted for analysis (n=3). The amount of DHA released was quantified using UV-Vis spectroscopy (Figure S1). Photothermal Efficiency Analysis of DHA@MFn After biomimetic mineralization of blank Fn with manganese ions (Mn2+), the resulting MFn demonstrated improved stability and enhanced NIR absorption capacity, leading to superior photothermal properties.37 To evaluate the temperature changes at different concentrations (CMFn=0, 100, 200, 300 and 400 μg/mL), which were exposure to NIR laser for 5 min at a power density of 1.5 W/cm2, with temperature measurements taken every 0.5 min. To assess the effect of laser power on DHA@MFn, 200 μL of MFn solutions (CMFn=200 μg/mL) were irradiated with NIR laser at three different power densities (0.5, 1, 1.5, and 2 W/cm2), with temperature changes recorded every 0.5 min. For evaluating photothermal stability, 200 μL of DHA@MFn solution (CMFn= 200 μg/mL) was transferred to a 96-well plate. The laser power was set to 1.5 W/cm2, and a temperature probe was placed into the solution to record temperature changes every 0.5 min over time. After 7.5 min of irradiation, the laser was turned off, allowing the solution to cool for 7.5 min. This cycle was repeated four times to assess stability. Cellular Internalization and Intracellular ROS Detection of DHA@MFn Ovarian tumor tissues exhibit high expression levels of TfR1. As a natural carrier targeting these receptors, Fn demonstrates notable tumor-targeting specificity compared to other natural proteins. The cellular internalization of Fn and MFn was evaluated using SKOV3 cells. To facilitate tracking, AF488 was encapsulated within Fn and MFn, resulting in AF488-loaded Fn and AF488-loaded MFn (AF488@Fn and AF488@MFn). SKOV3 cells in the logarithmic growth phase were inoculated uniformly into 6-well plates. The cells were then incubated with AF488@Fn and AF488@MFn for 3, 6, and 9 h, respectively. After incubation, cells were harvested, and the intracellular fluorescence intensity of AF488@Fn and AF488@MFn was quantified via flow cytometry (FACS) (n=3). To further elucidate subcellular localization following endocytosis, SKOV3 cells in the logarithmic growth phase were evenly distributed into 6-well plates at a pH 6.5 to simulate the tumor environment. The cells were treated with Cy3@MFn for 2 and 4 h. Subsequently, nuclear staining with DAPI was performed, and the distribution of Cy3@MFn fluorescence was observed using confocal laser scanning microscopy (CLSM). Intracellular ROS levels were measured using the fluorescent probe DCFH-DA. SKOV3 cells were treated with various formulations, including PBS, DHA, MFn, DHA@MFn, DHA combined with NIR irradiation (DHA+NIR), MFn combined with NIR (MFn+NIR), and DHA@MFn with NIR (DHA@MFn+NIR), each at a DHA equivalent concentration of 50 μg/mL, for 4 h. Subsequently, cells were stained with DCFH-DA for 30 min. After washing twice with PBS, ROS levels were assessed through CLSM and quantified by FACS (n=3). Synergistic Antitumor Mechanism of DHA@MFn The mitochondrial membrane potential (Δψm) in SKOV3 cells was evaluated using the JC-1 assay kit. Specifically, SKOV3 cells were seeded in Petri dishes at a density of 1 × 105 cells per dish. The cells were then incubated with various treatments: serum-free Mccoy’s 5A medium, DHA, MFn, DHA@MFn, DHA combined with NIR irradiation (DHA+NIR), MFn combined with NIR (MFn+NIR), and DHA@MFn combined with NIR (DHA@MFn+NIR). After 3 h of incubation, the cells were subjected to NIR laser irradiation (1.5 W/cm2 for 5 minutes). Subsequently, JC-1 working solution was added to the dishes and the cells were stained at 37°C for 20 minutes in the dark. Following washing steps, the cells were analyzed using confocal laser scanning microscopy (CLSM) and quantified by flow cytometry (FACS) (n=3). Following intervention with DHA@MFn, the alterations in intracellular lipid peroxidation and ferroptosis-related protein expressions were examined to assess the compound’s ability to induce ferroptosis in ovarian cancer cells. Ferroptosis is characterized by the excessive accumulation of ROS-induced lipid peroxides and the upregulation of key ferroptosis markers, which can alter cell membrane fluidity and permeability, ultimately disrupting cellular structure and function, leading to programmed cell death of ovarian cancer cells.38 In this study, we employed the Lipid Peroxidation Assay Kit with BODIPY 581/591 C11 to monitor changes in lipid peroxidation levels within ovarian cancer cells. Logarithmic-phase SKOV3 cells were treated with various treatments: serum-free McCoy’s 5A medium, DHA, MFn, DHA@MFn, DHA+NIR, MFn+NIR, and DHA@MFn+NIR (10 μg/mL DHA equivalent). After treatment, cells were stained with the lipid peroxidation sensor, BODIPY 581/591 C11. Following two washes with PBS, lipid peroxidation levels were assessed via CLSM and quantified using FACS (n=3). Additionally, total cellular proteins were extracted from treated cells for Western blot analysis to evaluate the expression levels of ferroptosis markers, including GPX4 and SLC7A11 (n=3). To avoid ineffective separation of bands caused by similar protein molecular weights, we selected different reference proteins for detecting different target proteins. For detecting the expression levels of GPX4 protein (22 kDa), we employed β-actin (42 kDa) as the reference protein. For detecting the expression levels of SLC7A11 protein (55 kDa), we used GADPH (37 kDa) as the reference protein. These proteins play critical roles in the regulation of ferroptosis—with GPX4 acting as a key enzyme that detoxifies lipid peroxides, and SLC7A11 regulating cystine intake essential for glutathione synthesis, thereby influencing the cellular redox status and ferroptosis sensitivity. Cytotoxicity Analysis of DHA@MFn The cytotoxicity of various treatment groups on SKOV3 cells was evaluated using the CCK-8 assay. Logarithmic-phase SKOV3 cells were seeded at 4,000 cells per well in 96-well plates. After incubation with different concentrations of serum-free Mccoy’s 5A medium, DHA, MFn, DHA@MFn, and combinations involving NIR irradiation, cells were subjected to irradiation (808 nm, 1.5 W/cm2 for 5 min) after 6 h of incubation with treatments containing NIR. Following a further 24 h culture, 100 μL of CCK-8 reagent was added to each well and incubated for 1.5 h. Absorbance at 450 nm was measured to determine cell viability (n=5). For apoptosis assessment, logarithmic-phase SKOV3 cells at 2 × 105 cells per well were seeded into 6-well plates. After 6 h of treatment with DHA, MFn, or DHA@MFn (15 μg/mL DHA), NIR-irradiated groups were exposed under the same conditions. Cells and supernatants were collected following 24 h, stained with Annexin V-FITC and PI for 20 minutes, and apoptosis levels were analyzed by FACS (n=3). In Vivo Biodistribution and Photothermal Evaluation of DHA@MFn To investigate the biodistribution and tumor-targeting capability of Fn and MFn, a SKOV3 xenograft tumor model was established via subcutaneous injection of BALB/c nude mice (female, 4–6 w, 18–22 g). Once tumor volumes reached approximately 200 mm3, mice were randomly assigned to three groups: Cy5.5, Cy5.5@Fn, and Cy5.5@MFn. The respective formulations were administered intravenously through the tail vein (10 mg/kg Fn equivalent, 100 μL). Fluorescent imaging was performed at 6, 12, 24, and 36 h post-injection to monitor the in vivo accumulation of labeled drug delivery system (n=3). The in vivo photothermal effects of Fn and MFn were evaluated using the SKOV3 tumor-bearing model. Mice were divided into three groups: control, Fn, and MFn (10 mg/kg Fn equivalent, 100 μL). At 24 h following intravenous administration, mice were anesthetized with 2% pentobarbital sodium and subjected to near-infrared laser irradiation (808 nm, 1.5 W/cm2, for 5 min). Real-time temperature changes at the tumor site were monitored using a thermal imaging camera (n=3). In Vivo Antitumor Efficacy Study The SKOV3 xenograft tumor model was established via subcutaneous injection of 1×106 SKOV3 cells into the right dorsal flank of female BALB/c mice. Once tumor volumes reached approximately 100 mm3, mice were randomly divided into six groups: PBS, DHA, MFn, DHA@MFn, MFn with irradiation and DHA@MFn with irradiation (3 mg/kg DHA, 100 μL, n=6). The corresponding treatments were administered via tail vein injection on days 1, 4, 7, 11, and 13, and then NIR laser irradiation (808 nm, 1.5 W/cm2 for 5 minutes) was performed 24 h post-injection in the irradiation groups. During the study, tumor and body weights were measured and recorded on days 3, 6, 9, 12, and 15. At the end of the treatment period (day 15), the mice were euthanized, and tumor tissues were harvested, photographed, and weighed. Major organs, including the heart, liver, spleen, lungs, kidneys, and tumors, were collected to evaluate tumor suppression efficacy and organ health by assessing changes in size and morphology. Additionally, blood samples from each group were analyzed biochemically to evaluate systemic organ functions and the biocompatibility of the treatment systems. Histopathological examinations were performed to identify potential toxic effects. To elucidate the anti-tumor mechanism and stromal depletion capability, tumor sections underwent immunohistochemical staining for Ki67, CD34, and α-SMA, and immunofluorescence staining for GPX4 and SLC7A11 to assess ferroptosis-related pathways, proliferation, vascularization, and stromal modulation. Statistical Analysis All experiments were performed at least three times. The normality of all continuous data was assessed using the Shapiro–Wilk test. Data are presented as mean ± standard deviation (SD) for normally distributed variables and as median (interquartile range) for non-normally distributed variables. Comparisons between the two groups were performed using the independent samples t-test for normally distributed data and the Mann–Whitney U-test for non-normally distributed data. Data analysis and graphing were conducted using GraphPad Prism 6 (San Diego, USA). Statistical significance is indicated as: *p < 0.05, **p < 0.01, and ***p < 0.001. Results and Discussion Fn Receptor Expression in Tumors Natural ferritin (Fn) is a cage-like protein with a central cavity that plays a crucial role in iron metabolism, maintaining iron homeostasis and cellular antioxidant defense. Ovarian cancer cells increasingly overexpress TfR1 to fulfill their increased iron demands for rapid proliferation and metabolic activity. Owing to its biocompatibility and innate tumor-targeting ability, ferritin shows promise as a delivery vector. To evaluate its potential for ovarian cancer therapy, our team collected clinical tissue samples and monitored TfR1 expression in tumor tissues. Our team analyzed clinical ovarian cancer tissues, and immunohistochemistry revealed high TfR1 expression, particularly on tumor cells (Figure 1A), indicating efficient ferritin uptake by cancer cells and supporting its potential as a targeted delivery system. Preparation and Characterization of DHA@MFn The expression level of TfR1 is generally low in normal cells. However, the liver, as the primary organ responsible for systemic iron homeostasis, facilitates non-specific accumulation of ferritin when used as a tumor delivery carrier, which compromises targeting efficiency. The design of biomimetic-mineralized Fn enables it to mask non-specific accumulation by transferrin receptors within the circulatory system, thereby enhancing its circulation stability, and then the “shield” dissolves under tumor-specific microenvironments characterized by mild acidity and redox imbalances promoting improved uptake by tumor cells. Inspired by this, we prepared a manganese-mineralized ferritin (Mn-Fn) drug delivery system via mineralization of Fn with manganese ions, and the specific design of DHA@MFn is illustrated in Scheme 1A. We initially expressed the human heavy-chain subunit in Escherichia coli and successfully purified self-assembled Fn from the lysates. Subsequently, Mn ion biomimetic mineralization was performed on Fn, and DHA was loaded to obtain DHA@MFn. As shown in Figure 1B and C, the blank Fn exhibited a uniform dispersion under TEM. After manganese mineralization, MFn displayed a high-contrast metal shell encapsulating the Fn core. To confirm the successful construction of the DHA@MFn nanoplatform, elemental mapping analysis (Figure 1D) demonstrated the presence of elements corresponding to DHA (O), the Mn shell (Mn), and the ferritin core (Fe, O).39 Dynamic light scattering (DLS) measurements confirmed that DHA@MFn had a monodisperse size distribution with an average diameter of approximately 12.2 nm, slightly larger than native ferritin (Figure 1E). Zeta potential analysis further verified the formation of the Mn mineralized shell, indicating successful fabrication and good stability (Figure 1F). The drug loading capacity of DHA@MFn was evaluated via UV-Vis spectroscopy, revealing a loading efficiency of 5.84 ± 0.05%, comparable to that of Fn loaded with small-molecule drugs. Additionally, the stability of the nanoparticles was assessed based on particle size and drug loading capacity (n=3). The results indicated that the particle size remained stable at around 14 nm without significant aggregation or sedimentation under physiological conditions, suggesting favorable stability and resistance to degradation (Figure S2). Fn’s cage-like architecture exhibits pH-dependent disassembly and reassembly, endowing it with intrinsic pH-sensitive drug release properties. This pH responsiveness arises from the loss of salt bridges and hydrogen bonds between protein dimers under highly acidic or alkaline conditions, leading to reversible depolymerization and reorganization of the protein subunits.40 The release profile of DHA from DHA@MFn was evaluated under different pH conditions using a dialysis experiment. The results showed that at pH 5.5 and 7.4, the cumulative release of DHA increased from approximately 10% to 80% over 72 h (n=3). The pH-dependent release behavior is advantageous for cancer therapy, as it minimizes drug release in the bloodstream and ensures sufficient drug delivery to tumor tissues, thereby promoting effective tumor cell killing after internalization of DHA@MFn (Figure 1G). Photothermal Efficiency Analysis of DHA@MFn Previous studies have indicated that Fn mineralized with tetravalent manganese exhibits a characteristic absorption peak near near-infrared region II (NIR-II), suggesting its potential for efficient photothermal conversion under 808 nm laser irradiation. The temperature elevation of DHA@MFn solutions correlates positively with both protein concentration and laser irradiation power. As shown in Figure 1H and I, when DHA@MFn solutions at a fixed concentration (CMFn = 200 μg/mL) were exposed to lasers of varying intensities within the range of 0.5–2 W/cm2, the temperature increased progressively with increasing laser power. Specifically, after 5 min of irradiation at 1.5 W/cm2 with an 808 nm laser, the temperature of DHA@MFn (CMFn = 200 μg/mL) rapidly rose by approximately 17°C, demonstrating its excellent photothermal conversion capability. Furthermore, when DHA@MFn suspensions at concentrations of 0, 100, 200, 300, and 400 μg/mL (all with CMFn) were irradiated at 1.5 W/cm2, the temperature increased steeply during the initial phase, reaching a plateau after approximately 5 min. These findings suggest that at a concentration of 200 μg/mL and an irradiation power of 1.5 W/cm2, DHA@MFn can achieve effective photothermal therapy. Additionally, the temperature response during four repeated heating-cooling cycles—achieved by alternating the laser on and off—remained consistent, indicating that the MFn-based drug delivery system possesses robust photothermal stability (Figure 1J). Cellular Internalization and Intracellular ROS Detection of DHA@MFn Typically, free chemotherapeutic drugs administered via intravenous systemic injection lack tumor-specific targeting capabilities, which often results in elevated systemic toxicity, side effects, and diminished efficacy in tumor cell eradication. Fn-based carriers can selectively accumulate in tumor cells by mediating transcytosis through the overexpressed TfR1 on ovarian cancer cells, thereby enhancing recognition and cytotoxicity toward cancerous tissues. However, the impact of biomimetic mineralization of Fn into MFn on its tumor-targeting enrichment potential remains to be thoroughly investigated, an aspect addressed in this study. Initially, we employed flow cytometry to compare cellular uptake of AF488-labeled ferritin (AF488@Fn) and manganese-mineralized ferritin (AF488@MFn) after incubation with SKOV3 cells for 3, 6, and 9 h (n=3). As shown in Figure 2A, within the 9 h observation period, there was no significant difference in the uptake of Fn versus MFn by the tumor cells, indicating that manganese mineralization and drug loading do not impair Fm’s intrinsic cellular internalization ability. To further evaluate the intracellular localization of Cy3-labeled MFn (Cy3@MFn), confocal fluorescence microscopy was performed after incubation at pH 6.5 for 2 and 4 h—mimicking the acidic tumor microenvironment. As depicted in Figure 2B, Cy3 fluorescence signals were significantly enriched within the cells, demonstrating effective internalization. Given that tumor cells generate higher levels of reactive oxygen species (ROS) due to their heightened metabolic activity, they are especially susceptible to intracellular ROS accumulation. DHA@MFn facilitates the delivery of the ferroptosis-inducing agent DHA and promotes iron overload through ferritin loading, which induces lipid peroxidation within tumor cells. Moreover, the MFn with iron and manganese ions in the tumor’s acidic and redox-perturbed microenvironment triggers Fenton-like reactions, generating substantial quantities of ROS to promote tumor cell apoptosis and ferroptosis. Using the DCFH-DA fluorescent probe, we detected intracellular ROS levels in drug-treated groups, with green fluorescence indicating ROS presence (n=3). Results showed that the DHA@MFn group exhibited the highest fluorescence intensity—corresponding to the most elevated ROS levels—compared with free DHA and the pure ferritin carrier. Notably, upon irradiation with an 808 nm laser, the fluorescence intensity in the DHA + NIR group showed no significant increase relative to DHA alone, suggesting DHA itself lacks notable photothermal properties. Conversely, the MFn + NIR group displayed a slight increase in fluorescence intensity, indicating some photothermal effect. Importantly, the ROS level in the DHA@MFn + NIR group was significantly elevated, signifying an enhanced induction of intracellular oxidative stress (Figure 2C and D). These findings demonstrate that DHA@MFn can effectively elevate ROS levels within tumor cells, supporting the induction of ferroptosis and emphasizing its potential for synergistic tumor therapy. Synergistic Antitumor Mechanism of DHA@MFn Ferroptosis is an iron-dependent form of programmed cell death characterized by lipid peroxidation and closely associated with mitochondrial dysfunction. Studies have demonstrated that lipid peroxidation—the core mechanism underlying ferroptosis—disrupts the integrity of the mitochondrial inner membrane, leading to dissipation of the proton gradient and depolarization of the mitochondrial membrane potential. Excessive production of ROS within cells can directly impair components of the electron transport chain, further diminishing mitochondrial membrane potential. To evaluate these changes, we employed the mitochondrial membrane potential probe JC-1 in SKOV3 cells treated with various formulations (n=3). JC-1 aggregates within mitochondria when the membrane potential is high, emitting red fluorescence, whereas at low membrane potential, it exists in monomeric form, emitting green fluorescence. As shown in Figure 2E, flow cytometric analysis revealed that the proportion of JC-1 monomers, indicative of mitochondrial depolarization, increased significantly after 24 hours of treatment across multiple groups: control (5.86%), MFn (13.4%), MFn + NIR (27.3%), DHA (36.5%), DHA + NIR (36.8%), DHA@MFn (58.7%), and DHA@MFn + NIR (82.7%). The most substantial reduction in mitochondrial membrane potential was observed in the DHA@MFn + NIR group. Further, inverted fluorescence microscopy demonstrated that this group exhibited the strongest green fluorescence, consistent with flow cytometry results, indicating pronounced mitochondrial depolarization. The decline in mitochondrial membrane potential represents an early marker of apoptosis, and these JC-1 results corroborate the flow cytometry data related to apoptosis (Figure 2F). Extensive lipid peroxidation is a hallmark downstream event in ferroptosis. Iron plays a pivotal role in this process by catalyzing the Fenton reaction, which produces ROS that attack polyunsaturated fatty acids (PUFAs) in cellular membranes, initiating lipid peroxidation cascades. In addition, inhibition of glutathione peroxidase 4 (GPX4)—a key selenoprotein responsible for detoxifying lipid hydroperoxides—leads to the accumulation of lipid peroxides, resulting in membrane destruction and cell death. To assess lipid peroxidation levels, we utilized BODIPY 581/591 C11, a fluorescent probe sensitive to lipid ROS (n=3). The oxidation of BODIPY shifts its fluorescence from red to green, enabling ratiometric analysis. Flow cytometry analysis (Figure 3A and B) demonstrated that after 24 hours of treatment, the groups receiving DHA@MFn and DHA@MFn + NIR exhibited significantly increased fluorescence intensities, indicating elevated lipid peroxidation (Figure S3). Fluorescence microscopy further supported these findings, showing the lowest red/green fluorescence ratio in these groups, which corresponds to maximal lipid peroxidation and ferroptosis induction. To elucidate the molecular mechanisms underlying ferroptosis, the expression levels of SLC7A11 and GPX4—two key regulators of ferroptosis—were examined via Western blot analysis. SLC7A11 mediates cystine import for glutathione synthesis, while GPX4 reduces lipid hydroperoxides, thereby preventing ferroptosis.41 A decrease in these protein levels signifies progression towards ferroptotic cell death. The results, shown in Figure 3C, indicated that the DHA@MFn + NIR group significantly suppressed the expression of both SLC7A11 and GPX4 compared to the control, confirming its capacity to induce ferroptosis in tumor cells (Figures S4 and S5). Cytotoxicity Analysis of DHA@MFn The cytotoxic effects of DHA, MFn, DHA@MFn with and without laser irradiation were evaluated in SKOV3 cells using the CCK-8 assay (n=5). As shown in Figure 3D, MFn exhibited a high cell viability even at elevated concentrations, demonstrating its excellent biocompatibility. Conversely, in the MFn + NIR group, cell survival decreased with increasing MFn concentration, indicating that MFn-based photothermal therapy can effectively induce tumor cell death in a dose-dependent manner.42 Both DHA and DHA + NIR groups showed a concentration-dependent decrease in cell viability; however, there was no significant difference in their IC50 values, suggesting that DHA alone does not possess inherent photothermal properties. Notably, the cytotoxicity of DHA@MFn combined with NIR irradiation was significantly higher than that of DHA@MFn alone, with substantially lower IC50 values compared to all other groups. To further investigate the role of oxidative stress in treatment-induced apoptosis, apoptosis levels were measured using flow cytometry and Annexin V-FITC/PI staining. The experimental setup included a control group and six treatment groups: MFn, DHA, DHA@MFn, MFn + NIR, DHA + NIR, and DHA@MFn + NIR (n=3). In the absence of light, the percentage of viable cells in the MFn group was approximately 90%, comparable to the control, confirming the biocompatibility of MFn. The proportion of early and late apoptotic cells in the free DHA group was 15%, whereas in the DHA@MFn group, apoptosis increased to 37.8%, indicating a synergistic effect of DHA and MFn in promoting cell death. The MFn + NIR group showed an apoptotic rate of about 13.8%, higher than MFn alone, consistent with its photothermal therapeutic effect. In contrast, the DHA + NIR group showed no significant increase in apoptosis, confirming that DHA lacks photothermal properties and that infrared irradiation does not appreciably enhance its effect. Importantly, the DHA@MFn + NIR group exhibited a significant increase in apoptotic rate to 53.86%, compared to 37.8% in the DHA@MFn group without light (Figure 3E). This suggests that MFn acts as an effective carrier to enhance cell killing under NIR irradiation through combined photothermal and ferroptotic mechanisms. Collectively, these results demonstrate that photothermal therapy exerts a highly effective tumoricidal effect. The synergistic induction of ferroptosis by MFn and DHA, mediated by increased ROS production within cells, further enhances apoptosis, providing a comprehensive approach to tumor eradication. In Vivo Biodistribution and Photothermal Evaluation of DHA@MFn The biodistribution and tumor-targeting capacity of the nanocarriers were further evaluated using SKOV3 tumor-bearing mouse models. Three experimental groups—free Cy5.5, Cy5.5@Fn, and Cy5.5@MFn—were administered via tail vein injection, and in vivo fluorescence imaging revealed a time-dependent accumulation within tumor tissues in all groups (n=3). Notably, both Cy5.5@Fn and Cy5.5@MFn exhibited higher fluorescence intensity and prolonged retention in tumors compared to free Cy5.5, which is consistent with its rapid hepatic metabolism and short systemic circulation half-life (Figure 4A). Within the observed time frame, Cy5.5@MFn demonstrated superior enrichment and retention in tumor lesions, observable up to 48 h post-injection. Conversely, Cy5.5@Fn showed a higher non-targeted accumulation in the liver, indicating that MFn through an insulating layer effectively minimizes off-target uptake in organs with high iron metabolic activity (Figure 4B). These findings suggest that the biomimetic mineralization endows MFn with enhanced tumor-specific targeting, likely owing to its adaptive response within the tumor microenvironment. Furthermore, the photothermal performance of MFn was assessed in vivo via infrared thermal imaging. Following systemic administration, temperature changes in PBS, Fn, and MFn groups were monitored to evaluate photothermal efficacy (n=3). Laser irradiation conducted at approximately 6 h post-injection—corresponding to peak tumor accumulation—demonstrated minimal temperature increases in the PBS and Fn groups, with only slight rises after 5 min of laser exposure. In contrast, the MFn group showed a significant, time-dependent temperature elevation. When administered at equivalent protein concentrations, MFn consistently attained higher tumor temperatures than Fn. Specifically, after 5 min of laser irradiation, the tumor temperature in the MFn group rose to approximately 50°C, satisfying the thermal threshold for effective photothermal therapy (Figure 4C and D). These in vivo results corroborate the biodistribution data and further validate that the biomimetic mineralization combined with surface shielding strategy effectively minimizes off-target accumulation, thereby improving tumor specificity and therapeutic efficacy. In Vivo Antitumor Efficacy Study This study demonstrates the promising therapeutic potential of DHA@MFn, a biomimetic manganese-mineralized Fn-based nanoplatform, for ovarian cancer therapy, combining ferroptosis induction and photothermal treatment. Building upon its excellent in vitro performance, in vivo experiments were conducted wherein DHA, MFn, and DHA@MFn were intravenously administered to SKOV3 tumor-bearing mice. Blood biochemical tests conducted on mice post-administration confirmed the biosafety of the manganese-mineralized ferritin platform, with no significant adverse effects observed, enabling the progression to pharmacodynamic studies (Figure 4E). Tumors were established by subcutaneously inoculating 1×106 SKOV3 cells into the upper left thigh. Once tumors reached approximately 100 mm3, the mice were randomized into six groups and received injections of PBS, DHA, MFn, or DHA@MFn. In designated groups, NIR irradiation at 1.5 W/cm2 for 5 min was performed 24 h post-injection (Figure 5A) (n=6). Tumor volume analysis revealed significant suppression, particularly in groups receiving combination therapy. The free DHA demonstrated a tumor inhibition rate of 44.91%, while DHA@MFn showed a slightly higher inhibition rate of 62.82%. The DHA@MFn treatment group supplemented with near-infrared light irradiation exhibited the most significant tumor suppression efficacy, reaching 83.14% (Figure 5B). Tumor weight measurements corroborated these findings, supporting the effective inhibition of tumor growth and reduction of tumor cell viability by DHA@MFn with or without NIR (Figure 5C). Over the two-week treatment period, all groups exhibited stable body weights, indicating minimal systemic toxicity (Figure 5D). Mechanistically, the biomimetic ferritin system mediated ferroptosis through iron overload, demonstrating measurable anti-tumor effects. Notably, DHA-induced ferroptosis and the combined DHA@MFn treatment significantly outperformed MFn alone, highlighting that DHA enhances ferroptosis via more effective lipid peroxidation. The synergistic chemo-photothermal therapy further markedly elevated antitumor efficacy, with the DHA@MFn plus NIR group achieving near-complete tumor remission, including the disappearance of tumors in two mice (Figure 5E–G). Immunohistochemical analyses of tumor tissues displayed decreased proliferation marker Ki67, reduced microvessel density (CD34), and diminished fibroblast populations (α-SMA), indicating that the combined therapy impairs neovascularization and stromal support, thereby restraining tumor progression (Figure 5H). Post-treatment assessments of organ histology confirmed the biosafety of the Mn-mineralized ferritin platform, with no apparent adverse effects (Figure 6A). Moreover, immunofluorescent staining of ferroptosis-related proteins GPX4 and SLC7A11 revealed a significant decrease in ferroptosis markers in DHA-treated and DHA-loaded groups, with the lowest levels observed in the biomimetic mineralized DHA@MFn group, underscoring a synergistic ferroptosis induction (Figure 6B). This work provides a comprehensive and innovative therapeutic strategy for ovarian cancer by integrating biomimetic mineralization, ferroptosis induction, and photothermal therapy within a biocompatible nanoplatform. The results demonstrate that DHA@MFn can effectively induce ferroptosis while enabling synergistic tumor ablation via photothermal effects, with minimal systemic toxicity. The approach modulates the tumor microenvironment by impairing neovascularization and reducing stromal cell populations, thereby enhancing overall therapeutic efficacy. These findings hold promise for developing targeted, minimally invasive, and combination nanotherapies that could overcome resistance mechanisms and improve clinical outcomes in ovarian and other cancers. Conclusions In this study, we developed a biomimetic manganese-mineralized Fn-based nanoplatform (DHA@MFn) capable of synergistically inducing ferroptosis and exerting photothermal effects for ovarian cancer therapy. The in vitro assessments demonstrated that DHA@MFn effectively triggered ferroptosis and showed potent photothermal capabilities. In vivo experiments confirmed that DHA@MFn achieved enhanced tumor accumulation with minimal off-target distribution, owing to its biomimetic design and surface shielding strategy. Systemic administration combined with NIR irradiation resulted in significant tumor growth suppression, with some cases of complete tumor eradication. Immunohistochemical analyses revealed decreased tumor proliferation, neovascularization, and stromal components, while ferroptosis markers such as GPX4 and SLC7A11 levels were markedly elevated, confirming ferroptosis induction. Importantly, the Mn-mineralized ferritin platform exhibited excellent biosafety, with no observable systemic toxicity. The results collectively indicate that DHA@MFn, through combined ferroptosis induction and photothermal therapy, provides a promising, minimally invasive approach for ovarian cancer treatment, effectively modulating the tumor microenvironment and overcoming resistance mechanisms. Although the nanomedicine we developed demonstrated promising antitumor activity and safety in a subcutaneous xenograft model, the current research cycle primarily evaluated tumor suppression rates and short-term safety. Its long-term toxicity, resistance development, and ultimate impact on survival remain unclear. Furthermore, while this model facilitates operational monitoring, it fails to replicate the complex biological behavior of ovarian cancer growth and metastasis within the peritoneal cavity, limiting its predictive value for clinical efficacy. Subsequent studies require models more closely aligned with clinical settings. Furthermore, while laboratory-scale nanoparticle drug preparation methods demonstrate good reproducibility, scaling up to clinical-grade production still faces challenges in sterility control, stability, and quality assurance. In summary, our biomimetic manganese-mineralized ferritin platform offers a promising multi-modal strategy for ovarian cancer therapy, with potential for further refinement to facilitate clinical translation. Continued research addressing the aforementioned challenges will be vital to realize its full therapeutic potential.