Carrier‐Free Nanoparticles Enhance Photothermal–Immune Therapy via Metabolic Reprogramming in Triple‐Negative Breast Cancer

ABSTRACT

A carrier‐free co‐assembled nanoplatform, designated as IR@PF‐M NPs, was developed to enhance photothermal‐immunotherapy against triple‐negative breast cancer (TNBC) by reprogramming methionine metabolism. The nanoplatform was constructed through the self‐assembly of the methionine adenosyltransferase 2A (MAT2A) inhibitor PF9366 with the near‐infrared (NIR)‐absorbing dye IR808, followed by surface modification with DSPE‐PEG2000‐Met. This formulation exhibited dual targeting capabilities, enabling tumor accumulation via the enhanced permeability and retention effect and selective uptake by tumor cells through overexpressed methionine transporters, which together enhanced intratumoral accumulation and photothermal efficiency. Upon NIR irradiation, IR808 generated potent photothermal effects with intratumoral temperature beyond 50°C that induced immunogenic cell death (ICD) and transformed the tumor microenvironment into an immune‐activated (“hot”) state. Simultaneously, PF9366 disrupted methionine metabolism, further amplifying ICD and suppressing TNBC progression and metastasis. By integrating photothermal therapy with metabolic intervention, IR@PF‐M NPs demonstrated superior antitumor efficacy and immunomodulatory activity, offering a promising therapeutic strategy for TNBC management.

A methionine‐guided co‐assembled nanoplatform integrates IR808‐mediated photothermal therapy with PF9366‐driven MAT2A inhibition for triple‐negative breast cancer. Targeted tumor uptake and NIR‐triggered hyperthermia cooperate with methionine metabolic disruption to amplify immunogenic cell death, convert an immunosuppressive microenvironment into an immune‐active state, and achieve potent antitumor and antimetastatic effects.

1. Introduction

Breast cancer (BC) remains one of the most prevalent malignancies among women and ranks as the second leading cause of cancer‐related mortality worldwide [1, 2, 3]. Based on immunohistochemical profiling, BC is typically categorized into five molecular subtypes: luminal A, luminal B, HER2‐overexpressing, basal‐like, and normal‐like [4]. Among these, basal‐like breast cancer—commonly referred to as triple‐negative breast cancer (TNBC)—accounts for approximately 15%–20% of all breast cancer cases [5, 6]. TNBC is characterized by the absence of estrogen receptor, progesterone receptor, and HER2 expression [7]. This distinct molecular phenotype is associated with aggressive clinical progression, poor prognosis, limited response to conventional therapies, and a lack of effective targeted treatment options [8]. Although novel agents such as the PARP inhibitor olaparib, the antibody‐drug conjugate sacituzumab govitecan, and the immune checkpoint inhibitor pembrolizumab have shown promise in clinical studies, the incidence, recurrence, and mortality of TNBC remain unacceptably high [9, 10]. Thus, there is an urgent need for more effective and durable therapeutic strategies.

Photothermal therapy (PTT) has emerged as a promising approach for cancer treatment, particularly in light of the unique features of the tumor microenvironment (TME). Due to hypoxia and aberrant vascularization, cancer cells exhibit increased sensitivity to thermal stress relative to normal tissues [11, 12, 13, 14, 15, 16]. PTT employs photothermal agents (PTAs) that preferentially accumulate in tumors and are activated by near‐infrared (NIR) irradiation to induce localized hyperthermia, effectively ablating tumor cells while minimizing damage to surrounding healthy tissues [17]. Beyond direct cytotoxicity, PTT can trigger immunogenic cell death (ICD), characterized by the release of damage‐associated molecular patterns (DAMPs), which in turn stimulate innate and adaptive immune responses [18, 19]. This process facilitates the transformation of the tumor from an “immune‐cold” to an “immune‐hot” phenotype, thereby enhancing immunogenicity and antitumor immunity [20, 21]. However, PTT alone may also activate compensatory mechanisms, such as the upregulation of immunosuppressive surface proteins, which contribute to tumor immune escape, recurrence, and metastasis [17]. Consequently, combining PTT with immunomodulatory strategies holds great potential to amplify its therapeutic efficacy, particularly in immunologically “cold” tumors like TNBC.

Metabolic reprogramming is a hallmark of malignancy and plays a key role in shaping the immunosuppressive TME, thereby promoting tumor progression and resistance to therapy [21, 22, 23, 24, 25, 26]. Targeting tumor‐specific metabolic pathways offers an attractive strategy to reverse immune evasion and recondition the TME [24, 27, 28]. Among various metabolic vulnerabilities, methionine (Met) metabolism has garnered increasing attention due to the heightened dependence of tumor cells on exogenous Met relative to normal cells [29, 30, 31]. Methionine fuels critical cellular processes, including methylation reactions for epigenetic regulation, maintenance of redox balance, and polyamine biosynthesis. Dietary methionine restriction has been explored as a potential anticancer strategy, showing synergistic effects with chemotherapy and radiotherapy in both preclinical and clinical studies [32, 33, 34]. Notably, methionine depletion has been reported to suppress the differentiation of regulatory T cells (Tregs) and enhance the infiltration of cytotoxic CD8⁺ T cells into tumors, thereby relieving immunosuppression and boosting antitumor immunity [35]. These findings support the therapeutic potential of methionine metabolic intervention in immunologically resistant tumors such as TNBC [36, 37, 38]. However, the integration of methionine metabolism targeting with PTT remains underexplored.

In this study, we designed a co‐assembled nanoplatform to synergistically modulate methionine metabolism and enhance photothermal‐immunotherapy in TNBC (Scheme 1). Specifically, methionine adenosyltransferase 2A (MAT2A)—a rate‐limiting enzyme in the methionine cycle—was selected as a target for metabolic inhibition [39]. The MAT2A inhibitor PF9366 was co‐assembled with a NIR‐absorbing photothermal dye, IR808 (a heptamethine cyanine), and the resulting nanoparticles were surface‐functionalized with DSPE‐PEG2000‐Met. This engineered nanoplatform, termed IR@PF‐M NPs, not only exploited the enhanced permeability and retention effect for tumor accumulation but also leveraged the overexpression of methionine transporters on cancer cells to achieve active cellular uptake [40, 41, 42]. Upon NIR irradiation, IR808 generated local hyperthermia with high photothermal conversion efficiency, inducing potent ICD and promoting an immune‐hot TME [43]. Simultaneously, PF9366 disrupted methionine metabolism, further relieving immunosuppression and enhancing the ICD response [44, 45]. Altogether, this photothermal‐metabolic synergistic approach achieved significant inhibition of TNBC growth and metastasis, offering a promising strategy for overcoming therapeutic resistance and immune escape in TNBC.

SCHEME 1.

Schematic illustration of the design, targeting mechanism, and therapeutic effects of IR@PF‐M nanoparticles for triple‐negative breast cancer treatment. The co‐assembled nanoplatform IR@PF‐M NPs was constructed by the self‐assembly of the photothermal agent IR808 and the methionine adenosyltransferase 2A (MAT2A) inhibitor PF9366, followed by surface functionalization with DSPE‐PEG2000‐Met. Upon intravenous injection, the nanoparticles accumulate at the tumor site via the enhanced permeability and retention effect, and are further taken up by tumor cells through overexpressed methionine transporters. Under near‐infrared (NIR) irradiation, IR808 generates localized hyperthermia to induce tumor cell damage and promote immunogenic cell death (ICD). Meanwhile, PF9366 disrupts methionine metabolism to enhance ICD and further inhibit tumor growth. The combination of photothermal ablation and metabolic intervention reprograms the tumor microenvironment into an immune‐active state, characterized by dendritic cell (DC) maturation, activation of cytotoxic CD8⁺ T cells, and repolarization of tumor‐associated macrophages (TAMs) toward the M1 phenotype. These synergistic effects result in robust antitumor efficacy against triple‐negative breast cancer.

2. Materials and Methods

2.1. Materials

1,2‐distearoyl‐sn‐glycero‐3‐phosphoethanolamine‐N‐[hydroxyl (polyethylene glycol)‐2000] (DSPE‐PEG2000) was purchased from AVT Pharmaceutical Tech Co., Ltd. 1,2‐distearoyl‐sn‐glycero‐3‐phosphoethanolamine‐N‐[hydroxyl (polyethylene glycol)‐2000]‐methionine (DSPE‐PEG2000‐Met) was purchased from Xian Ruixi Biological Tech Co., Ltd. 1‐(5‐Carboxypentyl)‐2‐(2‐(3‐(2‐(1‐(5‐carboxypentyl)‐3,3‐dimethylindolin‐2‐ylidene)ethylidene)‐2‐chlorocyclohex‐1‐en‐1‐yl)vinyl)‐3,3‐dimethyl‐3H‐indol‐1‐ium bromide (IR808, Cas no.: 172971‐76‐5) and 7‐Chloro‐N,N‐dimethyl‐5‐phenyl‐[1,2,4]triazolo[4,3‐a]quinoline‐1‐ethanamine (PF9366, Cas: 72882‐78‐1) were purchased from Macklin. Protease inhibitors, BCA protein assay kit, ECL chemiluminescence kit, antifade mounting medium (with DAPI), Calcein/PI cell viability/cytotoxicity assay kit, and ATP content detection kit were purchased from Shanghai Beyotime Biotechnology Co., Ltd. Acrylamide, buffers, and ammonium persulfate were purchased from Beijing Solarbio Science & Technology Co., Ltd. Mouse S‐adenosylmethionine (SAM) ELISA kit, Mouse Interleukin‐12 p70 (IL‐12 p70) ELISA Kit, and Mouse Interferon‐gamma (IFN‐γ) ELISA Kit were purchased from LunChangShuo Biotechnology Co., Ltd. Mouse inducible Nitric Oxide Synthase (iNOS) ELISA Kit was purchased from Shanghai Enzyme‐linked Biotechnology Co., Ltd. (China). Mouse Spermidine ELISA Kit was purchased from Wuhan Fine Biotech Co., Ltd. (China). HMGB1 (High Mobility Group Box‐1 protein) detection kit was obtained from Yantai Boyun Biological Technology Co., Ltd. GAPDH antibody and CRT (Calreticulin) antibody were obtained from Shanghai Abcam Trading Co., Ltd. MAT2A (S‐adenosylmethionine synthase isoform type‐2) antibody, TNF‐α (Tumor Necrosis Factor‐alpha) antibody, and HSP70 (Heat Shock Protein 70) antibody were purchased from Affinity Biosciences. CD86 antibody, CD206 antibody, CD3 antibody, CD4 antibody, CD8 antibody, and FOXP3 antibody were obtained from Boster Biological Technology Co., Ltd.

2.2. Preparation and Characterization of Nanoparticles

A one‐step nanoprecipitation method was employed to synthesize the nanoparticles, and all procedures were carried out under dark conditions. Initially, the optimal molar ratios of IR808 and PF9366 for co‐assembly were screened. These two substances were separately dissolved in dimethyl sulfoxide (DMSO) at a concentration of 1 mM to prepare stock solutions. Then, they were mixed at volume ratios of 1:1, 2:1, 3:1, 4:1, and 5:1 (IR808:PF9366). Subsequently, DSPE‐PEG₂₀₀₀ was added at a weight ratio of 20%. The resulting mixture was slowly dropped into ultrapure water while being stirred at 500 revolutions per minute (rpm) for 10 min. After that, the mixture was dialyzed against ultrapure water at 4°C for 4 h to remove the organic phase. The particle sizes and Polydispersity Index (PDI) of the samples were measured to identify the optimal molar ratio for the co‐assembly of the two molecules. The obtained nanoparticles were designated as IR@PF NPs. Additionally, DSPE‐PEG₂₀₀₀‐Met was used to prepare IR@PF‐M NPs following the same formulation ratio and procedures. The surface morphology of the nanoparticles was characterized using a transmission electron microscope. The encapsulation efficiency of the nanoparticle

To determine the IR808 and PF9366 content in the nanoparticles, the concentration of IR808 was quantified using a multimode microplate reader via fluorescence spectroscopy (λ_ex = 808 nm, λ_em = 830 nm), while the PF9366 amount was determined using a validated high‐performance liquid chromatography (HPLC) method (UV detection at 230 nm). The drug loading capacity (DL, %) and encapsulation efficiency (EE, %) were calculated according to the following equations:

$$\mathrm{DL}\left(\%\right)=\frac{\mathrm{Mass}\mathrm{of}\mathrm{loaded}\mathrm{drug}}{\mathrm{Total}\mathrm{mass}\mathrm{of}\mathrm{nanoparticles}}$$

$$\mathrm{EE}\left(\%\right)=\frac{\mathrm{Mass}\mathrm{of}\mathrm{loaded}\mathrm{drug}}{\mathrm{Mass}\mathrm{of}\mathrm{feeding}\mathrm{drug}}$$

where the “Total weight of nanoparticles” refers to the cumulative mass of the loaded drugs and the carrier materials, including IR808, PF9366, and DSPE‐PEG2000/DSPE‐PEG2000‐Met.

To evaluate the long‐term storage stability, the nano‐assemblies were stored at 4°C for 14 days, during which their hydrodynamic diameters and polydispersity index (PDI) were monitored periodically. Furthermore, the physiological stability was assessed by incubating the nanoparticles in PBS containing 10% fetal bovine serum (FBS) at 37°C. The changes in particle size were recorded at predetermined time intervals to simulate the behavior of the nanoplatform under physiological conditions. Following the stability tests, the drug leakage rate was determined to evaluate the structural integrity of the nanoparticles. Briefly, the samples were transferred to centrifugal filter units (MWCO: 3000 Da) and centrifuged to separate the leaked free drugs from the nano‐assemblies. The concentrations of IR808 and PF9366 in the filtrate were then quantified. The drug leakage rate (%) was calculated using the following equation:

$$\mathrm{Leakage}\mathrm{rate}\left(\%\right)=\frac{\mathrm{Mass}\mathrm{of}\mathrm{drug}\mathrm{in}\mathrm{filtrate}}{\mathrm{Total}\mathrm{mass}\mathrm{of}\mathrm{loaded}\mathrm{drug}}$$

To probe the assembly mechanism of the nanoparticle. The ligand models were constructed using ChemDraw and subjected to energy minimization and geometric optimization using the MM2 module within Chem3D software to obtain the optimal 3D conformations. The refined structures were saved in SDF format as receptors for subsequent docking. The structure of IR808 was retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/), and its energy was similarly minimized using Chem3D before docking. The processed IR808 was defined as the small‐molecule ligand, while the self‐constructed compound models served as the target receptors. Molecular docking and conformational optimizations were performed using the Dock module of MOE 2019.01 software. Initially, the receptor models were pre‐processed and subjected to tethered energy minimization under the Amber 14: EHT force field. For the binding site definition, due to the absence of pre‐determined active pockets, a Blind Docking strategy was employed, where the entire molecular surface of the receptor was designated as the potential binding site. An Induced Fit protocol was applied during the docking process to fully account for the conformational flexibility between the receptor and ligand. Regarding the energy calculation and scoring method, the initial placement of ligand conformations was performed using the Triangle Matcher algorithm, followed by a primary evaluation via the London dG empirical scoring function. Subsequently, a rigid‐body refinement of the poses was conducted using the Forcefield method, followed by a secondary assessment with the London dG function. A total of 300 docking poses were generated for each ligand, and the final binding free energies were calculated using the GBVI/WSA dG algorithm based on Generalized Born Volume Integral/Weighted Surface Area theory. The top five docking poses with the most favorable GBVI/WSA dG scores were selected for further interaction analysis and visualization, which were conducted using MOE 2019.01 and PyMOL 2.1. Detailed docking parameters are summarized in Table S1.

To investigate the optical properties of the nano‐assemblies, the IR808, PF9366, DSPE‐PEG2000, and DSPE‐PEG2000‐Met before and after co‐assembly were prepared in DMSO and ultrapure water (as IR@PF NPs and IR@PF‐M NPs) at 12 µg/mL. Then, the UV absorption spectra were recorded from 400–1000 nm, while the fluorescence emission spectra were recorded after emission at the maximum absorption wavelength of the IR808. Pure water was used as the blank solvent.

To evaluate the photothermal heating capability, IR808 alone, IR@PF NPs, and IR@PF‐M NPs were dispersed in PBS at a concentration of 12 µg/mL (equivalent to the IR808 concentration). The samples were irradiated with an 808 nm laser at a power density of 0.5 W cm^{− 2} for 5 min, during which the temperature variations and real‐time infrared thermal images were recorded using an infrared thermal camera at 30 s intervals. Furthermore, the samples were subjected to three consecutive laser‐on/off cycles. In each cycle, the samples were irradiated for 5 min, followed by a 10 min cooling period. The consistency of the temperature increments across the three cycles was used to determine the photostability of the nano‐assemblies compared to free IR808.

2.3. Cell Culture

Mouse triple‐negative breast cancer cells (4T1), murine mammary epithelial cells (HC11), and monocyte macrophage cells (RAW264.7) were obtained from the cell bank of the Chinese Academy of Sciences Type Culture Collection (Shanghai, China). All cell lines were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 50 units/mL streptomycin, and 100 units/mL penicillin. The cells were incubated at 37°C in a 5% CO₂ atmosphere.

2.4. Cellular Uptake

The 4T1 cells were seeded in 12‐well plates at a density of 4 × 10⁴ cells/well. After incubation for 24 h, IR808, IR@PF NPs, and IR@PF‐M NPs (equivalent to 3 µg/mL IR808) were added, and the cells were further incubated for 0.5, 1.0, 2.0, and 4.0 h. The cells treated with different formulations were then digested, collected, centrifuged, and resuspended in 500 µL PBS. Finally, intracellular fluorescence was measured by flow cytometry. For fluorescence microscopy analysis, the treated cells were washed and stained with Hoechst 33342 for 15 min. The cells were then washed again, and intracellular fluorescence was examined under a fluorescence microscope. The mean fluorescence intensity was quantified using ImageJ software. For the competitive inhibition assay, 4T1 cells were pre‐incubated with culture medium containing an excess of free methionine (1 mg/mL) for 2 h. Subsequently, the aforementioned formulations were added. Following the same experimental procedure described above, the cells were observed under a fluorescence microscope. The mean fluorescence intensity (MFI) was then quantified using ImageJ software.

2.5. Key Transporters Analysis

The 4T1 and HC11 cells (5 × 10⁵) were lysed in RIPA buffer supplemented with protease/phosphatase inhibitors (Cell Signaling Technology, USA), followed by centrifugation at high speed at 4°C. The protein concentrations were determined by the BCA assay (Beyotime). Equal amounts of protein (30 µg) were fractionated by SDS‐PAGE and transferred to a polyvinylidene difluoride membrane. Membranes were blocked with 10% nonfat milk in TBST for 2 h and incubated with the appropriate primary antibody against LAT1 overnight at 4°C. Membranes were washed three times with TBST and incubated with HRP‐conjugated secondary antibodies at room temperature for 2 h. Blots were developed using chemiluminescence substrates (BeyoECL Moon, P0018 FS, Beyotime) and visualized using a gel imaging analysis system (12003153, BIO EQUIPRAD).

2.6. Cell Viability Analysis

The 4T1 cells were seeded into 96‐well plates at a density of 3000 cells/well and cultured for 24 h. Then, a series of dilutions of IR808, PF9366, IR808/PF9366 combination, IR@PF NPs, and IR@PF‐M NPs were added to fresh medium. Cells cultured in medium without therapeutic agents were used as a control. After 4 h of incubation, the cells were subjected to irradiation at 808 nm with an intensity of 0.5 W/cm² or kept in the dark. After another 24 h of incubation, 20 µL of MTT (5 mg/mL) was added to each well. After 4 h of incubation at 37°C, the medium was replaced with 200 µL of DMSO, and the absorbance at 570 nm was measured. Moreover, RAW 264.7 cells were seeded in 96‐well plates at a density of 5 × 10³ cells per well. These cells were incubated with the above formulations in the dark to assess their biocompatibility. Cell viability was calculated using the following equation, and the half‐maximal inhibitory concentration (IC50) values were calculated using GraphPad Prism software. Moreover, the live ratio of the cells after different treatments was evaluated by flow cytometry. Specifically, 4T1 cells were seeded in a 12‐well plate at a density of 1 × 10⁵ cells/well and cultured for 24 h to allow adherence. The cells were divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. The cells were washed with PBS and stained with 500 µL of the Calcein AM working solution for 30 min. Finally, the cells were harvested and analyzed using a flow cytometer to quantify the green fluorescence (Calcein AM). Untreated cells served as the negative control.

2.7. Cell Colony Assay

The 4T1 cells were seeded in 12‐well plates at a density of 4 × 10⁴ cells per well and cultured for 24 h. Afterward, the cells were divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. Next, the cell culture medium was replaced with fresh medium every 24 h for three consecutive days. Subsequently, the cells were washed with PBS, fixed with 4% paraformaldehyde, and stained with 1× Giemsa solution. The samples were then washed with PBS three times, air‐dried, and imaged. Furthermore, the samples were dissolved in 1 mL of lysis solution, and the absorbance of the sample was recorded.

2.8. Cell Migration Scratch Assay

The 4T1 cells were seeded in 12‐well plates at a density of 4 × 10⁴ cells per well and cultured for 24 h. Once the monolayer reached 90% confluence, the medium was discarded, and cells were washed with sterile PBS. Three vertical scratches were made, and the cells were washed again with PBS to remove floating debris. The cells were then divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. After culturing for 24 h, the cells were incubated in fresh medium with 2.5% FBS for another 24 h, during which the width of the scratched gap was recorded. Relative migration was calculated as follows:

$$\begin{array}{cc}& \mathrm{Relative}\mathrm{mobility}\left(\%\right)\hfill \\ & =\frac{\mathrm{Sample}\mathrm{group}0\mathrm{h}\mathrm{area}-\mathrm{sample}\mathrm{group}24\mathrm{h}\mathrm{area}}{\mathrm{Control}\mathrm{group}0\mathrm{h}\mathrm{area}-\mathrm{Control}\mathrm{group}24\mathrm{h}\mathrm{area}}\times 100\%\hfill \end{array}$$

2.9. Immunofluorescence Experiment

The 4T1 cells (4 × 10⁴ cells/well) were seeded in 12‐well plates and cultured for 12 h. The cells were then divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. The cells were then washed three times with PBS, fixed with 4% paraformaldehyde, permeabilized with 0.2% Triton X‐100, and blocked with 5% BSA. Afterward, the samples were incubated with primary antibodies against HMGB1 or CRT at 4°C for 12 h, followed by incubation with fluorescent secondary antibody (Alexa Fluor 488) at 37°C for 1 h. The samples were stained with DAPI and analyzed by confocal microscopy. The fluorescence intensities were quantified using ImageJ software.

2.10. Metabolic Marker Analysis

The 4T1 cells were seeded in 12‐well plates at a density of 4 × 10⁴ cells per well and cultured for 24 h. The cells were then divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF, (8) IR@PF plus irradiation, (9) IR@PF‐M, (10) IR@PF‐M plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. After incubation for 24 h, the supernatants were collected and centrifuged at 1500 rpm for 5 min. Then, SAM and Spermidine levels were measured using ELISA kits according to the manufacturer's protocol. What is more, the protein expression of tri‐methylated histone H3 lysine 4 (H3K4me3) was analyzed via Western blot.

2.11. ATP Level Detection

The 4T1 cells were seeded in 12‐well plates at a density of 4 × 10⁴ cells per well and cultured for 24 h. The cells were then divided into ten groups and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were incubated with cells at an equivalent dose of 3 µg/mL IR808 and 0.35 µg/mL PF9366 for 2 h. The irradiation‐treated groups were exposed to 808 nm irradiation (0.5 W/cm², 5 min) after 2 h of incubation. After incubation for 24 h, the supernatants were collected and centrifuged at 1500 rpm for 5 min. Then, ATP levels were measured using kits according to the manufacturer's protocol.

2.12. Animal Studies

Female BALB/c mice (18‐22 g) were purchased from Charles River Laboratories (Beijing, China). This study obtained ethical approval from Wenzhou Research Institute of Guoke (ID number: WIUCAS25060502). All experimental procedures were conducted following the ARRIVE guidelines and in accordance with the Regulations of the Experimental Animal Administration, issued by the State Committee of Science and Technology of the People's Republic of China.

2.13. Hemolysis Test

Approximately 2 mL of blood was collected from healthy Balb/c mice into a plastic tube with a little heparin. Subsequently, the sample was centrifuged at 2000 rpm for 15 min, and the pellet was resuspended in 3 mL saline. This centrifuge‐ resuspension process was repeated 3 times to separate the integral erythrocytes. After that, they were dispersed in saline at a concentration of 5% (v/v) and incubated with IR808 alone, IR@PF NPs, and IR@PF‐M NPs. Erythrocytes suspended in saline and pure water served as negative and positive controls, respectively. Then, all samples were stored at 37°C for 1 h with a shaking frequency of 100 rpm, and the absorbance at 570 nm (OD₅₇₀) was measured using a Varioskan lux multimode microplate reader (Thermo, USA). Then, the hemolysis rate (%) was calculated as follows to assess the hemolytic activity of different formulations.

$$\mathrm{Hemolysis}\mathrm{rate}\left(\%\right)=\frac{\mathrm{O}{\mathrm{D}}_{\mathrm{experimental}\mathrm{group}}-\mathrm{O}{\mathrm{D}}_{\mathrm{negative}\mathrm{group}}}{\mathrm{O}{\mathrm{D}}_{\mathrm{positive}\mathrm{group}}-\mathrm{O}{\mathrm{D}}_{\mathrm{negative}\mathrm{group}}}\times 100\%$$

2.14. Tumor Model Establishment

Approximately 5 × 10⁶ 4T1 cells were subcutaneously implanted into the back of each mouse. The mice were housed in a specific pathogen‐free environment until the tumor volume reached approximately 100 mm³, at which point the experiments were initiated.

2.15. Biodistribution

The tumor‐bearing mice were randomly divided into three groups (n = 3) and treated with free IR808, IR@PF NPs, and IR@PF‐M NPs at an IR808 dose of 1 mg/kg, respectively. At 2, 4, 8, 12, and 24 h post‐injection, the mice were imaged using a Small Animal In Vivo Imaging System, and fluorescence in the tumor region was recorded. After the final in vivo imaging, the mice were sacrificed, and major organs and tumors were excised for fluorescence imaging.

2.16. In Vivo Photothermal Imaging

The in vivo photothermal efficiencies of different formulations were studied using an infrared thermal imaging camera. IR808 alone, IR@PF NPs, and IR@PF‐M NPs were intravenously injected into BALB/c mice bearing 4T1 xenograft tumors at a dose of 1 mg/kg, equivalent to IR808. The mice treated with IR808 received irradiation 2 h after administration, while those treated with nano‐assemblies received irradiation 12 h after administration. The irradiation (808 nm, 0.5 W/cm²) lasted for 5 min, and the temperature change of the tumor site was monitored using an IR thermal camera every 30 s. The mice treated with PBS served as the control.

2.17. In Vivo Photo‐Therapeutic Effects

The tumor‐bearing mice were randomly divided into ten groups (n = 5) and treated as follows: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combination, (6) IR808/PF9366 combination plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, (10) IR@PF‐M NPs plus irradiation. These formulations were intravenously administered on days 1, 3, and 5 at an IR808 dose of 1 mg/kg and a PF9366 dose of 0.12 mg/kg, respectively. The mice treated with IR808 received irradiation 2 h after administration, while those treated with nano‐assemblies received irradiation 12 h after administration. The laser‐treated groups received light irradiation (808 nm, 0.5 W/cm²) 12 h post‐injection. Tumor volume and body weight were measured daily. Finally, the mice were euthanized on day 14, and blood was collected and centrifuged to separate the serum for hepatic and renal function analysis. All tumors were harvested, photographed, and weighed. Three of the tumors were minced and digested at room temperature for 30 min, then filtered through a 400‐mesh sieve. The filtered samples were centrifuged, the supernatant was discarded, and the samples were resuspended (1 × 10⁶/mL). The cells were stained according to the manufacturer's protocol. Finally, the samples were analyzed using a flow cytometer. To further evaluate the functional status of the immune microenvironment, the concentrations of IFN‐γ, IL‐12 p70, and iNOS in the tumor tissues were quantified using commercial ELISA kits according to the manufacturer's instructions. All results were normalized to the total protein concentration of each sample, as determined by a BCA protein assay kit, to ensure consistency across samples. Finally, the tumor and major organs from each group were harvested, fixed, embedded, and stained with hematoxylin and eosin to investigate pathological changes.

2.18. Statistical Analysis

All in vitro experiments, including cellular uptake, cytotoxicity, and ICD marker expression, were performed in at least three independent biological replicates. Data are reported as mean ± standard deviation (SD). Comparisons between groups were analyzed using a t‐test and one‐way analysis of variance (ANOVA). p‐values < 0.05 (p< 0.05) were considered statistically significant.

3. Results

3.1. Preparation and Characterization of IR@PF‐M NPs

The co‐assembly behavior of IR808 and PF9366 was first examined at various molar ratios to optimize nanoparticle formation. As shown in Figure S1, the ratio of the two components markedly influenced the resulting particle size. When the molar ratio of IR808 to PF9366 was adjusted to 4:1 (corresponding to a weight ratio of 8.7:1), the resulting nanoparticles—referred to as IR@PF NPs—exhibited a relatively small hydrodynamic diameter of 142.1 ± 1.26 nm and a narrow size distribution, with a polydispersity index (PDI) of just 0.02. This nanoscale size and uniformity are considered ideal for promoting tumor accumulation via the enhanced permeability and retention (EPR) effect. Consequently, this formulation was identified as the optimal co‐assembly condition and selected for the preparation of methionine‐modified nanoparticles termed IR@PF‐M NPs. As shown in Figure 1A, IR@PF NPs exhibited a homogeneous appearance with no macroscopic aggregates or precipitation. Transmission electron microscopy (TEM) further revealed that the nanoparticles possessed a uniform spherical morphology with a well‐dispersed distribution. After surface modification with DSPE‐PEG2000, the zeta potential of the nanoparticles was measured to be nearly neutral (Figure 1C), a characteristic favorable for systemic circulation and intravenous administration. To further enhance tumor‐targeting specificity, the nanoparticles were functionalized with DSPE‐PEG2000‐Met—a methionine‐conjugated polymer—leveraging the upregulation of methionine transporters on tumor cells. The prepared IR@PF‐M NPs showed no significant differences in particle size, size distribution, appearance, or morphology compared to the non‐targeted version (Figure 1B). Notably, the zeta potential of IR@PF‐M NPs exhibited a significant decrease compared to the non‐targeted version. This reduction is probably attributed to the deprotonation (ionization) of the methionine carboxyl groups into COO⁻ in aqueous environments (Figure S2). Nevertheless, the overall surface charge remained within a physiologically acceptable range for biomedical applications.

FIGURE 1.

Physicochemical characterization and molecular modeling of IR@PF‐M NPs. (A,B) Hydrodynamic size distribution (measured by DLS) and representative appearance: TEM images showing the morphology of (A) IR@PF NPs and (B) IR@PF‐M NPs. (C) Zeta potential of different nanoparticle formulations in aqueous solution. (D) Chemical structures of IR808 and PF9366. (E) The binding mode of the receptor (PF9366, green) and IR808 (cyan). Left: Electrostatic potential surface of the receptor with the IR808 ligand encapsulated within the hydrophobic cavity. Right: Detailed view of the binding pocket. Hydrogen bonds (2.4 Å) and π–π stacking (4.2 Å) interactions are represented by yellow and grey dashed lines, respectively. (F) UV–vis absorption and (G) fluorescence emission spectra of individual components dissolved in DMSO. (H) UV–vis absorption and (I) fluorescence emission spectra of different formulations dispersed in deionized water. (J) Photothermal heating curves and (K) Temperature variation of different formulations during three consecutive laser‐on/off cycles (808 nm, 0 W cm^{− 2}, 5 min).

Based on these, the hemocompatibility of the formulations was evaluated via a hemolysis assay. As shown in Figure S3, free IR808 exhibited pronounced hemolytic activity. This effect is likely attributable to two factors: first, its poor aqueous solubility leads to the formation of micro‐precipitates that adsorb onto the surface of red blood cells (RBCs), causing mechanical damage; second, the molecular structure of IR808, characterized by a hydrophobic cyanine backbone and hydrophilic hexanoic acid carboxyl groups, endows it with surfactant‐like amphiphilicity, which facilitates its insertion into the RBC membrane and induces pore formation or rupture. In stark contrast, both IR@PF NPs and IR@PF‐M NPs demonstrated negligible hemolytic effects, with hemolysis rates well below the 5% threshold. This improved safety profile is attributed to the effective shield of the hemolytic IR808 molecules within the hydrophobic core of the nano‐assemblies. Furthermore, the nearly neutral or slightly negative surface charge of the nanoparticles minimizes non‐specific interactions with the negatively charged RBC membranes. Coupled with the excellent biocompatibility of the surface‐exposed PEG and methionine moieties, these results suggest that the nanoformulations possess superior hemocompatibility, making them highly suitable for intravenous administration and systemic circulation.

Subsequently, the encapsulation efficiency (EE) and drug loading (DL) of the two therapeutic agents within the nanostructure were determined (Table S2). Both IR808 and PF9366 exhibited high EE, exceeding 90%. Benefiting from the “drug‐as‐carrier” characteristic of the carrier‐free assembled nanoplatform, the DL of IR808 reached approximately 70%, while that of PF9366 was around 8%. These values are substantially higher than those achieved by traditional nanocarriers via physical encapsulation. Such high drug‐loading capacity is expected to significantly minimize carrier‐associated systemic toxicity and enhance the therapeutic window for clinical translation. Additionally, the storage stability and physiological stability of the nano‐assemblies were evaluated. As shown in Figures S4 and S5, the particle size of both nanoparticles exhibited no significant changes after storage at 4°C for 14 days or incubation in a simulated physiological environment (PBS containing 10% FBS) at 37°C for 24 h. This exceptional structural stability ensures that the nanoparticles maintain their optimized size and narrow distribution during long‐term storage and throughout the systemic circulation, thereby avoiding premature clearance by the reticuloendothelial system (RES) and ensuring efficient EPR‐mediated tumor accumulation. Furthermore, the drug leakage rates of the nanoparticles were investigated following the stability tests. The results showed that the leakage of both encapsulated drugs was less than 5% (Table S1). This high degree of payload retention is crucial for minimizing “off‐target” toxicity caused by premature drug release in the bloodstream. By ensuring that the majority of the therapeutic cargo remains sequestered within the nanocarrier until it reaches the tumor site, this formulation significantly enhances the safety profile and maximizes the therapeutic efficacy of the dual‐drug delivery system.

To elucidate the self‐assembly mechanism of IR808 and PF9366, molecular docking studies were performed. As shown in Table S3 and Figure 1E, the IR808 molecule exhibited high structural complementarity and strong binding affinity with the target receptor, yielding a binding free energy of −5.321 kcal/mol. The optimized annotations showed that the IR808 ligand was found to be deeply encapsulated within the hydrophobic cavity of the receptor, forming a highly stable assembly. Notably, a hydrogen bond was identified between the IR808 molecule and the receptor with a bond distance of 2.4 Å, which significantly contributed to the structural stability of the IR808 within the pocket. Furthermore, the aromatic benzene ring moiety of the IR808 compound engaged in π–π stacking interactions with the receptor framework. These synergistic interactions, including hydrophobic interaction, hydrogen bonding and π–π stacking, play a pivotal role in stabilizing the supramolecular complex and enhancing its functional performance.

The optical characteristics of the nanoparticles were systematically evaluated. As shown in Figure 1F, free IR808 (dissolved in DMSO) displayed a strong absorption peak at 808 nm in the near‐infrared (NIR) region, which is advantageous for deep tissue light penetration. Upon excitation, a corresponding fluorescence emission peak was observed at 830 nm (Figure 1G). In contrast, PF9366, DSPE‐PEG2000, and DSPE‐PEG2000‐Met exhibited negligible absorbance and fluorescence within the 400–900 nm range, confirming that IR808 is the primary contributor to the optical signature of the formulation. This intrinsic fluorescence was thus employed in subsequent studies to monitor nanoparticle uptake and distribution. When IR808 was dispersed in aqueous solution, its absorption and fluorescence signals were significantly attenuated due to poor solubility and precipitation (Figure 1H,I). In comparison, the co‐assembled nanoparticles maintained spectral signals, albeit with slight reductions in intensity. Notably, a blue shift and peak broadening were observed in the absorbance spectra of the nanoparticles, indicative of H‐type aggregation—a common phenomenon among cyanine dyes—which is known to enhance photothermal conversion via the aggregation‐induced photothermal effect. In agreement, the photothermal performance of IR@PF‐M NPs significantly surpassed that of free IR808 (Figure 1J). Upon NIR laser irradiation (808 nm, 0.5 W/cm², 5 min), the temperature of the nanoparticle solution rose by approximately 30°C, while free IR808 under identical conditions generated only a ∼15°C increase, underscoring the superior heating efficiency of the nanoformulations. Furthermore, the nanoformulations exhibited enhanced photothermal stability. As shown in Figure 1K, IR@PF‐M NPs maintained consistent heating rates and peak temperatures across three consecutive laser‐on/off cycles. In stark contrast, the photothermal efficiency of free IR808 declined sharply, showing negligible temperature elevation by the third cycle, similar to that of pure water. This sustained photothermal capability offers significant therapeutic advantages. The robust resistance to photobleaching ensures consistent and reproducible heating during repeated or prolonged irradiation, which is essential for complete tumor ablation. Moreover, the molecular confinement within the nanostructure protects IR808 from rapid degradation, allowing for a reduced therapeutic dose while maintaining high treatment efficacy and safety.

3.2. Cellular Uptake and In Vitro Cytotoxicity

The cellular internalization efficiency of IR808 and the two nanoformulations was assessed using both flow cytometry and fluorescence microscopy. As shown in Figure 2A _i and Figures S6 and S7, both analytical methods confirmed that the nanoparticle assemblies exhibited significantly higher cellular uptake than free IR808. This enhancement is likely attributable to their improved colloidal stability and aqueous dispersibility. Time‐course studies revealed a progressive increase in uptake from 0.5 to 2 h, followed by a plateau between 2 and 4 h. Based on these findings, a 2 h incubation period was selected as the standard condition for subsequent cellular experiments. Notably, the methionine‐modified IR@PF‐M NPs showed markedly greater cellular accumulation compared to the non‐targeted IR@PF NPs, demonstrating the feasibility of using methionine as a targeting ligand for specific delivery. This superior uptake was presumably facilitated by the overexpression of methionine transporters (specifically LAT1) on the surface of 4T1 tumor cells, as confirmed by Western blot analysis (Figure S8). To further elucidate the mechanism underlying this enhanced internalization, a competitive inhibition assay was conducted. As illustrated in Figure 2A _ii, when cells were pre‐incubated with an excess of free methionine (1 mg/mL), the intracellular fluorescence intensity of the IR@PF‐M NPs treated group was significantly attenuated. In contrast, the uptake of both free IR808 and the non‐targeted IR@PF NPs remained largely unchanged under the same competitive conditions. Importantly, after pre‐treatment with excess methionine, the fluorescence intensity in the IR@PF‐M NPs treated group decreased to a level comparable to that of the IR@PF NPs treated group, with no statistically significant difference between them. These results collectively indicate that the enhanced accumulation of IR@PF‐M NPs is primarily mediated by methionine transporter‐dependent endocytosis, thereby confirming the active targeting specificity of the nanoformulations toward tumor cells.

FIGURE 2.

In vitro cellular uptake and cytotoxicity of IR@PF‐M NPs. (A) Representative fluorescence images of 4T1 cells treated with IR808, IR@PF NPs, and IR@PF‐M NPs for the indicated time points (scale bar = 50 µm). Sub‐panels (Ai) and (Aii) represent cells without and with methionine pre‐incubation (1 mg/mL, 2 h), respectively. Cell viability of (B) RAW 264.7 cells and (C) 4T1 cells treated with various formulations in the dark for 24 h. (D) Cell viability of 4T1 cells after different treatments under NIR irradiation (808 nm, 0.5 W cm^{− 2}, 5 min) and (E) the corresponding IC₅₀ values. (F) Flow cytometry analysis of the cells stained with Calcein‐AM (live) and PI (dead) after different treatments. The numbers (1–10) represent cells treated with: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combo, (6) IR808/PF9366 combo plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, and (10) IR@PF‐M NPs plus irradiation. Scale bar = 100 µm. Data are presented as mean ± SD (n = 3). Images are representative of three independent observations.

The cytotoxicity of all formulations in the absence of laser irradiation (“dark toxicity”) was evaluated in both murine macrophages (RAW 264.7) and murine breast cancer cells (4T1) using the MTT assay. As shown in Figure 2B,C, all treatments exhibited negligible cytotoxicity, with cell viability remaining above 90% even at IR808 concentrations of 3 µg/mL and PF9366 concentrations of 0.35 µg/mL. These results indicate excellent biocompatibility under non‐irradiated conditions.

Subsequently, photothermal cytotoxicity was assessed in 4T1 cells upon exposure to an 808 nm laser (0.5 W/cm² for 5 min). As shown in Figure 2D,E, IR808 alone induced moderate photothermal‐induced cell death. When combined with PF9366, the cytotoxic effect was further enhanced, likely due to simultaneous inhibition of methionine metabolism. Both nanoparticle formulations exhibited significantly greater cytotoxicity than free IR808, consistent with their superior cellular uptake. In particular, IR@PF‐M NPs achieved nearly complete cell death at a 3 µg/mL IR808 equivalent concentration, highlighting the advantage of methionine‐mediated active targeting.

To further validate these results, the cells following different treatments were stained with Calcein AM to visualize viable cells. As illustrated in Figure 2F, all non‐irradiated groups exhibited dense and intense green fluorescence signals, comparable to the control group, confirming the favorable biosafety and minimal dark toxicity of the formulations. In contrast, the cell viability in all irradiation‐treated groups showed a significant decline, evidenced by the marked reduction in Calcein AM signals and the emergence of a new cell population with weak fluorescence intensity on the left side of the plots. Notably, the cells treated with IR@PF‐M NPs plus laser irradiation exhibited an almost complete disappearance of green fluorescence, indicating widespread cell death. These results collectively demonstrate the potent and selective photothermal cytotoxicity of the IR@PF‐M nanoformulations in vitro.

3.3. Metabolic Inhibition and Synergistic Immunogenic Effects

PTT is well known for its ability to induce ICD, a form of regulated cell death characterized by the extracellular release of damage‐associated molecular patterns (DAMPs)—including adenosine triphosphate (ATP) and high mobility group box 1 (HMGB1)—and the surface exposure of calreticulin (CRT). These signals promote DC maturation and antigen presentation, ultimately leading to tumor‐specific T‐cell activation and systemic antitumor immune responses. This shift in the tumor immune landscape is often described as a transition from an “immune‐cold” to an “immune‐hot” phenotype. However, the therapeutic efficacy of PTT may be hampered by tumor‐intrinsic adaptive resistance. In certain models, PTT‐induced thermal stress has even been reported to accelerate tumor proliferation and metastasis [46].

To overcome these limitations, we hypothesized that combining PTT with methionine metabolism inhibition could enhance the ICD response and improve therapeutic outcomes. MAT2A, a key enzyme catalyzing the conversion of methionine to S‐adenosylmethionine (SAM)—a central methyl donor in cellular metabolism—was selected as the metabolic target. As shown in Figure 3A, treatment with PF9366 markedly reduced intracellular SAM levels, confirming effective inhibition of MAT2A activity. Crucially, the depletion of the SAM pool led to significant downstream metabolic and epigenetic remodeling. As a key polyamine derived from SAM, the concentration of spermidine was substantially decreased Figure 3B, indicating the disruption of the methionine‐polyamine axis essential for tumor cell proliferation. Furthermore, Western blot analysis revealed a marked reduction in the levels of tri‐methylated histone H3 lysine 4 (H3K4me3, Figure 3C and Figure S9), a hallmark of active chromatin that is highly sensitive to methyl donor availability. Moreover, nanoparticle‐mediated delivery further enhanced cellular uptake of PF9366, leading to a more substantial decrease in SAM, spermidine, and H3K4me3 levels. These results collectively demonstrate that our formulation can effectively interfere with tumor cell metabolism at multiple levels, thereby potentially amplifying the ICD‐inducing effects of PTT.

FIGURE 3.

Metabolic inhibition and anti‐proliferation/migration properties of IR@PF‐M NPs in vitro. (A‐C) Mechanistic validation of MAT2A inhibition and its downstream consequences, including: (A) intracellular SAM levels quantified by ELISA; (B) concentration of the polyamine metabolite spermidine; and (C) Western blot analysis of H3K4me3 expression. (D) Representative images of colony formation assays and scratch wound healing assays of 4T1 cells after the indicated treatments. The numbers (1–10) represent cells treated with: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combo, (6) IR808/PF9366 combo plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, and (10) IR@PF‐M NPs plus irradiation. Scale bar = 500 µm. Data are presented as mean ± SD (n = 3). Images are representative of three independent observations. Statistical significance: ^*** p< 0.001, ^**** p< 0.0001.

To evaluate the downstream functional consequences of methionine metabolism blockade, we next assessed the anti‐proliferative and anti‐migratory properties of the formulations using colony formation and scratch wound healing assays. As shown in Figure 3D and Figure S10, untreated 4T1 cells exhibited robust clonogenic potential, with a colony formation rate of approximately 95%. Treatment with PF9366 alone modestly reduced this rate by ∼20%, indicating a moderate inhibition of proliferation. IR808‐mediated PTT further suppressed colony formation to ∼65%, and a synergistic effect was observed when PF9366 was combined with IR808, reducing the rate to ∼50% in the IR@PF NPs + laser group. Notably, the methionine‐modified nanoformulations (IR@PF‐M NPs) led to a dramatic reduction in colony formation to ∼10% upon irradiation, reflecting both enhanced cellular internalization and potent combinatorial cytotoxicity. Given that tumor cell migration often accompanies proliferation, we further evaluated cell motility using a scratch wound healing assay. The observed anti‐migratory effects across treatment groups closely paralleled the trends in colony suppression, supporting a mechanistic correlation between inhibited proliferation and impaired migration.

Subsequently, the capacity of methionine metabolism inhibition to modulate ICD was then examined. As shown in Figure 4, PF9366 alone failed to induce substantial ICD, as evidenced by minimal changes in CRT exposure, ATP secretion, and HMGB1 release. In contrast, IR808‐mediated PTT alone did trigger ICD to a limited extent. Importantly, the combination of PTT and methionine metabolic blockade significantly enhanced ICD marker expression. Among all groups, IR@PF‐M NPs combined with laser irradiation induced the most pronounced ICD response, highlighting the strong synergistic effect of this dual‐modality strategy (Figure 4A–D).

FIGURE 4.

Induction of immunogenic cell death (ICD) in 4T1 cells by IR@PF‐M NPs. (A) Representative fluorescence images of calreticulin (CRT) surface exposure and HMGB1 efflux in 4T1 cells following different treatments. Quantitative analysis of the relative fluorescence intensity for (B) CRT and (C) HMGB1 expression. (D) Extracellular ATP levels secreted by 4T1 cells after various treatments. The numbers (1–10) represent cells treated with: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combo, (6) IR808/PF9366 combo plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, and (10) IR@PF‐M NPs plus irradiation. Scale bar = 100 µm. Data are presented as mean ± SD (n = 3). Images are representative of three independent observations. Statistical significance: ^** p< 0.01, ^*** p< 0.001.

Together, these findings demonstrate that IR@PF‐M nanoparticles exert potent anti‐proliferative and anti‐metastatic effects while simultaneously promoting immunogenic cell death. By integrating photothermal therapy with metabolic reprogramming, this nanoplatform effectively reshapes the tumor microenvironment and holds strong potential for enhancing antitumor immunity in triple‐negative breast cancer.

3.4. Tumor Accumulation and Photothermal Efficiency

Tumor‐targeting capability and photothermal performance were assessed in vivo using a small animal imaging system. In mice administered free IR808, only weak and gradually declining fluorescence signals were detected at the tumor site between 2 and 24 h post‐injection, indicating poor tumor accumulation (Figure 5A,B). In contrast, both IR@PF and IR@PF‐M nanoparticles exhibited markedly stronger and progressively increasing tumor fluorescence, peaking at approximately 12 h, which was selected as the optimal time point for subsequent laser irradiation. Notably, methionine‐functionalized IR@PF‐M NPs achieved significantly higher tumor enrichment than their non‐targeted counterparts, as confirmed by enhanced fluorescence in both in vivo imaging and ex vivo tumor analysis (Figure 5B,C). These findings underscore the effectiveness of methionine ligand modification in enhancing tumor‐specific nanoparticle uptake.

FIGURE 5.

In vivo tumor accumulation and photothermal performance of IR@PF‐M NPs. (A) Representative in vivo fluorescence images of 4T1 tumor‐bearing mice at indicated time points and ex vivo fluorescence images of major organs and tumors harvested at 24 h post‐injection. (B) Semi‐quantitative fluorescence analysis of the tumor region over 24 h and (C) fluorescence biodistribution in major organs and tumors at the 24 h time point. (D) Photothermal (infrared) images and (E) corresponding heating curves of the tumor region in mice treated with PBS or various formulations under 808 nm laser irradiation (0.5 W cm^{− 2}, 5 min). Data are presented as mean ± SD (n = 3). Statistical significance: ^** p< 0.01, ^*** p< 0.001, ^**** p< 0.0001.

At the identified optimal accumulation time, tumors were irradiated with an 808 nm near‐infrared laser to assess photothermal conversion. As shown in Figure 5D,E, laser irradiation alone or in combination with free IR808 resulted in minimal temperature elevation, with the tumor surface temperature reaching only ∼37°C due to insufficient accumulation of the free dye. In contrast, nanoparticle‐treated groups exhibited a significant thermal response, reflecting their superior tumor retention and IR808 loading. Among these, IR@PF‐M NPs generated the highest intratumoral temperature—exceeding 50°C—indicating excellent photothermal conversion efficiency and strong potential for effective tumor ablation.

3.5. In Vivo Antitumor Efficacy, Immune Activation, and Biosafety Evaluation

The therapeutic performance of IR@PF‐M nanoparticles was systematically evaluated in a 4T1 murine breast tumor model. As illustrated in Figure 6A, treatment was initiated when tumors reached approximately 100 mm³ (designated as day 1), with intravenous administrations on days 1, 3, and 5. Mice receiving free IR808 or IR808/PF9366 combination were subjected to NIR laser irradiation 2 h post‐injection, while nanoparticle groups (IR@PF NPs and IR@PF‐M NPs) received laser treatment 12 h after injection to allow for tumor accumulation. As shown in Figure 6B, tumors in the PBS group progressed rapidly, reaching ∼1200 mm³ within 14 days. Free IR808 or PF9366, even with irradiation, failed to significantly inhibit tumor growth, likely due to poor tumor‐targeting capacity in their unencapsulated forms. Similarly, the combination of free IR808 and PF9366 offered limited improvement, underscoring the need for an efficient co‐delivery system. By contrast, both nanoformulations demonstrated enhanced therapeutic performance. Even without irradiation, IR@PF‐M NPs slowed tumor progression, likely due to the intrinsic metabolic inhibitory effect of PF9366. Upon NIR laser exposure, IR808‐mediated photothermal ablation synergized with PF9366‐induced methionine metabolism blockade, resulting in marked tumor growth inhibition. Notably, the IR@PF‐M NPs group exhibited the most pronounced tumor suppression, with average tumor volumes reduced to below 100 mm³. The tumor weight and appearance were consistent with the final tumor volume on day 14 (Figure 6C,D). Histological analysis by H&E staining confirmed widespread nuclear fragmentation and tissue destruction in this group (Figure 6E), further validating the potent therapeutic efficacy of this dual‐functional platform.

FIGURE 6.

In vivo synergistic antitumor efficacy of IR@PF‐M NPs. (A) Schematic illustration of the therapeutic schedule and dosing regimen in 4T1 tumor‐bearing mice. (B) Tumor volume growth curves of mice over the 14‐day treatment period (n = 5). (C) Average weights and (D) representative photographs of the excised tumors harvested on day 14 post‐treatment. (E) Representative H&E‐stained histological sections of tumor tissues from different treatment groups. The numbers (1–10) represent mice treated with: (1) PBS, (2) IR808, (3) IR808 plus irradiation, (4) PF9366, (5) IR808/PF9366 combo, (6) IR808/PF9366 combo plus irradiation, (7) IR@PF NPs, (8) IR@PF NPs plus irradiation, (9) IR@PF‐M NPs, and (10) IR@PF‐M NPs plus irradiation. Scale bar = 100 µm. Data are presented as mean ± SD. Statistical significance: ^** p< 0.01, ^**** p< 0.0001.

To investigate whether the observed tumor suppression was associated with immune activation, tumor‐infiltrating immune cells were profiled by flow cytometry (Figure 7 and gating in Figures S11–S13). Following laser irradiation, IR@PF‐M NPs significantly increased the release of immunogenic DAMPs such as CRT and HMGB1, indicative of robust ICD induction. This effect facilitated DC maturation and recruitment, as reflected by a 2‐fold increase in intratumoral DCs compared to the PBS group (Figure 7A,B). Consistently, the level of interleukin‐12 p70 (IL‐12 p70) was markedly up‐regulated (Figure S14A), confirming the functional maturation of DCs and their capacity to initiate immune responses. In parallel, polarization of tumor‐associated macrophages toward the pro‐inflammatory M1 phenotype was observed, accompanied by reduced M2 populations (Figure 7C,D). The successful M1‐type polarization was further validated by the significant elevation of inducible nitric oxide synthase (iNOS) (Figure S14B). As a hallmark functional enzyme of M1 macrophages, iNOS promotes the production of cytotoxic nitric oxide to directly inhibit tumor growth. These changes created a more immunostimulatory tumor microenvironment through enhanced antigen presentation and pro‐inflammatory signaling. In addition, CD3⁺ T cell infiltration was significantly enhanced (Figure 7E,F), with marked expansion of effector CD8⁺ T lymphocytes (Figure 7G,H). To verify the cytotoxic potency of these recruited T cells, we quantified the secretion of interferon‐gamma (IFN‐γ). The results showed a substantial increase in IFN‐γ levels in the IR@PF‐M NPs group (Figure S14C), a key effector cytokine that drives cell‐mediated antitumor immunity. Collectively, these findings suggest that the IR@PF‐M NPs plus laser treatment not only increased the infiltration of immune cells but also effectively triggered their functional activation, establishing a robust and sustained antitumor immune response.

FIGURE 7.

Modulation of the tumor immune microenvironment (TIME) by IR@PF‐M NPs. (A) Representative flow cytometric plots and (B) corresponding quantitative analysis of mature dendritic cells (mDCs, gated on CD80⁺ and CD86⁺). (C) Flow cytometric analysis and (D) quantitative percentages of tumor‐associated macrophage (TAM) polarization, including M1‐like macrophages (CD80⁺/CD86⁺) and M2‐like macrophages (CD206⁺). (E, F) Flow cytometric plots and quantitative analysis of (E, F) total T cell infiltration (CD3⁺) and (G, H) cytotoxic T lymphocyte (CTL) infiltration (CD3⁺CD8⁺) in tumor tissues following different treatments. Data are presented as mean ± SD (n = 3). Statistical significance: ^** p< 0.01 and ^**** p< 0.0001.

Biosafety was concurrently evaluated to assess the translational potential of this treatment. Throughout the 14‐day treatment period, no significant body weight loss was observed across all groups, indicating acceptable systemic tolerance (Figure S15). H&E staining of major organs (heart, liver, spleen, lung, kidney) revealed no signs of pathological damage (Figure S16). Furthermore, hematological and biochemical parameters remained within normal physiological ranges, confirming minimal hematotoxicity and nephrotoxicity (Figure S17).

Taken together, these findings demonstrate that IR@PF‐M NPs exhibit excellent in vivo antitumor efficacy by integrating photothermal therapy with metabolic intervention. This strategy not only inhibits tumor proliferation and growth but also triggers strong immunogenic responses, all while maintaining favorable systemic safety, underscoring its promise for clinical translation in TNBC therapy.

4. Discussion

In this study, we developed a co‐assembled nanoplatform (IR@PF‐M NPs) that integrates PTT with metabolic inhibition to treat TNBC. By co‐loading the photothermal agent IR808 and the MAT2A inhibitor PF9366, and surface‐modifying with methionine‐conjugated DSPE‐PEG2000, the nanoparticles achieved tumor‐specific accumulation and active cellular uptake via overexpressed methionine transporters. Upon NIR laser irradiation, IR808 induced local hyperthermia to promote ICD, while PF9366 disrupted methionine metabolism to further sensitize tumor cells and reshape the tumor microenvironment. The combination strategy significantly inhibited tumor proliferation and metastasis, triggered potent immune activation, and exhibited excellent biocompatibility in vivo.

PTT has attracted increasing attention for its ability to induce ICD and reshape the immunological landscape of tumors. However, its standalone efficacy is often limited by compensatory immunosuppressive mechanisms within the TME, which can facilitate immune evasion and recurrence [47, 48]. Our results demonstrate that metabolic intervention, when rationally integrated with photothermal treatment, offers a synergistic strategy to amplify antitumor efficacy. The mechanistic basis of the observed therapeutic benefit lies in the synergistic interplay between photothermal injury and metabolic reprogramming. PTT alone can initiate ICD by disrupting membrane integrity and promoting the release of immunogenic DAMPs [49]. However, tumor cells may respond to such stress by activating metabolic adaptation pathways that limit immune recognition [50, 51]. In particular, MAT2A inhibition via PF9366 significantly reduced intracellular SAM levels, impairing one‐carbon metabolism and sensitizing tumor cells to thermal stress [52]. Inhibiting methionine metabolism via MAT2A blockade not only impairs methylation‐dependent oncogenic signaling but also weakens the immunosuppressive barriers maintained by regulatory T cells and tumor‐associated macrophages. The dual insult from heat‐induced damage and metabolic blockade greatly enhanced ICD hallmarks—such as ATP release, CRT exposure, and HMGB1 secretion—thereby promoting DC maturation, macrophage repolarization, and CD8⁺ T cell infiltration. Importantly, tumor metabolism plays a critical role in sustaining tumor proliferation and progression, and combining metabolic intervention with other therapeutic modalities has been shown to yield improved outcomes [53, 54, 55, 56, 57, 58, 59, 60, 61, 62]. Our findings support this paradigm and also provide a mechanistic basis for expanding metabolic‐photothermal strategies into broader therapeutic contexts.

The design of IR@PF‐M NPs offers several translational advantages. First, the co‐assembled nanostructure ensures synchronized delivery of both therapeutic agents to the tumor site, overcoming the limitations of separate drug administration. Second, the near‐neutral surface charge and methionine‐targeting functionality enable effective circulation, reduced RES clearance, and enhanced cellular uptake [63]. Third, the NIR‐responsiveness of IR808 allows for spatiotemporal control over therapeutic activation, minimizing systemic side effects. Compared with conventional immunotherapies or chemotherapy‐photothermal combinations, our strategy leverages a tumor‐intrinsic metabolic vulnerability to augment therapeutic efficacy while maintaining safety. Given the pressing clinical need for effective TNBC therapies and the increasing recognition of metabolic modulation in oncology, this work provides a compelling basis for future development of personalized nanomedicine approaches integrating immune and metabolic axes.

A notable advantage of IR@PF‐M NPs lies in their carrier‐free co‐assembly design, where both IR808 and PF9366 serve as the intrinsic structural building blocks. Unlike conventional nanocarriers (e.g., liposomes or polymeric scaffolds) that often struggle with drug loading capacities below 10%, our platform achieves a high loading efficiency. This quantitative leap ensures a superior therapeutic payload per unit mass, allowing for a significant reduction in the administration dose of exogenous excipients. Consequently, this minimizes systemic metabolic burdens and mitigates potential excipient‐related toxicity—a common bottleneck for dose‐sensitive treatments like metabolic inhibition [64, 65]. Beyond high loading, this streamlined formulation demonstrates excellent stability. Our assays (Figures S4 and S5) confirm that these assemblies maintain structural integrity in physiological media for over 3 days, exhibiting colloidal stability comparable to established PEGylated lipid systems. By eliminating extraneous inert materials, the IR@PF‐M NPs reduce the risk of carrier‐induced immunogenicity and off‐target accumulation, thereby enhancing overall biocompatibility. Furthermore, the simplified one‐step assembly process facilitates large‐scale manufacturing and regulatory consistency, representing a clinically relevant innovation that balances high therapeutic efficacy with translational feasibility.

Despite these promising results, several limitations warrant further investigation to facilitate clinical translation. First, while the one‐step co‐assembly process ensures excellent scalability for large‐scale manufacturing, the long‐term in vivo degradability of cyanine‐based nano‐assemblies must be rigorously tracked via longitudinal studies to guarantee complete clearance and avoid potential cumulative toxicity, and comprehensive assessments regarding long‐term toxicity (>14 days) and systemic nanoparticle degradation/clearance kinetics remain to be systematically performed. Furthermore, although the biomimetic methionine‐coating (M‐layer) effectively reduces RES sequestration, the chronic immunogenicity of these synthetic assemblies in humans remains to be comprehensively assessed to ensure long‐term biosafety. Second, our therapeutic evaluation was primarily conducted in a subcutaneous 4T1 syngeneic murine model. While useful, this subcutaneous model does not fully replicate the intricate complexity and anatomical niche of orthotopic or metastatic breast cancer tumors, nor does it employ advanced patient‐derived xenograft or humanized immune system mouse models. Future studies utilizing these clinically relevant models will be essential to validate the translational applicability and the durability of the anti‐tumor immune memory response. Moreover, although the elevated infiltration of CD8⁺ T cells and increased levels of IFN‐γ observed herein suggest the activation of adaptive immunity, CD8⁺ T‐cell depletion experiments were not performed in the current study; consequently, the conclusion that antitumor efficacy is truly dependent on adaptive immunity remains indirect, representing a limitation that warrants T‐cell depletion assays in future mechanistic studies. Additionally, while MAT2A inhibition offers higher specificity than global methionine depletion, its broader systemic metabolic consequences require more granular evaluation through metabolomic profiling. Finally, the modular nature of our co‐assembly strategy provides a versatile template for personalized nanomedicine. Future research could explore the integration of this metabolic‐photothermal platform with immune checkpoint blockade (ICB) or adoptive T‐cell therapies to further amplify the systemic abscopal effect. By substituting IR808 or PF9366 with other functional NIR dyes or targeted inhibitors, this highly adaptable platform could be tailored to address a diverse range of tumor types and heterogeneous patient populations.

5. Conclusion

In this study, we developed a novel carrier‐free co‐assembled nanoplatform (IR@PF‐M NPs) that synergistically integrates photothermal therapy and methionine metabolism inhibition for the treatment of TNBC. Leveraging both the EPR effect and the overexpression of methionine transporters in tumor cells, the system enables efficient and targeted co‐delivery of the photothermal agent IR808 and the metabolic inhibitor PF9366. Following tumor accumulation and cellular uptake, IR@PF‐M NPs exert potent therapeutic effects through IR808‐mediated photothermal ablation and PF9366‐induced suppression of methionine metabolism. Importantly, this dual‐action strategy not only directly impairs tumor proliferation and migration but also amplifies ICD and remodels the tumor immune microenvironment, culminating in enhanced systemic antitumor immunity. Furthermore, the carrier‐free design ensures high drug loading, reduced formulation complexity, and improved translational potential. Taken together, this work presents IR@PF‐M NPs as a robust and versatile therapeutic platform, offering a promising strategy to overcome current limitations in TNBC treatment through the integration of metabolic targeting and immune‐potentiated phototherapy.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adhm71434‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.