Cantharidin-manganese based cocktail nanoplatform Co-activating ferroptosis and STING for enhanced HCC immunotherapy Authors: Shengmei Wang, Mengtian Ma, Qiuyan Guo, Guiming Deng, Xinhua Xia, Jianyu Li, Xueqing Tao, Dong Zhong, Kun Zhang Published: 2026 ABSTRACT Hepatocellular carcinoma (HCC) is often identified during its advanced phases, where therapeutic choices are constrained and patient prognosis is generally poor, creating a critical need for innovative treatment approaches. Despite the promise of immunotherapy, its efficacy is frequently constrained by the immunosuppressive tumor microenvironment (TME). To address this, we developed a multifunctional nanoplatform (CTD/MM@BSA) via a facile “one-pot” method, co-delivering the chemotherapeutic agent cantharidin (CTD) and manganese ions (Mn2+) using bovine serum albumin (BSA). This platform leverages the synergistic interplay between ferroptosis induction and cGAS-STING pathway activation to remodel the TME. Upon internalization and subsequent release within tumor cells, CTD and Mn2+ act synergistically to boost intracellular ROS levels. This elevation promotes the induction of ferroptosis and ICD. The process culminates in the emission of DAMPs, which in turn activate the maturation of dendritic cells. Simultaneously, DNA damage triggered by CTD acts synergistically with the heightened sensitivity of cGAS to DNA in the presence of Mn2+, leading to a marked amplification of cGAS-STING pathway activation and the subsequent production of type I interferons. This bidirectional ferroptosis-immunity activation cascade effectively reverses immunosuppression, promoting cytotoxic T-cell responses. It markedly suppresses tumor growth and metastasis while extending the survival of mice. Moreover, when combined with an anti-PD-L1 antibody (aPD-L1), the therapy demonstrates a stronger synergistic anti-tumor effect. This study overcomes the limitations of free CTD (hepatorenal toxicity) and Mn2+ (poor tumor accumulation), offering a potent and targeted strategy to remodel the TME and boost anti-tumor immunity for HCC therapy. Introduction Hepatocellular carcinoma (HCC), which represents 75-90% of primary liver cancers, is a globally prevalent malignancy with persistently rising incidence. The absence of distinct early clinical signs and a lack of sensitive biomarkers for early detection mean that the majority of cases are identified only after the disease has progressed to an advanced phase. Consequently, the five-year survival rate for these patients remains under 20% [1,2]. Current treatment options for advanced HCC demonstrate only modest effectiveness and are frequently associated with problems such as drug resistance and undesirable side effects [3]. Consequently, there is a critical need to devise innovative therapeutic approaches that can address these limitations and enhance patient outcomes in HCC. Immunotherapy aims to strengthen the host's anti-tumor immune responses, offering potential for both tumor eradication and recurrence prevention [4,5]. However, its clinical effectiveness is often constrained by the immunosuppressive tumor microenvironment (TME), characterized by limited immune cell infiltration [6]. Remodeling the TME to enhance anti-tumor immunity remains a key research focus. The cGAS-STING pathway plays a central role in innate immunity by sensing cytosolic DNA and triggering the production of type I interferons (IFN-I) and pro-inflammatory cytokines, thereby promoting anti-tumor immune responses [[7], [8], [9], [10], [11]]. Moreover, research indicates that ferroptosis eliminates tumor cells by promoting an accumulation of lipid peroxides, thereby triggering immunogenic cell death (ICD). This process leads to the emission of damage-associated molecular patterns (DAMPs) [12]. These molecules function as “danger signals”, promoting the maturation of dendritic cells (DCs) and activating cytotoxic T cell responses, which in turn enhances anti-tumor immune defenses [[13], [14], [15]]. It is noteworthy that a strong synergistic relationship exists between ferroptosis and the cGAS-STING pathway in the context of anti-tumor immunity. Activation of the cGAS-STING pathway results in the production of IFN-I, which are instrumental in activating T cells. Consequently, these activated CD8+ T cells release substantial amounts of IFN-γ. This IFN-γ subsequently promotes the onset of ferroptosis through the inhibition of SLC7A11/SLC3A2 [[16], [17], [18], [19]]. This interplay underscores the therapeutic potential of simultaneously targeting cell death pathways and innate immunity to overcome immune tolerance. Despite their promise, STING agonists face clinical limitations, including degradation of natural cyclic dinucleotides, poor cellular uptake, and risk of systemic immune activation [20]. Thus, developing tumor-targeted STING agonists with potent immune-activating capability is crucial for improved efficacy. Manganese ions (Mn2+), have emerged as an effective strategy to enhance cGAS sensitivity and amplify STING activation [21,22]. In addition, Mn2+ in the TME can promote Fenton-like reactions, generating hydroxyl radicals (·OH) that trigger lipid peroxidation and ferroptosis [21,23]. However, solely relying on Mn2+-mediated cGAS-STING activation is insufficient to fully trigger tumor cell death [24]. Combined targeting of multiple nodes within this pathway is therefore key to enhancing efficacy. Traditional chemotherapy drugs can enhance the responsiveness of the cGAS-STING pathway to Mn2+ by inducing DNA damage in tumor cells [25]. Cantharidin (CTD), a bioactive compound derived from the traditional Chinese medicine blister beetle, offers a novel opportunity for this synergy. This agent demonstrates a wide range of anti-cancer effects, showing notable efficacy against various solid malignancies, including HCC, gastric cancer, and lung cancer [[26], [27], [28]]. Its advantage lies in inhibiting cancer cell proliferation, metastasis, and angiogenesis while avoiding the myelosuppression and immune system inhibition common with conventional chemotherapeutics [26]. According to studies, the anti-tumor activity of CTD is mediated through the generation of DNA damage and the suppression of proteins involved in DNA repair [29]. Based on our preliminary finding that CTD induces ROS production, we hypothesize that CTD complements Mn2+-mediated cGAS-STING pathway activation. It is expected that this dual strategy will not only improve the eradication of tumor cells but also transform the immune microenvironment, thereby fostering long-lasting immunity against tumors. However, CTD's strong dose-dependent hepatorenal toxicity severely limits its clinical application [30], and free Mn2+ struggles to achieve effective therapeutic concentrations due to insufficient tumor accumulation. To overcome these challenges, nanodelivery system capable of co-loading CTD and Mn2+ presents an ideal solution. Recent progress in nanomedicine has demonstrated that nanomaterials can deliver drugs directly to tumor sites, boosting accumulation in cancer cells while reducing toxicity to healthy tissues [31]. Furthermore, such systems allow for the codelivery of synergistic agents and enable controlled release, which is crucial for modulating complex biological pathways [32]. Consequently, constructing a delivery system based on CTD and Mn2+ synergy, enabling precise modulation of the bidirectional ferroptosis-immunity activation pathway to remodel the TME, represents a vital research direction to overcome current HCC treatment barriers. Inspired by these pioneering insights, we developed a multifunctional nanoplatform (CTD/MM@BSA) (Scheme 1A) using a convenient “one-pot” method. This approach efficiently encapsulates CTD within the hydrophobic domain of BSA, while simultaneously utilizing BSA's thiol groups for specific coordination with Mn2+, alongside the interaction between 2-methylimidazole (2-MI) and Mn2+. This “one-pot” method offers advantages of simplicity and high drug loading capacity. Following cellular internalization, CTD/MM@BSA decomposes under acidic conditions releasing CTD and Mn2+. Subsequently, the synergistic activity of CTD and Mn2+ markedly increases the levels of reactive oxygen species (ROS) inside tumor cells, leading to ferroptosis. This ferroptosis then initiates ICD and facilitates the emission of DAMPs. Concurrently, the dual effects of CTD-induced DNA damage and Mn2+-enhanced cGAS sensitivity to DNA synergistically amplify the activation of the cGAS-STING signaling pathway (Scheme 1B). Through these coordinated actions, a potent anti-tumor immune response is generated, which effectively suppresses both tumor growth and metastatic spread. Moreover, the anti-tumor efficacy is further enhanced synergistically when this nanoplatform is administered in combination with an anti-PD-L1 antibody (aPD-L1).Scheme 1(A) Illustrates the fabrication procedure for the CTD/MM@BSA nanoparticles. (B) Depicts the underlying mechanisms through which the nanodrug exerts its anti-tumor effects. (A) Illustrates the fabrication procedure for the CTD/MM@BSA nanoparticles. (B) Depicts the underlying mechanisms through which the nanodrug exerts its anti-tumor effects. Materials and methods Materials CTD was from Hunan Xuan Chuang Biological Company Phospholipids (Changsha, China). The Annexin V-FITC/PI double staining kit and ECL chemiluminescence detection kit were sourced from BestBio (Shanghai, China) and Biosharp (Hefei, China), respectively. Ferrostatin-1 (Fer-1) was from MACKLIN (Shanghai, China). Antibodies against GAPDH, SLC7A11/xCT, STING, TBK1, IRF3, and calreticulin were purchased from Proteintech Group Inc (Wuhan, China). Phospho-STING (P-STING), Phospho-IRF3 (P-IRF3), and Phospho-TBK1 (P-TBK1) antibodies were from Cell Signaling Technology (Boston, USA). CD4 and CD8 antibodies were obtained from Affinity (China). Mouse ELISA kits for TNF-α, IFN-γ, IFN-β, IL-10, and TGF-β were supplied by Jianglai (Wuhan, China). The following flow cytometry antibodies were purchased from Elabscience (Wuhan, China): PE-conjugated anti-mouse MHCII, APC-conjugated anti-mouse CD11c, PE-conjugated anti-mouse CD86, APC-conjugated anti-mouse CD4, FITC-conjugated anti-mouse CD3, PE-conjugated anti-mouse CD8a, and purified anti-mouse CD16/32. Animals The C57BL/6 mice, male SD rats, and nude mice used in this study were provided by Hunan Slaik Jingda Laboratory Animal Co., Ltd. All animal experiments were strictly conducted in accordance with the protocols approved by the Experimental Animal Ethics Committee of the First Affiliated Hospital of Hunan University of Chinese Medicine (No. 202503026) and the Experimental Animal Ethics Committee of Hunan Academy of Chinese Medicine (Sy2025-0116 and Sy2026-0020). Synergistic effect of CTD and Mn2+ To investigate the respective roles of CTD and Mn, cell viability was initially assessed using a CCK-8 assay. Intracellular ROS levels and cell death were then evaluated. Further analyses included the examination of DNA damage and ferroptosis-related markers through flow cytometry, western blotting, and a GSH detection kit. In a separate experiment, Hepa1-6 cells were exposed to increasing concentrations of CTD, Mn2+, or their combination and cell viability was measured by CCK-8. The combined effect of CTD and Mn2+ was analyzed using CompuSyn software to calculate the combination index (CI). Preparation of CTD/MM@BSA Add 200 μL of 2-MI solution dropwise to the BSA solution. Then, 100 μL of CTD solution was introduced, and stirring was maintained for 1 h. Next, 200 μL of MnCl2 solution was added, and the mixture was stirred at 37 °C for another 2 h. The final product, CTD/MM@BSA, was obtained by centrifuging the solution at 16000 rpm for 10 min to remove unreacted BSA and inorganic ions. To optimize the nanoparticle synthesis, key parameters were varied systematically: BSA concentration (0.45, 0.91, 1.82, 4.55, 7.27 mg/mL), 2-MI (4.55, 9.09, 18.18, 27.27, 36.36 mM), MnCl2 (9.09, 18.18, 27.27, 36.36 mM), and CTD (50, 100, 200, 300, 400 μg/mL). The particle size and polydispersity index (PDI) were measured for each condition to determine the optimal formulation. Characterization of CTD/MM@BSA The size and zeta potential of the nanoparticles (NPs) were determined through dynamic light scattering (DLS). The morphology and elemental composition of CTD/MM@BSA were characterized using Transmission Electron Microscopy (TEM) and X-ray Photoelectron Spectroscopy (XPS). The drug-loading capacity (LD%) and encapsulation efficiency (EE%) of CTD and Mn2+ were analyzed separately using high-performance liquid chromatography (HPLC) and inductively coupled plasma optical emission spectrometry (ICP-OES). The stability of CTD/MM@BSA nanoparticles was evaluated by incubating them in water and 10% FBS for 24 h respectively. Storage stability was evaluated by keeping the nanoparticles at 4 °C and measuring the particle size and polydispersity index (PDI) after 1, 3, 6, and 9 days. The in vitro release profile was studied using a dialysis method, and the concentration of Mn2+ was quantitatively determined by ICP-OES to calculate the cumulative release amount. Experimental verification of the formation mechanism of CTD/MM@BSA To verify the interaction between CTD and BSA, CTD/MM@BSA was added to a dialysis bag with a molecular weight cut-off (MWCO) of 3400. The dialysis bag was then immersed in dialysis solutions containing NaCl, Urea, or sodium dodecyl sulfate (SDS). After 8 h of incubation, the dialysis solutions were collected, freeze-dried, and subjected to methanol extraction. The released CTD content was determined by HPLC. To verify the coordination between Mn2+ and the sulfhydryl groups in BSA, CTD/MM@BSA was incubated with NaCl, urea, and glutathione (GSH) for 8 h. Subsequently, the nanoparticles were separated by centrifugation, and the BSA content in the supernatant was determined. To verify the coordination between Mn2+ and 2-MI, 2-MI and Mn2+ were mixed with stirring, followed by the addition of ethylenediaminetetraacetic acid (EDTA). After standing for 1 h, the phenomenon was observed. Cellular uptake Hepa1-6 cells were seeded in 6-well plates (2 × 105 cells/well). The cells were then exposed to different treatments (Control, FITC, and CTD/MM@BSA-FITC) for 4 h. After the incubation, the cells were washed three times with PBS to remove the drugs and dyes, and then the cells were collected for quantitative analysis using flow cytometry. In parallel, cellular internalization of the nanoparticles was directly visualized by confocal laser scanning microscopy (CLSM). Furthermore, nanoparticle uptake by BMDCs, RAW 264.7 cells, and other cell types was evaluated using flow cytometry. Lysosomal escape of CTD/MM@BSA Hepa1-6 cells were plated in confocal dishes (5 × 104 cells/dish). The cells were then subjected to incubation with FITC-labeled CTD/MM@BSA for either 1 or 6 h. Following three washes with PBS, lysosomes were stained using LysoTracker Red for 30 min. The intracellular colocalization of CTD/MM@BSA with lysosomes was finally visualized and analyzed by CLSM. In vitro cytotoxicity Hepa1-6 cells were plated in 96-well plates (5000 cells/well). Following a 24-h treatment with a series of concentrations of CTD, MM@BSA, or CTD/MM@BSA, cell viability was evaluated using a CCK-8 assay. The CI for CTD/MM@BSA, indicating the synergistic effect of the combined components, was derived from the viability data using CompuSyn software. Intracellular ROS generation Hepa1-6 cells were plated in 6-well plates (2 × 105 cells per well). Following an 8-h incubation period with Control, CTD, MM@BSA, or CTD/MM@BSA, the cells were exposed to DCFH-DA (10 μmol/L) for 30 min. ROS levels were subsequently measured using either fluorescence imaging or flow cytometry. Effects of ferroptosis modulators on relative cell viability To evaluate the effects of various ferroptosis modulators, Hepa1-6 cells were plated in 96-well plates. The cells were then exposed to CTD, MM@BSA, or CTD/MM@BSA. Following this treatment, ferrostatin-1 (fer-1, 1 μM, an inhibitor of ferroptosis), glutamate (Glu, 2 mM, an inducer of ferroptosis), cysteine (Cys, 2 mM, a precursor for cysteine), or glutathione (GSH, 2 mM, a GPX4 cofactor) was added to the respective wells. A control group that received no ferroptosis-modulating agents was also established. After 24 h of incubation, cell viability was evaluated using the CCK-8 assay. Measurement of intracellular Fe2+ content Hepa1-6 cells were seeded in confocal dishes and cultured overnight to allow adherence. Subsequently, the cells were treated with CTD, MM@BSA, or CTD/MM@BSA for 12 h. After collection, the cells were washed with PBS and then incubated with a staining solution containing 1 μM FerroOrange for 30 min at 37 °C in the dark. Finally, the cells were imaged using CLSM, and the intracellular Fe2+ content was quantitatively analyzed by flow cytometry. Measurement of lipid peroxidation Hepa1-6 cells were seeded in confocal dishes and cultured overnight to allow adherence. Subsequently, the cells were treated with CTD, MM@BSA, or CTD/MM@BSA for 12 h. After treatment, the cells were washed with PBS and fixed with fixative solution. Then, they were stained with 10 μM BODIPY lipid probe for 30 min, followed by DAPI staining for 10 min. Finally, imaging was performed using CLSM, and the level of lipid peroxidation was quantitatively analyzed by flow cytometry. Western blot (WB) After treatment, cells were lysed in RIPA buffer following PBS washes. Proteins were quantified by BCA assay, separated by SDS-PAGE, transferred to PVDF membranes, and probed with the indicated primary and HRP-conjugated secondary antibodies. Bands were visualized by chemiluminescence (BioSpectrum 300; UVP). γ-H2AX expression Hepa1-6 cells (2 × 105 cells/well) were plated in 6-well plates. Cells were treated with the indicated formulations for 12 h, washed with PBS, and incubated with anti–γ-H2AX primary antibody at 4 °C overnight, followed by a FITC-conjugated secondary antibody for 30 min. After washing, γ-H2AX levels were quantified by flow cytometry. In vitro calreticulin (CRT) expression Hepa1-6 cells were plated in confocal dishes at a density of 200,000 cells per milliliter and exposed to PBS, CTD, MM@BSA, or CTD/MM@BSA for 24 h. Following treatment, cells were then fixed with 4% formaldehyde for 15 min, blocked with goat serum for 1 h, and incubated overnight at 4 °C with an anti-CRT primary antibody. After a 2 h incubation with a FITC-conjugated rabbit secondary antibody at room temperature, cells were imaged using a confocal microscope. CRT expression was further quantified by flow cytometry. In vitro BMDCs stimulation Bone marrow–derived dendritic cells (BMDCs) were isolated from 8-week-old female mice. Hepa1-6 cells were seeded in the upper chamber of a Transwell insert, with BMDCs in the lower chamber. The co-culture system was exposed to the indicated treatments for 24 h. BMDCs maturation was assessed by flow cytometry following staining with CD86-PE, CD11c-APC, and MHC II-PE antibodies. Cytokine levels in the culture supernatants were measured by ELISA. Transcriptome sequencing Hepa1-6 cells were exposed to either a control treatment or CTD/MM@BSA for 24 h, after which cell samples were harvested for subsequent transcriptome sequencing analysis. In vivo biodistribution of CTD/MM@BSA A mouse model of hepatocellular carcinoma was established by subcutaneously injecting 100 μL of a Hepa1-6 cell suspension (1 × 107 cells/mL in PBS mixed 1:1 with Matrigel) into the right axilla of each mouse. This cell-Matrigel mixture was administered via subcutaneous injection into the right axilla of each mouse (100 μL per mouse). After tumor volume grew to around 200 mm3, mice bearing tumors (n = 3 per group) were given tail vein injections of either free Cy5.5 or CTD/MM@BSA-Cy5.5 complexes. Then, the mice were imaged and analyzed using the IVIS Lumina XRMS Series III imaging system. In vivo magnetic resonance imaging (MRI) The tumor-bearing mice were randomly divided into the MnCl2 and the CTD/MM@BSA. These animals received intravenous injections of either MnCl2 or CTD/MM@BSA ([Mn] = 3 mg/kg). A GE SIGNA HDxt 3.0T MRI scanner was utilized to capture the images. The imaging protocol included a T1WI Flair sequence (TR/TE = 1500/24 ms, matrix = 192 × 192, averages = 1, FOV = 80 × 80 mm) and a T2WI sequence (TR/TE = 3825/68 ms, matrix = 192 × 192, averages = 1, FOV = 80 × 80 mm). MRI scans were performed prior to the injection, as well as at 2-h and 24-h intervals following administration (with n = 4 mice in each group). To assess the degree of signal enhancement, the signal-to-noise ratio (SNR) was determined by the equation SNRtumor = SItumor/SDnoise, where SI denotes signal intensity and SD signifies the standard deviation. The alteration in SNR, expressed as ΔSNR, was calculated using the formula ΔSNR= (SNRtime-SNRpre) SNRpre. Additionally, ICP-OES was used to measure the manganese concentration in various tissues and tumors harvested from the two groups of mice. Pharmacokinetics SD rats were randomly divided into two groups with three rats in each group. The rats were administered MnCl2 solution or CTD/MM@BSA formulation, respectively. Blood samples were collected from the orbital venous plexus at predetermined time points. After centrifugation, plasma was separated, and the Mn content in the plasma at each time point was determined using ICP-OES. Evaluation of the in vivo antitumor activity and immunological impact of CTD/MM@BSA Tumor-bearing mice (∼100 mm3) were randomly assigned to receive PBS, CTD, MM@BSA, CTD/MM@BSA, or CTD/MM@BSA + aPD-L1 (CTD: 400 μg/kg) via tail vein injection on days 0, 3, 6, and 9. Tumor dimensions were recorded at three-day intervals throughout the treatment period, and growth curves were generated accordingly. On day 21, all animals were humanely euthanized. Tumor tissues were then excised, imaged, and weighed, followed by histological examination using H&E staining of both tumors and major organs. Additionally, changes in the body weight of the mice were monitored regularly during the study. The levels of liver and kidney functions in serum were quantitatively determined by using test kits. To evaluate immune responses, the activation state of DCs and T cells within tumor tissues was examined using flow cytometry. Levels of specific cytokines were quantified in tumor homogenates via ELISA. Furthermore, WB analysis was employed to detect CD4 and CD8 protein expression in the tumor samples. The expression intensity of proteins associated with the immune response in tumors were additionally visualized by immunofluorescence. Synergistic antitumor activity of CTD/MM@BSA in combination with anti-PD-L1 antibody therapy Tumor-bearing mice were randomly assigned to four groups (PBS, aPD-L1, CTD/MM@BSA, or CTD/MM@BSA + aPD-L1) and treated on days 0, 3, 6, and 9. aPD-L1 was administered intraperitoneally (7.5 mg/kg), while CTD-containing formulations were delivered via tail vein injection (CTD: 400 μg/kg). The dosage of aPD-L1 was directly adopted from our published study [33], where it was confirmed to effectively block the PD-1/PD-L1 pathway. Given that this immune mechanism is conserved across different cancer types, it was applied as a validated effective dose in the hepatocellular carcinoma model and combination therapy in this study. Subsequent analyses were performed as described in Section 2.22. Inhibition of Ferroptosis/STING on tumor growth and immune activation Tumor-bearing mice (approximately 100 mm3) were randomly divided into 6 groups and injected via the tail vein with PBS, Fer-1, C-176 (STING inhibitor), CTD/MM@BSA, CTD/MM@BSA + Fer-1, or CTD/MM@BSA + C-176 on days 0, 3, 6, and 9, respectively. During the treatment period, tumor sizes were recorded every three days to generate growth curves. On day 21, all animals were euthanized. Tumor tissues were then removed, imaged, and weighed. To assess the immune response following ferroptosis inhibition or STING inhibition, the activation status of DCs and T cells in tumor tissues was detected by flow cytometry. Survival rate analysis In this experiment, the mice were modeled, grouped and administered drugs as shown in 2.22. Mice were monitored for 45 days from the start of treatment, with tumor volumes exceeding 2000 mm3 considered as mortality. Remaining mice were euthanized at the end of the study. The anti-metastatic effect of CTD/MM@BSA To create a model of lung metastasis, Hepa1-6 cells (5 × 105/mL) were injected intravenously into C57BL/6 mice, marking this time point as Day −14. On Day 0, the mice were randomly assigned to four different groups and received intravenous injections of PBS, CTD, MM@BSA, CTD/MM@BSA (CTD: 400 μg/kg) on Days 0, 3, 6, and 9. The lung tissues were collected on Day 21. Subsequent histological examination was performed on lung sections to evaluate the extent of tumor metastasis. Results and discussion Synergistic amplification of ROS by CTD and Mn2+ enhances anti-tumor efficacy According to earlier research, Mn2+ triggers the production of ROS through a mechanism resembling the Fenton reaction, subsequently activating the cGAS–STING signaling pathway and promoting the infiltration of cytotoxic T cells [34]. Meanwhile, CTD has been reported to suppress HCC progression primarily by inducing DNA damage. Based on these findings, we hypothesized that CTD synergizes with Mn2+ to promote cell death. To validate the synergistic effect of CTD and Mn2+ (5 μg/mL CTD and 3.95 μg/mL Mn2+), intracellular ROS were initially tracked in real-time using the DCFH-DA fluorescent probe. The data revealed a marked increase in fluorescence intensity in cells treated with Mn2+ alone relative to the control, demonstrating that Mn2+ successfully boosts intracellular ROS levels through a Fenton-like reaction. The CTD also exhibited markedly increased fluorescence signals, confirming its independent ROS-inducing capability. Notably, the CTD + Mn2+ combination group demonstrated substantially higher fluorescence intensity than either monotherapy group, revealing a significant synergistic effect in ROS generation (Fig. 1A–B). CCK-8 assay evaluating the proliferative inhibition on Hepa1-6 cells showed significantly lower cell viability in the combination group (48.8 ± 1.5%) compared to Mn2+ (79.8 ± 3.7%) and CTD (64.3 ± 5.0%) monotherapies (Fig. 1C). Quantification of cell death by Annexin V-FITC/PI dual staining flow cytometry demonstrated that the proportion of dead cells in the combination treatment group rose to 40.0 ± 8.6%. This value was substantially greater than those achieved with individual therapies, showing a 1.7-fold increase over CTD alone (24.2 ± 0.8%) and a 2.5-fold increase relative to Mn2+ (16.3 ± 1.0%) monotherapy (Fig. 1D and E). These results systematically demonstrate that CTD and Mn2+ synergistically amplify ROS signal to significantly enhance anti-tumor efficacy. To further elucidate the specific mechanism by which CTD and Mn interact, we assessed the levels of γ-H2AX, a nuclear DNA (nDNA) damage indicator [35], as well as markers linked to ferroptosis, such as GSH concentrations and glutathione peroxidase 4 (GPX4) protein expression. Intracellular GSH levels and GPX4 decreased significantly following combination treatment (Fig. 1F–H), suggesting an enhancement of ferroptosis. Flow cytometric analysis further demonstrated a more substantial increase in γ-H2AX expression levels in the combined CTD and Mn treatment group compared to groups receiving CTD or Mn alone (Fig. 1I and J). This observation suggests that the combined treatment synergistically intensified DNA damage within the nucleus. Taken together, the collective antitumor activity of CTD and Mn is likely attributable to a synergistic mechanism that concurrently promotes ferroptosis and aggravates nuclear DNA damage.Fig. 1(A) The levels of ROS in Hepa1-6 cells under various treatment conditions were assessed using flow cytometry. (B) The relative mean fluorescence intensity (MFI) in Figure A, (n = 4). (C) Cell viability of CTD, Mn and CTD + Mn. (D) The cell death conditions of Hepa1-6 cells after treatment with CTD, Mn, and their combination. (E) Quantitative analysis of cell mortality from Figure D. (F) The expression of GPX4 proteins. (G) Quantitative expression of GPX4 protein (n = 3). (H) Relative GSH levels in Hepa1-6 cells after treatment with the indicated formulations. (I) Expression of γ-H2AX protein in Hepa1-6 cells treated with different formulations, detected by flow cytometry. (J) Relative expression intensity of the results shown in panel (I). Data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. (A) The levels of ROS in Hepa1-6 cells under various treatment conditions were assessed using flow cytometry. (B) The relative mean fluorescence intensity (MFI) in Figure A, (n = 4). (C) Cell viability of CTD, Mn and CTD + Mn. (D) The cell death conditions of Hepa1-6 cells after treatment with CTD, Mn, and their combination. (E) Quantitative analysis of cell mortality from Figure D. (F) The expression of GPX4 proteins. (G) Quantitative expression of GPX4 protein (n = 3). (H) Relative GSH levels in Hepa1-6 cells after treatment with the indicated formulations. (I) Expression of γ-H2AX protein in Hepa1-6 cells treated with different formulations, detected by flow cytometry. (J) Relative expression intensity of the results shown in panel (I). Data are presented as mean ± SD. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001. Synthesis and characterization of CTD/MM@BSA Given the dual delivery challenges of poor aqueous solubility for the hydrophobic drug CTD and limited transmembrane transport efficiency of free Mn2+ this study innovatively developed a multifunctional nano-coordination polymer, CTD/MM@BSA, based on BSA, aiming to achieve simultaneous hydrophobic drug loading and metal ion delivery. The optimal concentrations of BSA, 2-MI, CTD, and Mn2+ were established by evaluating the particle size and PDI. The final formulation consisted of 1.82 mg/mL BSA, 36.36 mM 2-MI, 27.27 mM Mn2+, and 200 μg/mL CTD (Figure S1 A-D). Drug loading characterization revealed that at a final CTD concentration of 200 μg/mL, the LD% reached 66.24% with an EE% of 26.05% (Fig. S1E), establishing this parameter as the standard for subsequent preparation. Quantification by ICP-OES revealed Mn2+ LD% and EE% of 1.65% and 15.2%, respectively. To confirm whether the identified loading ratio (1.51:1) exhibits biological synergy, we evaluated the combination effects of free CTD and free Mn2+ at this specific ratio on Hepa1-6 cells by calculating CI. The free drug combination exhibited CI < 1, confirming a synergistic effect between the two agents (Fig. S2). This finding provided a crucial pharmacological basis for employing this ratio in our nanoparticle. DLS characterization showed a hydrodynamic diameter of approximately 165 nm (PDI <0.2) and a zeta potential of −16.5 mV (Fig. 2A and S3), with the excellent dispersibility attributed to the electrostatic repulsion provided by BSA. TEM image clearly reveals individual nanoparticles (Fig. 2A). Size agrees well with the core size estimated from the DLS measurements. SDS-PAGE electrophoresis detected intact BSA bands in depolymerized samples (Fig. S4), confirming protein structural integrity and successful encapsulation. To elucidate the role of BSA in the formation of the nanoparticles, UV-Vis absorption spectra of 2-MI, BSA, and MM@BSA were compared. As shown in Fig. S5, pure BSA exhibited a characteristic absorption peak at 278 nm, whereas 2-MI showed no absorbance at this wavelength. Upon the addition of 2-MI and Mn2+, a marked hyperchromic effect was observed at 278 nm, indicating that Mn2+ interacted with BSA. Furthermore, in the absence of BSA, the reaction system failed to yield uniform and well-dispersed nanoparticles, instead forming irregular, large aggregates (Fig. 2B and C). These findings demonstrate that BSA is not merely a passive stabilizer preventing nanoparticle aggregation after formation. To elucidate the molecular interaction mechanisms driving nanoparticle assembly and drug loading, we conducted a series of mechanistic studies. First, CTD/MM@BSA were co-incubated with three probes (NaCl, Urea, and SDS). Significant drug release (∼34.1%) was observed only in the presence of SDS, indicating that CTD loading relies on hydrophobic interactions with BSA (Fig. S6A). Second, to verify the coordination between Mn2+ and sulfhydryl groups in BSA, NaCl, Urea, and GSH were added. NaCl or urea caused only about 10% BSA detachment from the nanoparticles, with the particles remaining largely intact. In contrast, glutathione led to near-complete dissociation of the nanoparticles, with approximately 49.86% BSA release, confirming the critical role of Mn-S bonds in structural stability (Fig. S6B). Finally, to investigate the binding mode between Mn2+ and 2-MI, EDTA was used as a strong metal chelator. In the absence of BSA, 2-MI and Mn2+ spontaneously formed a precipitate, which dissolved upon EDTA addition (Fig. S6C). This indicates that EDTA competitively chelates Mn2+, disrupting its coordination with 2-MI, thereby demonstrating that the Mn/2-MI interaction is based on metal coordination. XPS analysis confirmed successful incorporation of Mn2+ and 2-MI into CTD/MM@BSA nanoparticles. The survey spectrum clearly showed characteristic signals of S, Mn and N (originating from 2-MI), confirming the coexistence of BSA, Mn, and 2-MI in the final nanoparticles (Fig. 2D). The high-resolution Mn 2p spectrum exhibited two main peaks at approximately 641.05 eV (Mn 2p3/2) and 653.04 eV (Mn 2p1/2), accompanied by additional peaks at 642.61 eV (Mn3+ 2p3/2) and 654.2 eV (Mn3+ 2p1/2), verifying the successful doping of manganese into the nanostructure. The N 1s spectrum was deconvoluted into two peaks at about 398.8 eV and 400.1 eV, which are assignable to pyridinic nitrogen and pyrrolic nitrogen on the imidazole ring of 2-MI, respectively, demonstrating that 2-MI was successfully assembled into CTD/MM@BSA (Fig. 2E and F).Benefiting from the stabilizing effect of BSA, the nanoparticles maintained excellent colloidal stability in both aqueous and serum environments, with negligible particle size fluctuation after 9 days of storage at 4 °C (Fig. 2G and H). Furthermore, the degradation behavior of CTD/MM@BSA nanoparticles was investigated under simulated physiological conditions. As shown in Fig. S7A, the color of the nanoparticle dispersion changed from yellow to transparent in the presence of GSH, indicating the breakdown of NPs. DLS measurements (Fig. S7B) showed that the particle size remained stable in Hepes buffer (pH 7.4) for up to 20 h, suggesting good colloidal stability suitable for prolonged circulation and tumor accumulation. In contrast, under simulated tumor microenvironment conditions (pH 5.0 + 10 mM GSH), the hydrodynamic diameter increased rapidly, and the nanoparticles subsequently disassembled completely. These results confirm that CTD/MM@BSA nanoparticles exhibit excellent stability under physiological conditions but undergo rapid and responsive degradation in an acidic and reductive tumor microenvironment. Subsequently, the release profile of Mn2+ from the nanoparticles was evaluated. As shown in Fig. 2I, under pH 7.4 conditions, the release of Mn2+ was very slow, with a cumulative release of less than 20% within 24 h, indicating minimal leakage under normal physiological conditions and a low risk of systemic side effects. In contrast, release was markedly accelerated at pH 5.0. Under the fully simulated tumor microenvironment (pH 5.0 + 10 mM GSH), a burst release was observed, with a cumulative Mn2+ release exceeding 80% within 24 h. These results demonstrate that the acidic pH and high GSH level synergistically trigger efficient drug release from the nanoparticles. Functional validation confirmed the significant Fenton-like catalytic activity of the nanosystem, under H2O2 conditions, monitoring the attenuation of the characteristic absorption peak of methylene blue (MB) at 660 nm (which decreases with increasing ·OH concentration) demonstrated that CTD/MM@BSA efficiently catalyzed MB degradation (Fig. S8), providing a crucial theoretical basis for subsequent antitumor applications.Fig. 2(A) Size distribution characteristics of CTD/MM@BSA. Illustration: TEM image of CTD/MM@BSA. (scale bar = 100 nm). (B) Schematic illustration and (C) Hydrodynamic size of CTD/MM and CTD/MM@BSA. (D) XPS survey spectrum of CTD/MM@BSA. (E) High-resolution Mn 2p spectrum of CTD/MM@BSA. (F) High-resolution N 1s spectrum of CTD/MM@BSA. (G) Stability of CTD/MM@BSA nanoparticles incubated in H2O and 10% FBS. (H) Variations in the size and PDI of the CTD/MM@BSA nanoparticles were monitored over a 9-day storage period at 4 °C. (I) Manganese release profile from CTD/MM@BSA over 24 h. (A) Size distribution characteristics of CTD/MM@BSA. Illustration: TEM image of CTD/MM@BSA. (scale bar = 100 nm). (B) Schematic illustration and (C) Hydrodynamic size of CTD/MM and CTD/MM@BSA. (D) XPS survey spectrum of CTD/MM@BSA. (E) High-resolution Mn 2p spectrum of CTD/MM@BSA. (F) High-resolution N 1s spectrum of CTD/MM@BSA. (G) Stability of CTD/MM@BSA nanoparticles incubated in H2O and 10% FBS. (H) Variations in the size and PDI of the CTD/MM@BSA nanoparticles were monitored over a 9-day storage period at 4 °C. (I) Manganese release profile from CTD/MM@BSA over 24 h. Multiscale delivery validation of CTD/MM@BSA To begin a systematic exploration of the cellular internalization of CTD/MM@BSA nanoparticles, we first constructed fluorescent nanoparticles using fluorescein isothiocyanate (FITC)-labeled BSA. Cell nuclei were counterstained with DAPI. According to the CLSM images presented in Fig. 3A, cells treated with CTD/MM@BSA displayed markedly stronger green fluorescence signals and a greater number of nanoparticle-positive cells compared with the control group, confirming efficient nanoparticle internalization. To further quantify intracellular fluorescence intensity, flow cytometry analysis was performed. The relative MFI of the CTD/MM@BSA–FITC group was significantly higher than that of free FITC. This result verifies that using the nanocarrier considerably improved the efficiency of cellular drug uptake (Fig. 3B and C). This result was in strong agreement with the CLSM observation. To track the intracellular trafficking of the nanoparticles, we further performed a lysosomal co-localization experiment. Staining with LysoTracker Red showed that after 1 h of incubation, although partial overlap between the FITC signal (nanoparticles) and the lysosomal signal was observed, most nanoparticles had already escaped from the lysosomes (Fig. 3D). At 6 h, a further increase in intracellular green fluorescence was observed, indicating continued nanoparticle uptake. Importantly, however, the red-green co-localization ratio at this time point was markedly lower than that at 1 h (Fig. S9), indicating that lysosomal escape is not a static event but a continuous and efficient process. Collectively, these results confirm that CTD/MM@BSA nanoparticles can achieve lysosomal escape, and deliver active components to their target sites. Having established the efficient cellular internalization and lysosomal escape of our NPs, we sought to contextualize these findings within our broader therapeutic strategy. While the non-specific uptake of NPs by tumor-resident immune cells is often viewed as a hurdle for targeted delivery, our NPs platform leverages this phenomenon as an integral component of its therapeutic strategy. Our in vitro results demonstrated that the nanoparticles were internalized by both cancer cells and immune cells (macrophages and dendritic cells), with markedly higher uptake efficiency observed in cancer cells (Fig. 3E and F). This specific cellular tropism is central to our mechanism: the high uptake in cancer cells initiates the core therapeutic cascade synergistic ferroptosis and cGAS-STING activation. Concurrently, the uptake by antigen-presenting cells like DCs is not a loss but a strategic gain; it directly contributes to their maturation, working in concert with DAMPs released from the dying cancer cells to potentiate the anti-tumor T cell response. Thus, the seemingly non-specific uptake profile is, in fact, a deliberately harnessed feature that ensures a coordinated kick-start and amplification of the cancer-immunity cycle. We further investigated the distribution of nanoparticles in vivo. CTD/MM@BSA was labeled with Cy5.5 to assess in vivo biodistribution. At 24 h post-injection, tumors in the CTD/MM@BSA group exhibited markedly higher fluorescence than those in the free Cy5.5 group, indicating enhanced tumor targeting and accumulation (Fig. 3G). Ex vivo imaging of major organs and tumor tissue, along with quantitative fluorescence intensity analysis, further corroborated these findings (Fig. 3H and I). The fluorescence intensity of nanoparticles within tumors was approximately 2.8-fold higher than that of free Cy5.5. To further validate and quantify manganese accumulation, we performed in vivo MRI and ex vivo tissue content analysis. In T1-weighted MRI, the CTD/MM@BSA group showed significantly stronger signal enhancement in tumor regions compared to the free MnCl2 group at equivalent manganese doses. Conversely, in T2-weighted imaging, the CTD/MM@BSA group exhibited more pronounced signal attenuation over time, consistent with nanoparticle accumulation and aggregation at the tumor site. Quantitative analysis of signal-to-noise ratio changes (ΔSNR) in T1-weighted imaging further validated these observations (Fig. 3J and K and S10A, 10B). Direct ICP-OES analysis of manganese levels in harvested tissues demonstrated a markedly greater accumulation of manganese within the tumors of the CTD/MM@BSA cohort compared to the MnCl2 group (Fig. S10C). In summary, systematic studies from the cellular to the whole-body level demonstrate that CTD/MM@BSA nanoparticles not only achieve efficient cellular uptake, lysosomal escape, and possess a cellular tropism conducive to activating anti-tumor immunity, but also accumulate in tumor tissue in vivo via the enhanced permeability and retention (EPR) effect. These properties collectively establish a solid delivery foundation for this nanoplatform to enhance anti-tumor efficacy while reducing systemic toxicity.Fig. 3Cellular uptake, lysosomal colocalization, cell type selectivity, and in vivo targeting and imaging performance of CTD/MM@BSA. (A) Cellular uptake of Hepa1-6 cells following treatment with various formulations. Scale bar: 20 μm. (B-C) Cellular uptake levels of the formulations quantified by flow cytometry. (D) CLSM image showing the lysosomal colocalization of CTD/MM@BSA. Scale bar: 20 μm. (E-F) Cellular uptake of CTD/MM@BSA in different cell types. (G) Fluorescence imaging of tumors performed in living animals and (H) after extraction. (I) Quantification of the measured fluorescence signal intensity (n = 3). (J) T1-and (K) T2-weighted MR images of tumor-bearing mice at 0, 2, and 24 h post-injection with MnCl2 and CTD/MM@BSA. Cellular uptake, lysosomal colocalization, cell type selectivity, and in vivo targeting and imaging performance of CTD/MM@BSA. (A) Cellular uptake of Hepa1-6 cells following treatment with various formulations. Scale bar: 20 μm. (B-C) Cellular uptake levels of the formulations quantified by flow cytometry. (D) CLSM image showing the lysosomal colocalization of CTD/MM@BSA. Scale bar: 20 μm. (E-F) Cellular uptake of CTD/MM@BSA in different cell types. (G) Fluorescence imaging of tumors performed in living animals and (H) after extraction. (I) Quantification of the measured fluorescence signal intensity (n = 3). (J) T1-and (K) T2-weighted MR images of tumor-bearing mice at 0, 2, and 24 h post-injection with MnCl2 and CTD/MM@BSA. Evaluation of in vitro anti-tumor efficacy of CTD/MM@BSA Upon confirming the effective delivery of nanoparticles, the cytotoxic responses of various formulations were assessed in Hepa1-6 and HepG2 cell lines via the CCK-8 assay. A 24-h incubation period revealed that all tested formulations induced cytotoxicity in a manner dependent on dosage (Fig. 4A and Fig. S11). Notably, the CTD/MM@BSA showed a significantly lower IC50 value compared to free CTD and MM@BSA (Fig. 4B and C). To further validate that the enhanced cytotoxicity was due to drug synergy rather than merely improved delivery, the CI of CTD/MM@BSA was quantified. The results demonstrated a CI < 1, indicating a robust synergistic effect between the components within the nanoparticles (Fig. S12). This synergistic cell killing, combined with the improved cellular uptake facilitated by the nanocarrier, underpins the superior therapeutic efficacy of our formulation. To investigate the mechanism of action of CTD/MM@BSA, we measured intracellular ROS levels in Hepa1-6 cells. Flow cytometric assessment indicated that the CTD/MM@BSA exhibited a greater level of intracellular fluorescence intensity relative to the MM@BSA and CTD (Fig. 4D and Fig. S13). Fluorescence imaging confirmed ROS production across the groups, MM@BSA generated weak fluorescence via Mn2+-mediated Fenton-like reactions, while CTD alone also induced ROS elevation. Notably, the fluorescence signal observed for the CTD/MM@BSA was substantially more intense compared to that of the individual CTD or MM@BSA (Fig. 4E), indicating that intracellularly accumulated Mn2+ synergizes with CTD to substantially enhance ROS generation, suggesting the formulation might trigger ferroptosis through an oxidative stress pathway. To validate the occurrence of ferroptosis, Hepa1-6 cells were treated with various ferroptosis inhibitors (Fer-1, Cys), a detoxifier (GSH), and an inducer (Glu) in combination with the different formulations, and cytotoxicity was compared to groups without these modulators. Results showed that ferroptosis inhibitors (Fer-1, Cys) and the detoxifier GSH significantly alleviated the cytotoxicity of CTD/MM@BSA. Conversely, the inducer Glu enhanced its toxicity (Fig. 4F). These findings demonstrate that CTD/MM@BSA effectively induces ferroptosis. Given that GSH depletion and inhibition of GPX4 activity are key hallmarks of ferroptosis, we further assessed intracellular GSH levels. Results indicated that GSH content was significantly lower in CTD/MM@BSA-treated cells compared to the CTD or MM@BSA groups (Fig. 4G). Additionally, GPX4 protein expression showed a decreasing trend. WB analysis confirmed that CTD/MM@BSA downregulates SLC7A11 protein expression, collectively demonstrating its ferroptosis-inducing mechanism (Fig. 4H and Fig. S14). After inhibiting GPX4 activity and elevating ROS levels, we evaluated intracellular Fe2+ accumulation and lipid peroxidation. Fe2+ was detected using the FerroOrange probe, which is non-fluorescent alone but emits orange fluorescence upon binding to Fe2+. CLSM revealed that the orange fluorescence in the CTD/MM@BSA-treated group was markedly stronger than that in the control, CTD, and MM@BSA groups, a result further confirmed by flow cytometry (Figs. S15A, C, D). Lipid peroxidation levels were measured using C11-BODIPY 581/591, a probe whose emission shifts from 590 nm (red) to 510 nm (green) upon oxidation. Both CLSM and flow cytometry results showed that CTD/MM@BSA treatment induced the most intense green fluorescence signal, indicating the highest production of lipid peroxides. Quantitative analysis revealed that the green fluorescence intensity in CTD/MM@BSA-treated cells was 3.90-fold that of the control group, and 1.71-fold and 1.55-fold those of the CTD and MM@BSA groups, respectively (Figs. S15B, E, F). These results directly confirm the accumulation of Fe2+ and the generation of lipid peroxides, thereby establishing the activation of ferroptosis.Fig. 4Evaluation of ROS induction and related cellular effects in Hepa1-6 cells following treatment with CTD/MM@BSA. (A) Viability of Hepa1-6 cells treated with different preparations. IC50 was determined for (B) CTD and (C) the Mn-based formulation after 24 h of treatment. (D) Flow cytometric analysis of intracellular ROS levels in Hepa1-6 cells exposed to different treatments. (E) Fluorescence imaging illustrating ROS production in treated Hepa1-6 cells. scale bar: 50 μm. (F) Comparative cell viability in Hepa1-6 cells cultured with different formulation groups, with or without specific inhibitors. (G) Relative GSH levels in Hepa1-6 cells under various treatment conditions. (H) The expression of SLC7A11 and GPX4 proteins. (I) Flow cytometry detection of γ-H2AX protein expression in cells treated with different formulations. (J) The expression of cGAS, P-TBK1, P-STING, and P-IRF3 protein. (K) Release of IFN-β after treating Hepa1-6 cells with different preparations. Data are presented as mean ± SD (n = 3). ∗∗∗∗p < 0.0001. Evaluation of ROS induction and related cellular effects in Hepa1-6 cells following treatment with CTD/MM@BSA. (A) Viability of Hepa1-6 cells treated with different preparations. IC50 was determined for (B) CTD and (C) the Mn-based formulation after 24 h of treatment. (D) Flow cytometric analysis of intracellular ROS levels in Hepa1-6 cells exposed to different treatments. (E) Fluorescence imaging illustrating ROS production in treated Hepa1-6 cells. scale bar: 50 μm. (F) Comparative cell viability in Hepa1-6 cells cultured with different formulation groups, with or without specific inhibitors. (G) Relative GSH levels in Hepa1-6 cells under various treatment conditions. (H) The expression of SLC7A11 and GPX4 proteins. (I) Flow cytometry detection of γ-H2AX protein expression in cells treated with different formulations. (J) The expression of cGAS, P-TBK1, P-STING, and P-IRF3 protein. (K) Release of IFN-β after treating Hepa1-6 cells with different preparations. Data are presented as mean ± SD (n = 3). ∗∗∗∗p < 0.0001. CTD mediates its anti-tumor activity primarily by inducing nDNA damage and downregulating DNA repair-associated proteins. Within the TME, the cGAS-STING signaling axis functions as a pivotal innate immune regulator. This pathway detects the presence of aberrant DNA in the cytosol, initiating a series of immune responses. Upon detection of abnormal DNA, the DNA sensor cGAS facilitates the synthesis of the secondary messenger cGAMP, which in turn stimulates the STING protein and attracts TBK1 kinase. Consequently, IRF3 undergoes phosphorylation, relocates to the nucleus, and prompts the expression of IFN-I [36]. Studies indicate that Mn2+ significantly enhances cGAS sensitivity to DNA, promotes cGAMP synthesis, and amplifies the STING-TBK1-IRF3 signaling cascade, thereby boosting IFN-I secretion. Based on this, we hypothesized that the combination of CTD and Mn2+ might activate the cGAS-STING pathway by inducing nDNA damage, consequently triggering an innate immune response [37]. To test this hypothesis, we examined the expression of nDNA damage markers and cGAS-STING pathway-related proteins. Flow cytometry analysis of γ-H2AX showed that both CTD and MM@BSA treatments caused a moderate upregulation of γ-H2AX expression, confirming their ability to induce nDNA damage. Notably, the upregulation of γ-H2AX expression was significantly more pronounced in the CTD/MM@BSA group compared to the CTD or MM@BSA (Fig. 4I and Fig. S16), suggesting that the combined action of CTD and Mn2+ synergistically amplifies nuclear DNA damage. To evaluate the activation status of the cGAS-STING pathway, we measured the expression levels of key pathway proteins (cGAS, P-STING, P-TBK1, P-IRF3) by WB (Fig. 4J and Fig. S17). Results demonstrated higher expression of these proteins in the CTD/MM@BSA compared to other. This is attributable to the dual effects of Mn2+ enhancing cGAS activity and the CTD/Mn2+ combination inducing increased nDNA release. Elevated IFN-β levels in the supernatant further corroborated these findings (Fig. 4K). Collectively, these results demonstrate that CTD/MM@BSA, leveraging the synergistic effect of CTD and Mn2+, effectively activates the cGAS-STING pathway by inducing more pronounced DNA damage and cooperatively enhancing cGAS activity, thereby promoting an innate immune response. In-vitro immunogenic death induction and immune activation Manganese (Mn)-involved ferroptosis releases DAMPs, thereby triggering ICD. Hallmark features include the high surface expression of CRT on dying tumor cells and the subsequent release of ATP and HMGB1 [38,39]. Consequently, we first assessed CRT surface exposure using flow cytometry and measured HMGB1 release via WB and ELISA. Results demonstrated that all treatment groups induced significant CRT exposure and HMGB1 release (Fig. 5A and B and S18). Notably, compared to CTD or MM@BSA, CTD/MM@BSA significantly enhanced the induction of ICD, highlighting the synergistic effect between CTD and Mn2+. To further confirm CRT exposure, we performed immunofluorescence assays using CLSM. The results indicated that the green fluorescence signal of CRT on the Hepa1-6 cell membrane was significantly increased after treatment with CTD, MM@BSA and CTD/MM@BSA (Fig. S19). Subsequently, we measured intracellular ATP release using an ATP assay kit, showing that all treatments significantly reduced ATP content compared with the Control, with CTD/MM@BSA inducing the greatest ATP release (Fig. 5C). This further supports the induction of complete immunogenic cell death. Considering these DAMPs can provide co-stimulatory signals to DCs, we further evaluated DCs maturation. Consistent with the above findings, all formulations promoted the expression of CD86 and MHC II on BMDCs. Critically, the CTD/MM@BSA treatment group induced BMDCs maturation significantly more effectively than the CTD or MM@BSA treatment groups (Fig. 5D–G). Additionally, we examined the expression of the co-stimulatory molecule CD80. Similarly, the CTD/MM@BSA group showed upregulation of CD80 expression compared to other treatment groups (Fig. S20), consistent with the patterns observed for CD86 and MHC II, further confirming that the combination of CTD and Mn2+ potently enhances tumor cell immunogenicity and effectively promotes full activation of BMDCs. Activation of the cGAS-STING pathway is critical for initiating anti-tumor innate immunity, leading to phosphorylation of TBK1 and IRF3 and subsequent production of type I interferons, especially IFN-β [40]. Mn2+ can enhance this pathway by increasing cGAS sensitivity to cytosolic dsDNA [21,41]. In addition to the anti-tumor immunity triggered by ICD, CTD/MM@BSA is expected to enhance immune responses against tumors through STING pathway activation in DCs following their uptake of the complex. To assess this, BMDCs were treated in vitro with either PBS or CTD/MM@BSA, and subsequent WB was performed to examine proteins associated with the STING signaling pathway. Results revealed that CTD/MM@BSA treatment significantly upregulated cGAS protein levels in BMDCs and promoted P-TBK1, P-IRF3 and P-STING(Fig. S21). Furthermore, significantly elevated levels of key cytokines IFN-γ, IFN-β, TNF-α, and IL-6 provided additional support for the concept that CTD/MM@BSA effectively activates mature BMDCs and drives their participation in active immune responses (Fig. 5H–K). The results indicate that CTD/MM@BSA potently increases tumor immunogenicity through a dual mechanism.Fig. 5Analysis of ICD-related marker expression and DCs activation induced by different formulations. (A) Flow cytometric analysis of CRT surface exposure on Hepa1-6 cells following treatment with the different formulations. (B) The expression of intracellular HMGB1 protein. (C) The release of intracellular ATP levels (n = 5). (D-G) Percentage of CD86+ and MHC II+ BMDCs co-cultured with Hepa1-6 cells pre-treated with different formulations. (H-K) The levels of inflammatory cytokines released by BMDCs (n = 3). Analysis of ICD-related marker expression and DCs activation induced by different formulations. (A) Flow cytometric analysis of CRT surface exposure on Hepa1-6 cells following treatment with the different formulations. (B) The expression of intracellular HMGB1 protein. (C) The release of intracellular ATP levels (n = 5). (D-G) Percentage of CD86+ and MHC II+ BMDCs co-cultured with Hepa1-6 cells pre-treated with different formulations. (H-K) The levels of inflammatory cytokines released by BMDCs (n = 3). Mechanism of CTD/MM@BSA-induced ferroptosis-STING in promoting antitumor immunity To clarify the complex molecular processes that drive the initiation and potentiation of the ferroptosis-STING signaling cascade, we performed RNA sequencing and transcriptomic analysis on cell samples from two independent cohorts. Differentially expressed genes (DEGs) were identified by applying a statistical significance threshold of P < 0.05 and a fold change cut-off of |log2FC| ≥ 1. Subsequent differential expression analysis was conducted on DEGs from the CTD/MM@BSA-treated group compared to the control group. Volcano plot analysis revealed a total of 866 DEGs, comprising 607 significantly upregulated genes (P < 0.05, log2FC > 1) and 259 significantly downregulated genes (P < 0.05, log2FC < −1) (Fig. 6A). Based on these results, we selected 14 genes potentially associated with CTD/MM@BSA-induced immunogenic cell death and immune response mechanisms. A heatmap visually illustrated the complex expression patterns of these genes (Fig. 6B). To comprehensively evaluate the integrated effects of CTD/MM@BSA on tumors, we performed Gene Ontology (GO) database enrichment analysis and KEGG pathway enrichment analysis on the DEGs. GO analysis revealed that DEGs were significantly enriched in ferroptosis and innate/adaptive immune responses at the Biological Process (BP) level. Notably, the innate/adaptive immune response pathways are typical downstream events following STING pathway activation [42]. Within the Molecular Function (MF) category, DEGs showed significant enrichment in pathways closely associated with ferroptosis and STING activation, such as glutathione transferase activity and cGMP-stimulated cyclic-nucleotide phosphodiesterase activity (Fig. 6C). This finding provides crucial clues regarding the ferroptosis mechanism. One of the core features of ferroptosis is the disruption of glutathione metabolism and inhibition of GPX4 activity, while alterations in glutathione transferase activity generally reflect cellular responses to oxidative stress and lipid peroxidation. Therefore, it can be inferred that CTD/MM@BSA induces ferroptosis by interfering with the cellular antioxidant defense system, particularly the glutathione system, leading to the accumulation of lipid ROS. This inference is consistent with our previous in vitro experimental results showing that CTD/MM@BSA inhibits GPX4 expression and promotes GSH depletion. KEGG pathway analysis further demonstrated significant enrichment of DEGs in pathways closely associated with STING activation, such as Cytokine-cytokine receptor interaction (Fig. 6D). Upon activation, the STING signaling cascade induces the expression and release of type I interferons, along with a range of pro-inflammatory cytokines [43]. Hence, the enrichment of this pathway provides strong evidence of STING activation. To further clarify CTD/MM@BSA's inducing or promoting effects on the Cell cycle, cGMP-PKG signaling pathway, DNA replication, and cAMP signaling pathway in tumors, we performed Gene Set Enrichment Analysis (GSEA) on the DEGs. GSEA results indicated that CTD/MM@BSA suppresses hepatocellular carcinoma cell proliferation by disrupting cell cycle progression and DNA replication functions, as well as by modulating the cGMP-PKG and cAMP signaling pathways (Fig. 6E). This direct inhibition of cell proliferation complements the immune-mediated cytotoxic effects driven by ferroptosis-STING activation. On one hand, CTD/MM@BSA directly impedes tumor cell growth; on the other hand, it activates the immune system via the ferroptosis-STING axis to clear tumor cells, achieving a dual anti-tumor effect. In summary, these transcriptomic analyses provide preliminary evidence that CTD/MM@BSA may initiate systemic anti-tumor immunity via ferroptosis-triggered STING signaling, and they establish a foundation for subsequent in vivo investigations.Fig. 6Transcriptomic profiling and functional enrichment analysis of CTD/MM@BSA-treated group versus control group. (A) A volcano plot illustrating the DEGs identified from a comparison of the CTD/MM@BSA-treated samples and the control group. (B) The expression patterns of immune-related genes that were differentially expressed are displayed in a heatmap, comparing the CTD/MM@BSA group with the control. (C) GO analysis was performed to identify the functional categories enriched among the DEGs following a 24-h exposure to CTD/MM@BSA, relative to the control. (D) KEGG pathway enrichment analysis of DEGs after 24-h treatment with CTD/MM@BSA versus control. (E) GSEA plot for the CTD/MM@BSA group versus control group after 24-h treatment. Transcriptomic profiling and functional enrichment analysis of CTD/MM@BSA-treated group versus control group. (A) A volcano plot illustrating the DEGs identified from a comparison of the CTD/MM@BSA-treated samples and the control group. (B) The expression patterns of immune-related genes that were differentially expressed are displayed in a heatmap, comparing the CTD/MM@BSA group with the control. (C) GO analysis was performed to identify the functional categories enriched among the DEGs following a 24-h exposure to CTD/MM@BSA, relative to the control. (D) KEGG pathway enrichment analysis of DEGs after 24-h treatment with CTD/MM@BSA versus control. (E) GSEA plot for the CTD/MM@BSA group versus control group after 24-h treatment. Evaluation of the in vivo anti-tumor performance of CTD/MM@BSA Following confirmation of cellular activity, we proceeded to assess their in vivo performance in mouse models bearing Hepa1-6 tumors. Hemocompatibility assays demonstrated that CTD/MM@BSA exhibited a hemolysis rate below 5% at concentrations up to 200 μg/mL (Fig. S22), indicating its suitability for intravenous administration. To further investigate the pharmacokinetic behavior of CTD/MM@BSA, we measured serum manganese concentrations by ICP-OES to evaluate its in vivo circulation. The results showed that CTD/MM@BSA exhibited an elimination half-life (t1/2) of 4.96 ± 0.74 h, significantly longer than that of the MnCl2 (0.94 ± 0.51 h). The total clearance (CL) of CTD/MM@BSA was 0.59 ± 0.04 L/h/kg, markedly lower than that of the MnCl2 (0.97 ± 0.1 L/h/kg). The area under the concentration-time curve from 0 to 24 h (AUC0→24) was 370.88 ± 28.15 mg/L∗h, approximately 1.68-fold higher than that of the MnCl2 (Fig. S23). These data indicate that CTD/MM@BSA significantly prolongs the circulation retention time of manganese and reduces its clearance rate. The in vivo anti-tumor effect of drug administration is shown in Fig. 7A. During the efficacy evaluation, it was observed that CTD and MM@BSA alone provided only moderate tumor growth inhibition. This indicates that while tumor growth speed was somewhat reduced, the inhibitory effect was limited and failed to effectively control tumor progression. In contrast, the composite formulation CTD/MM@BSA, formed by loading CTD onto MM@BSA nanoparticles, demonstrated significant tumor-inhibitory capability, with efficacy far superior to that of free CTD or MM@BSA. This marked enhancement is primarily attributed to two key factors. The pronounced EPR effect of the nanoparticles, which promotes passive tumor targeting and increases intratumoral CTD accumulation. A synergistic antitumor action occurs between CTD and Mn2+. This synergy is potentially mediated by mechanisms such as enhanced immune modulation and promotion of ROS generation collectively contributing to tumor cell killing. Notably, combining CTD/MM@BSA with the anti-PD-L1 antibody (aPD-L1), an immunotherapy widely used for liver cancer, further enhanced the antitumor effect (Fig. 7B and C). Direct assessment of antitumor efficacy by measuring final tumor weights and corresponding macroscopic tumor images confirmed these results (Fig. 7D and E). The most substantial suppression of tumor growth was observed in the combined treatment group receiving CTD/MM@BSA + aPD-L1. The efficacy of the other groups, in descending order, was CTD/MM@BSA, MM@BSA, and free CTD. H&E staining of tumor sections for pathological analysis showed extensive necrotic areas and a lower density of nuclei in both the CTD/MM@BSA + aPD-L1 and CTD/MM@BSA groups (Fig. 7F), which confirms their strong antitumor activity. Concurrently, survival time was significantly prolonged in mice treated with CTD/MM@BSA + aPD-L1 and CTD/MM@BSA (Fig. 7G), demonstrating the favorable therapeutic outcome and good biocompatibility of CTD/MM@BSA. Collectively, the CTD/MM@BSA demonstrates significant efficacy in suppressing tumor progression and extending the survival of tumor-bearing mice, with its efficacy being significantly superior to that of the CTD or MM@BSA. To further validate the combined effect of CTD/MM@BSA and aPD-L1, tumor-bearing mice were randomly assigned to four groups (PBS, aPD-L1, CTD/MM@BSA, and CTD/MM@BSA + aPD-L1, [aPD-L1] = 7.5 mg/kg, [CTD] = 400 μg/kg). Each group was administered a different formulation once every three days, for a total of four administrations (Fig. S24A). To assess the therapeutic effectiveness, measurements of tumor volume were taken at three-day intervals. In contrast to the aPD-L1 monotherapy group, which demonstrated only modest inhibition of tumor growth, the combination treatment with CTD/MM@BSA and aPD-L1 resulted in substantially reduced tumor volumes and weights (Fig. S24B–E). This outcome, indicates a more potent suppression of tumor progression and points to a synergistic effect between the two therapies. Potential toxicity was assessed by monitoring liver and kidney function markers (AST, ALT, BUN, CRE) as well as body weight (Fig. S24F–H). Encouragingly, the combination therapy group showed no notable alterations in these parameters compared to the PBS control group, suggesting a favorable safety profile. In conclusion, when combined with the immunotherapeutic agent aPD-L1, CTD/MM@BSA produces a synergistic effect, leading to more potent tumor suppression and a more substantial survival benefit.Fig. 7Evaluation of the therapeutic effectiveness of various formulations on tumor-bearing mice. (A) Schematic diagram of tumor model establishment and drug administration. (B) Individual tumor growth kinetics in mice administered distinct formulations. (C) The growth curves of tumors in each group of mice (n = 5). (D) Visual images of tumors in each group of mice. (E) Weight measurements of the tumors. (F) Histological sections of tumor tissue stained with H&E. Scale bar = 50 μm. (G) Analysis of survival rates in mice across various formulation groups (n = 6). (H) Individual tumor volume change curves of mice treated with different formulations. (I) Tumor growth curves of mice in different dosage form groups. Data presented as mean ± SD, n = 5, ∗∗p < 0.01. (J) Representative photographs of tumors from each group at the end of treatment. (K) Tumor weight. n = 5. Evaluation of the therapeutic effectiveness of various formulations on tumor-bearing mice. (A) Schematic diagram of tumor model establishment and drug administration. (B) Individual tumor growth kinetics in mice administered distinct formulations. (C) The growth curves of tumors in each group of mice (n = 5). (D) Visual images of tumors in each group of mice. (E) Weight measurements of the tumors. (F) Histological sections of tumor tissue stained with H&E. Scale bar = 50 μm. (G) Analysis of survival rates in mice across various formulation groups (n = 6). (H) Individual tumor volume change curves of mice treated with different formulations. (I) Tumor growth curves of mice in different dosage form groups. Data presented as mean ± SD, n = 5, ∗∗p < 0.01. (J) Representative photographs of tumors from each group at the end of treatment. (K) Tumor weight. n = 5. To investigate the tumor suppression mechanism of CTD/MM@BSA, we further evaluated its in vivo antitumor efficacy in immunodeficient mouse models. The results showed that even under immunodeficient conditions, each treatment group still exhibited certain antitumor effects. Among them, CTD/MM@BSA demonstrated significantly superior antitumor efficacy compared to free CTD and MM@BSA, achieving a tumor inhibition rate of 58.16%, indicating that the antitumor effect of CTD/MM@BSA is not entirely dependent on adaptive immunity (Fig. 7H–K). However, compared with its efficacy in immunocompetent mice (tumor inhibition rate of 82.58%), a certain degree of reduction was observed, suggesting that adaptive immunity plays a key synergistic and amplifying role in this therapeutic system. In summary, CTD/MM@BSA exerts its potent antitumor effect through a dual synergistic mechanism involving the induction of ferroptosis and the activation of adaptive immunity. CTD/MM@BSA activates anti-tumor immunity in vivo via STING-ferroptosis induction Given that the ability of CTD/MM@BSA to robustly activate host immune responses through a ferroptosis-immunotherapy axis is critical for its high antitumor efficacy, we initially assessed the levels of ferroptosis-associated markers using WB analysis. Results showed that compared to the MM@BSA, CTD, and control groups, CTD/MM@BSA treatment significantly downregulated GPX4 and SLC7A11 protein expression in tumor tissues (Fig. 8A and Fig. S25), indicating effective promotion of ferroptosis by CTD/MM@BSA. The expression levels of GPX4 and SLC7A11 were further directly examined in tumor tissue sections from each group by immunofluorescence staining. The results revealed that the fluorescence intensities of GPX4 and SLC7A11 in tumor tissues from the CTD/MM@BSA-treated group were markedly lower than those in the control and free drug groups, demonstrating that CTD/MM@BSA triggers ferroptosis in vivo (Fig. S26). Moreover, this trend was highly consistent with the changes in ferroptosis-related pathway proteins detected by WB, further and visually confirming the induction of ferroptosis by CTD/MM@BSA. Further WB examination of proteins associated with the STING signaling cascade indicated that, consistent with expectations, tumors treated with CTD/MM@BSA displayed elevated expression levels of cGAS, P-STING, P-TBK1, and P-IRF3 (Fig. 8B and Fig. S27). ELISA analysis of IFN-β expression in tumor tissues demonstrated that CTD/MM@BSA treatment elevated IFN-β levels to 6.6-fold of the PBS group, significantly exceeding increases observed in the MM@BSA (4.2-fold) and CTD (4.1-fold) groups (Fig. 8C). This suggests CTD/MM@BSA possesses dual capabilities in synergistically inducing ferroptosis and activating the cGAS-STING pathway.Fig. 8Alterations in molecular expression profiles and the immune microenvironment within tumor tissues following therapeutic intervention using various formulations. (A) The levels of GPX4 and SLC7A11 protein expression detected in tumor tissue. (B) The expression of cGAS, P-TBK1, P-STING and P-IRF3 proteins in tumor tissues. (C) The levels of IFN-β within tumor tissues was measured. Flow cytometry analysis of tumor tissues: (D, E) Proportion of CD11c+CD86+ cells; (F, G) Proportion of CD3+CD8+ T cells; (H, I) Proportion of CD3+CD4+ T cells. (J) The expression levels of CD8 and CD4 proteins were quantified within tumor tissue. Changes in cytokine levels in tumor tissues: (K) IL-10, (L) TGF-β, (M) TNF-α and IFN-γ (n = 5). (N) The expression of HMGB1 and CRT in tumor tissues across different treatment groups. Alterations in molecular expression profiles and the immune microenvironment within tumor tissues following therapeutic intervention using various formulations. (A) The levels of GPX4 and SLC7A11 protein expression detected in tumor tissue. (B) The expression of cGAS, P-TBK1, P-STING and P-IRF3 proteins in tumor tissues. (C) The levels of IFN-β within tumor tissues was measured. Flow cytometry analysis of tumor tissues: (D, E) Proportion of CD11c+CD86+ cells; (F, G) Proportion of CD3+CD8+ T cells; (H, I) Proportion of CD3+CD4+ T cells. (J) The expression levels of CD8 and CD4 proteins were quantified within tumor tissue. Changes in cytokine levels in tumor tissues: (K) IL-10, (L) TGF-β, (M) TNF-α and IFN-γ (n = 5). (N) The expression of HMGB1 and CRT in tumor tissues across different treatment groups. Given that increased levels of IFN-β are known to enhance dendritic cell maturation, we evaluated the percentage of mature dendritic cells (mDCs) by flow cytometric analysis. The CTD/MM@BSA showed significantly higher mDCs infiltration versus controls, whereas CTD and MM@BSA exhibited only modest increases (Fig. 8D and E). This effect likely stems from the synergistic interaction between ferroptosis and potent Mn2+-mediated STING pathway activation. Given the pivotal function of DCs in presenting antigens to T cells, the observed enhancement of DCs maturation suggests that CTD/MM@BSA successfully counteracts the immunosuppressive tumor microenvironment to elicit antigen-specific immune activation. Subsequently, the extent of T-cell infiltration was assessed. Our findings revealed that both the CTD/MM@BSA regimen and the CTD/MM@BSA combined with aPD-L1 treatment led to a marked increase in the number of CD8+ T cells infiltrating the tumor tissue. Specifically, these treatments resulted in 8.70- and 10.69-fold elevations, respectively, when compared to the PBS control group. In contrast, the CTD and MM@BSA exhibited substantially lower increases of only 3.39-fold and 3.17-fold (Fig. 8F and G). Similarly, infiltration of CD4+ T cells was markedly elevated in the CTD/MM@BSA and CTD/MM@BSA + aPD-L1 groups, showing 8.30-fold and 12.22-fold increases over PBS, respectively. In contrast, the CTD and MM@BSA exhibited substantially lower increases of only 3.80-fold and 4.60-fold (Fig. 8H and I). The results from immunofluorescence staining for CD4 and CD8 T cell markers (Fig. S28), along with WB assessments of CD4 and CD8 protein levels (Fig. 8J and S29), provided further supporting evidence for these observations. Analysis of the tumor microenvironment revealed that all treatment groups showed an upregulation of the pro-inflammatory cytokines TNF-α and IFN-γ, and a downregulation of the immunosuppressive cytokines IL-10 and TGF-β compared to the PBS. This clearly demonstrates that our nanoparticles successfully activated a local anti-tumor immune response, which is the core mechanism of their therapeutic action. Importantly, these favorable changes were markedly more pronounced in the combination therapy group (Fig. 8K–M and S30). This provides a mechanistic basis for the enhanced efficacy, indicating that the combination more effectively promotes an anti-tumor immune response. Furthermore, increased surface exposure of CRT and reduced intracellular HMGB1 were observed in tumors from CTD/MM@BSA and combination groups (Fig. 8N and S31), demonstrating that coordinated cGAS-STING activation and ferroptosis induction elicit potent immunogenic responses in vivo. To further validate the key roles of ferroptosis and the cGAS-STING signaling pathway in the antitumor efficacy of CTD/MM@BSA, additional groups were established in mice bearing Hepa1-6 tumors: PBS, Ferrostatin-1 (Fer-1, a ferroptosis inhibitor), C-176 (a STING inhibitor), CTD/MM@BSA, CTD/MM@BSA + Fer-1, and CTD/MM@BSA + C-176. Following treatment, tumor growth, DC maturation, and infiltration of CD4+/CD8+ T cells were evaluated. The results showed that, compared with the CTD/MM@BSA, the CTD/MM@BSA + Fer-1 and the CTD/MM@BSA + C-176 exhibited significantly increased tumor volume and weight (Fig. S32A–D). Flow cytometry analysis revealed that in the combination treatment groups, the proportion of mature dendritic cells (CD86+CD11c+) in tumor tissues was significantly reduced, and the infiltration of both CD4+ and CD8+ T cells within the tumors was markedly decreased (Fig. S32E–J). These findings indicate that inhibiting ferroptosis or the STING signaling pathway partially reverses the antitumor effect of CTD/MM@BSA, confirming that induction of ferroptosis and activation of the cGAS-STING pathway are key mechanisms underlying the in vivo antitumor performance of this nanoplatform. Notably, compared with the PBS, the CTD/MM@BSA + Fer-1 and CTD/MM@BSA + C-176 still exhibited some degree of tumor growth inhibition, further demonstrating the therapeutic advantage of the dual-pathway design of CTD/MM@BSA. Anti-lung metastasis effect of CTD/MM@BSA As extrahepatic metastasis deprives many patients of the opportunity for curative treatment, preventing metastasis is a critical objective in the systemic therapy of HCC [44]. Research has indicated that enhancing oxidative stress within cells is an effective strategy for suppressing the spread of tumors. Additionally, immunotherapeutic approaches have shown promise in their ability to combat metastatic processes [45,46]. Given that CTD/MM@BSA enhances ROS levels and induces significant anti-tumor immune responses, we evaluated its anti-metastatic capability. Given the critical role of cancer cell migration in tumor metastasis, we evaluated the suppressive impact of various formulations on Hepa1-6 cell migration through an in vitro scratch wound healing assay. Results showed that scratches in the control group were almost completely healed after 24 h, indicating strong migratory ability of Hepa1-6 cells. In contrast, CTD/MM@BSA treatment significantly inhibited cell migration (Fig. S33). Enhanced tumor migratory capacity is often associated with the activity of enzymes that degrade the extracellular matrix, such as matrix metalloproteinase-9 (MMP-9) and MMP-2 [47,48]. Based on this, we further investigated MMP-9 and MMP-2 expression levels via WB analysis. Results revealed that CTD/MM@BSA treatment significantly downregulated the protein levels of MMP-9 and MMP-2 (Fig. 9A–C), suggesting that CTD/MM@BSA may impede cancer cell metastasis by inhibiting these proteins.Fig. 9Impact of different formulations on invasion-related protein expression in Hepa1-6 cells and lung metastasis. (A) Protein expression levels of MMP9 and MMP2. Quantitative analysis of (B) MMP2 and (C) MMP9 protein expression. (D) A schematic illustration depicting the protocol for generating the mouse lung metastasis model and the corresponding treatment schedule. (E) Histological sections of metastatic lung lesions, stained with hematoxylin and eosin, are shown following therapeutic intervention with various formulations. Scale bar = 1000 μm. Impact of different formulations on invasion-related protein expression in Hepa1-6 cells and lung metastasis. (A) Protein expression levels of MMP9 and MMP2. Quantitative analysis of (B) MMP2 and (C) MMP9 protein expression. (D) A schematic illustration depicting the protocol for generating the mouse lung metastasis model and the corresponding treatment schedule. (E) Histological sections of metastatic lung lesions, stained with hematoxylin and eosin, are shown following therapeutic intervention with various formulations. Scale bar = 1000 μm. To further assess the anti-metastatic potential in vivo, a lung metastasis model was generated through the intravenous administration of Hepa1-6 cells (Fig. 9D). To evaluate metastatic spread, lung tissue samples were harvested from the various treatment cohorts. Histological evaluation using H&E staining was conducted on pulmonary tissue samples obtained from all experimental groups. In the corresponding figures (Fig. 9E), metastatic lesions are demarcated by red dashed lines. Large metastatic foci were evident in the PBS group. The CTD monotherapy group and the MM@BSA group exhibited only a few small residual metastatic foci, while no significant metastatic foci were observed in the CTD/MM@BSA group. These results confirm that CTD itself possesses certain anti-lung metastatic activity, while CTD/MM@BSA likely further enhances this effect through synergistic activation of the immune response. Biosafety evaluation of CTD/MM@BSA Based on the significant therapeutic efficacy demonstrated by the nanoparticles, we systematically evaluated their biosafety in vivo. Throughout the course of the treatment, changes in body weight were regularly recorded. Mice administered CTD/MM@BSA exhibited body weight trends consistent with the PBS control group, with no significant fluctuations or signs of weight loss (Fig. 10A). This suggests that the nanoparticle treatment did not induce systemic toxicity or adversely affect the overall physiological condition of the animals. Serum biochemical analysis further revealed that levels of AST, ALT, BUN, and CRE remained within normal physiological ranges (Fig. 10B–C), confirming the absence of detectable hepatotoxicity or nephrotoxicity caused by the formulations. Histopathological evaluation of major organs via H&E staining demonstrated intact tissue architecture in the heart, liver, spleen, lungs, and kidneys, with no pathological alterations observed (Fig. 10D), thereby excluding systemic toxicity at the tissue level. In many potent immunotherapies, cytokine release syndrome is a frequent and dangerous side effect, characterized precisely by the uncontrolled elevation of inflammatory cytokines in the peripheral blood. To investigate whether CTD/MM@BSA could induce autoimmune responses, we measured serum levels of key inflammatory cytokines (IL-10, TGF-β and TNF-α) in tumor-bearing mice following various treatments using ELISA. Sustained systemic inflammation is a known risk factor for triggering autoimmune disorders. Compared with the PBS group, the levels of cytokines in the treatment groups showed no significant changes (Fig. 10E–G). These findings indicate that although strong immune activation was triggered locally within the tumor, this response was effectively confined to the TME and did not induce a harmful systemic inflammatory storm. Furthermore, we also evaluated the long-term toxicity of CTD/MM@BSA. The results showed that over the entire 12-week observation period, the body weight growth trend of mice in the CTD/MM@BSA was consistent with that of the PBS, with no significant decrease. Serum biochemical tests at week 12 indicated that the levels of ALT, AST, BUN, and CRE in the nanoparticle group showed no statistical difference compared with the PBS. Meanwhile, the histological structure of major organs remained intact, with no obvious pathological changes such as inflammation or necrosis (Fig. S34). Collectively, these findings strongly demonstrate the excellent biocompatibility and favorable safety profile of our nanoparticle system. It should also be noted that BSA, used here as a carrier material to coordinate with Mn2+ and encapsulate CTD, is a foreign protein that may possess certain immunogenic potential. However, BSA-based nanocarriers have been extensively reported to exhibit good biosafety and minimal immune activation in preclinical studies. Consistent with these reports, our results showed no signs of hypersensitivity, accelerated clearance, or systemic inflammation, suggesting negligible immunogenicity of BSA under our experimental conditions. Nonetheless, for future translational consideration, replacing BSA with human serum albumin or albumin-mimetic carriers could further minimize potential immune recognition in clinical applications.Fig. 10In vivo safety evaluation of treatments with different formulations. (A) Body weight fluctuations in mice were monitored throughout the course of the experimental treatment. Post-treatment serum indicators of hepatic and renal functions: (B) ALT and AST levels; (C) BUN and CRE levels (n = 5). (D) H&E-stained tissue sections from key organs of mice are displayed, representing the various formulation groups. Scale bar = 50 μm. (E-G) Changes in serum cytokine levels post-treatment: (E) IL-10, (F) TNF-α, (G) TGF-β (n = 5), ns: not significant. In vivo safety evaluation of treatments with different formulations. (A) Body weight fluctuations in mice were monitored throughout the course of the experimental treatment. Post-treatment serum indicators of hepatic and renal functions: (B) ALT and AST levels; (C) BUN and CRE levels (n = 5). (D) H&E-stained tissue sections from key organs of mice are displayed, representing the various formulation groups. Scale bar = 50 μm. (E-G) Changes in serum cytokine levels post-treatment: (E) IL-10, (F) TNF-α, (G) TGF-β (n = 5), ns: not significant. Conclusion In summary, we constructed a multifunctional nanoplatform (CTD/MM@BSA) that triggers an anti-tumor immune response by synergistically activating the cGAS-STING pathway and inducing ferroptosis. Within the TME, CTD and Mn2+ in CTD/MM@BSA act synergistically to stimulate robust generation of ROS, thereby inducing ferroptosis. Ferroptosis triggers ICD in malignant cells, leading to the emission of DAMPs. In parallel, the combined actions of DNA damage caused by CTD and the heightened sensitivity of cGAS to DNA, potentiated by Mn2+, synergistically intensify the stimulation of the cGAS-STING cascade. This amplified signaling cooperates with ICD to robustly provoke an anti-tumor immune response, resulting in the effective suppression of both tumor progression and metastatic spread. Notably, IFN-γ released by pro-inflammatory immune cells further promotes ferroptosis in tumor cells, establishing a positive feedback loop. Furthermore, combining this nanoplatform with an aPD-L1 significantly enhances its anti-tumor efficacy. Therefore, this strategy of reinforcing the mutual regulation between ferroptosis and anti-tumor immunity paves the way for designing novel immunomodulatory anti-cancer agents. Funding This work was supported by 10.13039/501100004735Hunan Provincial Natural Science Foundation of China (2023JJ30477, 2025JJ60525), Project Supported by Scientific Research Fund of Hunan Provincial Education Department (22B0401, 24B0357), Hunan Province Science and Technology Talent Support Project-Young Outstanding Science and Technology Worker Training Program (2022TJ-N05), 10.13039/100018696Health Commission of Hunan Provence (20256917), and The Project of Hunan Provincial Key Laboratory of Vascular Biology and Translational Medicine (2024XG002). CRediT authorship contribution statement Shengmei Wang: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Writing – original draft. Mengtian Ma: Data curation, Formal analysis. Qiuyan Guo: Formal analysis. Guiming Deng: Investigation. Xinhua Xia: Investigation. Jianyu Li: Data curation. Xueqing Tao: Methodology. Dong Zhong: Formal analysis. Kun Zhang: Funding acquisition, Project administration, Supervision, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.