Triethylamine (TEA, anhydrous and sealed with molecular sieves, 4 A), dichloromethane (DCM, anhydrous) and N, N-Dimethylformamide (DMF, anhydrous and sealed with molecular sieves, 4 A) were purchased from J&K Chemical Reagent Inc (Beijing, China) and used as received without further purification. AIBN (99%, recrystallized twice from ethanol), methacryloyl chloride (95%), hydroxymethylferrocene (98%) were obtained from Macklin Chemical Company (Shanghai, China). Poly (ethylene glycol) methyl ether methacrylate (PEGMA, 475 g/mol) was obtained from Aladdin Chemical Company (Shanghai, China) and passed through a basic aluminum oxide column to remove the inhibitor and stored in the fridge at −22 °C. 4-Cyano-4-(phenylcarbonothioylthio)pentanoic acid (>97%) was obtained from TCI. IR808 was purchased from AbMole (USA). DMSO was purchased from MP (USA). Fetal bovine serum was purchased from Zeta Life (USA). Citrate Antigen Retrieval solution was purchased from BOSTER (Wuhan, China). ROS kit, ATP kit and Hoechst were purchased from Beytime (Shanghai, China). Live Dead Staining Kit was purchased from Biosharp (Wuhan, China). C11 bodipy 581/591 was purchased from Thermo Fisher (USA). PE anti-mouse CD11c, FITC-anti-mouse CD3ε, APC or Alexa Fluro 700 anti-mouse CD8a Recombinant, APC/Cyanine7 or Percp-Cyanine5.5 anti-mouse I-A/I-E Antibody, PE/Cyanine7 anti-mouse CD44, Brilliant Violet 650TM anti-mouse CD62L were purchased from Biolegend (USA). Anti-GPX4 and anti-Ki67 antibody was purchased from Abcam (UK). anti-CRT antibody, anti-HMGB1 antibody, anti-MNT antibody, anti-SAT1 antibody, anti-PD1 antibody were purchased from ProteinTech (Wuhan, China) and anti-NRF2 antibody were purchased from CST (USA). HMGB1 kit was purchased from Novus (USA). DU145, A549 and RM1 cell lines were gained from American Type Culture Collection (USA). Immunohistochemistry kits and DAB staining reagents were purchased from MXB biotechnologies (Fuzhou, China). All animal experiments were approved by the Animal Ethics and Welfare Committee (approval number: IACUC-AEWC-F240730008, IACUC-AEWC-F250601001).

The preparation of ferrocenylmethyl methacrylate (FMMA)

In a flame-dried 100 ml Schlenk flask, hydroxymethylferrocene (2.16 g, 10 mmol, 1 equiv) was dissolved in anhydrous DCM (50 ml) and triethylamine (1.21 g, 12 mmol, 1.2 equiv) was added. Methacryloyl chloride (1.25 g, 12 mmol, 1.2 equiv) dissolved in DCM (15 ml) was added dropwise over 30 min at 0 °C and stirred overnight at room under Argon atmosphere. The organic solvent was transferred into a separatory funnel and washed with 10% sodium bicarbonate (150 ml for 3 times) and brine (100 ml for 1 time), collected the organic solvent and dried with anhydrous sodium sulfate. The crude product as yellow powder was obtained after the removal of the DCM. In order to remove the inhibitor in the methacryloyl chloride, the crude product was redissolved in DCM and passed through a basic aluminum oxide column with DCM. After removal of the DCM, a transparent, orange viscous oil was obtained. After placing the product in the refrigerator, an orange-yellow solid (2.4 g, 85%) was obtained and stored at −22 °C.

General synthesis of PPEGMA-b-PFMMA (PF) via reversible addition-fragmentation chain transfer (RAFT) copolymerization

In a nitrogen atmosphere glover-box, PEGMA (2.375 g, 5 mmol, 15 equiv), 4-cyano-4-(phenylcarbonothioylthio) pentanoic acid (0.093 g, 0.3 mmol, 1 equiv) as chain transfer agent (CTA) and AIBN (5 mg, 0.03 mmol, 0.1 equiv) were weighted and added in a 20 ml Schlenk flask containing 5 ml DMF. The glass vial was sealed with rubber stopper and placed in an oil bath at 70 °C. After 16 h, the reaction was quenched by putting the Schlenk flask into ice-water and opening the rubber stopper, exposing it to air. Then transfer the mixture into a dialysis tube (MWCO: 1000 Da) to remove the DMF. After lyophilization, the PPEGMA was obtained as a red viscous-liquid.

Then, PPEGMA as macroinitiator for the RAFT polymerization of FMMA. In a typical example, PPEGMA-CTA (0.2 g, 0.028 mmol) and different weights (0.10 g, 0.35 mmol), (0.2 g, 0.70 mmol) and (0.25 g, 0.88 mmol) of FMMA were weighted and dissolved in a 10 ml Schlenk flask with 3 ml DMF, respectively. 10 mg AIBN was weighed and dissolved into 5 ml DMF (2 mg/ml), then 1 ml of the AIBN stock solution was added into each Schlenk flask. The glass vials were sealed with rubber stopper. Then, the oxygen was removed through a Freeze-Pump-Thaw method. In brief, the Schlenk flasks were connected to the Schlenk flasks line, and the solvent is frozen into a solid state using liquid nitrogen. A high-vacuum pump is then applied to evacuate the system for several minutes. Afterward, the vacuum system is closed, and nitrogen was introduced into the flask. The solvent is allowed to thaw gradually under the nitrogen atmosphere. This cycle-freezing, evacuating, and thawing is repeated three times to ensure thorough removal of dissolved oxygen. After that, the Schlenk flask was placed in an oil bath at 70 °C, the RAFT polymerization lasted for 12 hours, then the reaction was quenched by immersing the glass vial into the liquid nitrogen. The DMF was removed by dialysis (MWCO: 3000 Da) against DIW for 2 days. After lyophilization, the PPEGMA-b-PFMMA were obtained as brown solid.

Characterization

FT-IR spectra of all samples were obtained in a transmission mode on a Perkin-Elmer Paragon1000 spectrometer under ambient condition. Samples were ground with KBr and then compressed into pellets. The spectra were taken at the frequencies range from 500 to 4000 cm−1. Typically, 32 scans at a resolution of 8 cm−1 were accumulated to obtain one spectrum. 1H NMR spectra measurements were executed on a Bruker NMR instrument (Bruker AV-500), using deuterated chloroform (CDCl3-d) or deuterated dimethyl sulfoxide (DMSO-d6) as solvent and tetramethylsilane (TMS) as an internal standard. The thermogravimetric analysis (TGA) was performed on a TA Instruments Q500. The heating range was from 25 °C to 800 °C at a constant rate of 10 °C/min. All the analysis was carried out under a high-purity nitrogen gas (flow rate: 40 ml/min) atmosphere.

Synthesis and characterization of PF NPs and P8D NPs

The blank PF NPs were prepared via nanoprecipitation method. Briefly, 10 mg of PF was dissolved in 1 ml DMSO. Using a 1000 μL pipette, the organic solution was slowly added into 5 ml of DIW over two minutes with continuous stirring at 1000 rpm. The resulting mixture was then stirred for 2 h and subsequently purified by dialysis (MWCO: 3500 Da) against DIW for 24 h to remove organic solvents.

To prepare the drug/ photosensitizer-loaded NPs, PF polymer, IR808, and Doc at predetermined mass ratios were dissolved in 1 ml of DMSO. Then the mixture was slowly added into 5 ml of DIW over 2 min with continuous stirring at 1000 rpm. The crude NPs solution was dialyzed (MWCO: 3500 Da) against DIW for 72 h, with water changes every 8 h, to facilitate NPs self-assembly and purify the final product. After dialysis, transfer the solution to sterile tubes and measure the concentrations of IR808 and Doc. Calculate the drug LE and drug LC using the following formulas:

where: A = Mass of the loaded IR808 or Doc; B = Mass of PF added to the solution; C = Mass of IR808 or Doc initially added to the solution

Drug release efficiency of P8D NPs

Divide the prepared P8D NPs into four groups (2 ml each) and load them into dialysis bags. Place the dialysis bags into beakers containing 50 ml of 10% fetal bovine serum (FBS) containing medium with different concentrations of H2O2 (0, 100 μM, 500 μM, and 1000 μM). Incubate the beakers in a 37 °C thermostatic shaking water bath. At specified time points (0, 4, 12, 24, 48, 72 and 96 h), collect 2 ml of the external solution and replenish with 2 ml of fresh medium containing the corresponding H₂O₂ concentrations (0, 100 μM, 500 μM or 1000 μM). Measure the concentration of IR808 in the collected samples using a fluorescence microplate reader (Varioskan Flash, Thermo, USA).

Photothermal conversion properties of P8D NPs

Firstly, 200 μL of PBS, Doc, PD, IR808, P8 or P8D NPs (IR808 concentration: 50 μg/ml) were added to 96-well plate, with 3 wells per group. Irradiate with an 808 nm NIR laser (PURI Materials, China) for 5 min and use a photothermal imaging system (FOTRIL 22OS, Shanghai, China) to monitor and record temperature changes in each group. Secondly, P8D NPs with varying IR808 concentrations (0 μg/ml to 50 μg/ml) were added to a 96-well plate (3 wells per group). Irradiate with the 808 nm NIR laser for 5 min and record temperature changes of P8D NPs at different concentrations. Thirdly, P8D NPs (IR808 concentration: 50 μg/ml) was added to a 96-well plate. Irradiate with the 808 nm NIR laser (Optical power density: 0.4 W/cm2 to 1.2 W/cm2) for 5 min, temperature changes of P8D NPs were recorded at different optical power density. Finally, P8D NPs (IR808 concentration: 50 μg/ml) was added to a 96-well plate. Irradiate with the 808 nm NIR laser for 5 min, allow the samples to cool naturally, then reheat them. Repeat this heating-cooling cycle for a total of 3 times and monitor temperature changes of P8D NPs using the photothermal imaging system.

Cellular uptake

1 × 10⁵ cancer cells (DU145, A549, RM1) and BPH1 were seeded in confocal dishes. After cell adhesion, 10% serum-containing medium supplemented with PBS, Doc, IR808, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 4 μg/ml) was added to each group. Following 8 hours of incubation, the supernatant was discarded and cells were washed three times with PBS, fixed with 4% paraformaldehyde for 15 min, washed three times with PBS again, incubated with 200 μL of DAPI-containing antifade mounting medium and imaged using laser scanning confocal microscopy (LSCM).

Cell viability assay

3 × 10³ cancer cells were seeded in a 96-well plate. After cell adhesion, 10% serum-containing medium supplemented with PBS, Doc, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 0–2.42 μg/ml; Doc concentration: 0–40 ng/ml) was added. Following 4 h of incubation, the P8 NPs and P8D NPs groups were irradiated with NIR laser (Optical power density: 0.8 W/cm2) for 5 min. Cells were further incubated for 24 h and cell viability was assessed using a CCK-8 kit and a microplate reader. For the investigation of ferroptosis inhibitors’ effects on cell viability, NAC (3 mM), DFO (10 μM) or FER1 (3 μM) were co-administered with P8D NPs. All subsequent steps were performed as before. The CI was calculated using the following formulas:

$$\mathrm=\frac_}_}+\frac_}_}$$

where D1 and D2 represent the doses of Doc and IR808 in the combination therapy respectively, while Dx1 and Dx2 are the doses of Doc or IR808 required to achieve the same effect when used alone.

Live/dead staining

3 × 10³ cancer cells were seeded in 6-well plates. After cell adhesion, 10% serum-containing medium supplemented with PBS, Doc, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml) was added. Following 4 h of incubation, P8 NPs and P8D NPs groups were irradiated with NIR laser for 5 min and further cultured for 24 h. Fluorescent imaging was performed using fluorescence microscopy while dead cells were calculated via Image J.

RNA-sequencing

A549 cells in the logarithmic growth phase were seeded in 10 cm dishes. After cell adhesion, the cells were treated with PBS or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml). The P8D NPs group was irradiated with 808 nm NIR laser for 5 min (Optical power density: 0.8 W/cm²). After 24 h, cells were collected and total RNA was extracted using a commercial service (Tsingke Biotechnology). RNA sequencing reads were aligned to the reference genome using the Hisat2 software, followed by differential gene expression analysis with DESeq2. Functional enrichment analyses, including Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses, were performed using the Xiantao Academic platform.

Transmission electron microscopy (TEM) Analysis

DU145 and A549 cells were seeded in 10 cm dishes. After cell adhesion, the cells were treated with PBS, Doc, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml). Following a 4-hour incubation, the P8D NPs-treated group was irradiated with 808 nm NIR laser (Optical power density: 0.8 W/cm²) for 5 min. After an additional 24 h of culture, cells were fixed with 2.5% paraformaldehyde for 24 h and 1% osmium tetroxide for 1 h. The samples were then dehydrated through a graded series of ethanol. The cells were embedded with Epon812 and sectioned by microtome (Leica, Germany), followed by double-staining with uranyl acetate and lead citrate. The stained sections were observed and imaged using a HITACHI TEM (HT7800, Tokyo, Japan).

ROS and lipid peroxidation detection

DU145, A549, and RM1 cells were seeded in confocal dishes. After cell adhesion, the cells were treated with PBS, Doc, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml). Following a 4-h incubation, P8 NPs and P8D NPs groups were irradiated with 808 nm NIR laser for 5 min (Optical power density: 0.8 W/cm²) and further cultured for 4 h. Subsequently, the cells were washed three times with PBS, incubated with DCFH-DA and BODIPY 581/591 C11 probes for 30 min, washed again three times with PBS and mounted in Hoechst-containing antifade mounting medium. Fluorescence imaging was performed using a confocal fluorescence microscope. For the investigation of ferroptosis inhibitors’ effects on LPO, FER1 (3 μM) was co-administered with P8D NPs. All subsequent steps were performed as before.

Western blot analysis

DU145 and A549 cells were seeded in 6-well plates. After cell adhesion, the cells were treated with PBS, Doc, PD NPs, P8 NPs, or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml). Following 4 h incubation, the cells were irradiated with 808 nm NIR laser (Optical power density: 0.8 W/cm²) for 5 min and further cultured for 10 h. Total protein was extracted using RIPA lysis buffer supplemented with protease inhibitors. Protein levels of GPX4, NRF2, MNT, and SAT1 were analyzed by Western blotting.

Damage-associated molecular patterns detection

DU145, A549, and RM1 cells were seeded in 6-well plates. After cell adhesion, the cells were treated with PBS, Doc, PD NPs, P8 NPs or P8D NPs (IR808 concentration: 1 μg/ml; Doc concentration: 16.56 ng/ml). Following 4 h incubation, the cells were irradiated with 808 nm NIR laser (Optical power density: 0.8 W/cm²) for 5 min and further cultured. HMGB1 within the cells was assessed at multiple time points (8, 12, or 16 h). For CRT and HMGB1 protein detection in cells. Cells were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton™ X-100 and blocked with 5% fetal bovine serum (FBS) for 60 min. Subsequently, cells were incubated with anti-CRT and anti-HMGB1 antibodies at 4 °C overnight. After washing three times with PBS, cells were incubated with fluorescent secondary antibodies for 2 h, washed again three times with PBS and stained with DAPI. Fluorescence imaging was performed using a confocal fluorescence microscope. For ATP and HMGB1 in culture supernatant. Cell culture supernatants were collected after 8 h incubation with PBS, Doc, PD NPs, P8 NPs or P8D NPs and NIR irradiation, centrifuged and analyzed for ATP and HMGB1 levels using commercial assay kits following the protocols. For the investigation of ferroptosis inhibitors’ effects on ICD, NAC (3 mM), DFO (10 μM), or FER1 (3 μM) were co-administered with P8D NPs. All subsequent steps were performed as before.

In vivo fluorescence imaging and photothermal imaging

Fluorescence Imaging: Subcutaneous tumor-bearing nude mice were intravenously injected with 100 μL of PBS, IR808, or PF@(IR808+Doc) (IR808 concentration: 5 µg/ml) via the tail vein. Fluorescence images were acquired at 0, 1, 4, 8, 24, 48, 7,2 and 96 h post-injection using a ChemiDoc™ MP imaging system (Bio-Rad, USA). At 96 h, mice were euthanized by cervical dislocation. Tumor tissues and major organs were harvested for fluorescence imaging. Photothermal Imaging: For photothermal imaging, 100 μL of PBS, IR808 or P8D NPs (IR808: 2 mg/kg) was administered via tail vein injection. Tumor regions were irradiated with NIR laser (Optical power density: 0.8 W/cm²) and photothermal images were captured using an infrared thermal camera. Temperature changes in the tumors were recorded in real-time.

Anti-tumor efficacy in vivo

For the nude mouse subcutaneous xenograft model: 4-week-old male Balb/c nude mice were subcutaneously injected with 1 × 107 DU145 cells. For the C57 mouse tumor model: Four-week-old male C57 mice were subcutaneously injected with 1 × 106 RM1 cells. When tumor volume reached approximately 60 mm³, the mice were randomly divided into six groups (n = 6 each): Ctrl, Doc, PD, P8D, P8-L, P8D-L. Each group received tail vein injections of 200 μL PBS, Doc, PD, P8, or P8D (IR808: 2 mg/kg), respectively. On the following day, tumor tissues in the P8-L and P8D-L groups were irradiated with NIR laser (Optical power density: 0.8 W/cm²). This protocol constituted one treatment cycle, with a total of 4 cycles administered in Balb/c nude mice model and a total of 2 cycles administered in C57 mouse tumor model. All mice were euthanized by cervical dislocation one week after the final treatment. Tumor tissues and organs were harvested, with TDLNs isolated from C57 mice for immune cell extraction and flow cytometry analysis.

For the abscopal effect model: C57 mice received 1 × 105 RM1 cells in the right hind limb. When the right limb tumor volume reached 60 mm³, the mice were randomly divided into six groups (n = 6 each): Ctrl, Doc, PD, P8D, P8-L, P8D-L. Each group received tail vein injections of 200 μL PBS, Doc, PD, P8 or P8D (IR808: 2 mg/kg), respectively. NIR irradiation was performed on right tumor tissues of P8-L and P8D-L groups the next day. Two treatment cycles were administered. One week after the final administration, 5 × 105 RM1 cells were rechallenged in the left hind limb. Tumor volume and body weight were monitored. All mice were euthanized by cervical dislocation two weeks after the final treatment.

For the immune memory model: RM1 xenografts were first established in the right hind limb of C57 mice (1 × 106 cells). When right limb tumors reached 60 mm³, mice were divided into two groups: Ctrl and P8D-L. Tail vein injections of 200 μL PBS or P8D (IR808: 2 mg/kg) were administered respectively. P8D-L group received NIR irradiation the next day. Two treatment cycles were completed. After complete regression of right limb tumors in the P8D-L group, mice were maintained for 21 days and the left limb was received 5 × 105 RM1 cells. Tumor volume was monitored and all mice were euthanized by cervical dislocation one week after the final treatment.

For lung metastasis model: Four-week-old male C57 mice were subcutaneously injected with 1 × 106 RM1 cells. When tumor volume reached approximately 60 mm3, the mice were randomly divided into six groups (n = 8): Ctrl, Doc, PD, P8D, P8-L, P8D-L. Each group received tail vein injections of 200 μL PBS, Doc, PD, P8 or P8D (IR808: 2 mg/kg), respectively. On the following day, tumor tissues in the P8-L and P8D-L groups were irradiated with NIR laser (Optical power density: 0.8 W/cm²). Two treatment cycles were administered. One week after the final administration, to establish lung metastases, 5 × 105 fluorescently labeled RM1 cells were administered intravenously via the tail vein. On days 6, 9, and 12 after model establishment, the mice were subjected to in vivo imaging to monitor metastasis. Survival rates were recorded throughout the experiment. On day 30, lung tissues were harvested for H&E staining analysis.

Flow cytometry analysis

We isolated immune cells from tumor tissues, spleen, and TDLNs using established protocols from the literature.63 DCs were labeled with anti-CD11c and anti-MHCII antibodies, while CD8+ T cells were identified using anti-CD3 and anti-CD8 antibodies. Central memory T cells and effector memory T cells were identified with anti-CD3, anti-CD8, anti-CD44, and anti-CD62L antibodies. Gating strategy of flow cytometry analysis is shown in Supplementary Fig. 17.

Immunohistochemical (IHC) analysis

Tumor tissue sections from each group underwent IHC staining to analyze the expression of PCNA, GPX4, PD-1, and CD8a using ImageJ and GraphPad Prism 8.

Biosafety analysis

For chronic toxicity analysis: 12 mice were divided into four groups (n = 3 per group): Ctrl, Doc, PD, P8D. Each group received a single tail vein injection of 100 μL of the following formulations: PBS, Doc, PD NPs, P8 NPs, or P8D NPs, 24 h post-injection. After the four cycles, mice were monitored for 14 days. For acute toxicity analysis: 28 mice were divided into four groups (n = 7 per group): Ctrl, Doc, PD, P8D. Each group received a single tail vein injection of 100 μL of the following formulations: PBS, Doc, PD NPs, P8 NPs, or P8D NPs. After the four cycles, mice were monitored for 60 days, the weight of mice was recorded. All mice were euthanized and major organs (heart, lung, liver, kidneys, spleen) were harvested. Hematoxylin and eosin (H&E) staining was performed on organ tissues, followed by histopathological analysis to evaluate potential nanodrug-induced toxicity.

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8. For comparisons between two groups, an independent samples t-test was applied, while one-way ANOVA was used for multi-group comparisons. Results are expressed as mean ± standard deviation (SD), with statistical significance defined as P < 0.05. Image analysis of protein expression, fluorescence imaging, and IHC was conducted using ImageJ software.