A novel photosensitizer-based photodynamic therapy reprograms the Kynurenine-AhR axis to boost antitumor immunity in breast cancer.

1
Introduction
Breast cancer is the leading causes of cancer-related morbidity and mortality among women worldwide [1]. Triple-negative breast cancer (TNBC), characterized by the absence of estrogen receptor, progesterone receptor, and HER2 expression, fails to show response to endocrine therapy or HER2-targeted treatments, resulting in a poor clinical prognosis [2,3]. To address this, emerging therapies such as immune checkpoint inhibitors (ICIs), particularly the programmed death receptor 1/programmed death ligand 1 (PD1/PD-L1) blockade, have been extensively explored [4,5]. While ICIs have demonstrated efficacy against several cancers, their effectiveness in TNBC remains limited, with only a small subset of PD-L1 positive patients achieving meaningful clinical benefit [6,7].
Increasing evidence indicates that an immunosuppressive tumor microenvironment (TME) plays a critical role in resistance to ICI therapy [8,9]. The immunosuppressive TME is shaped not only by insufficient immune cell infiltration, but also by tumor-induced metabolic reprogramming [10]. Rapidly proliferating tumor cells deplete essential nutrients like glucose and amino acids while producing immunosuppressive metabolites, creating an acidic and nutrient-deprived environment [[11], [12], [13]]. This disrupts the TME and restricts the function of infiltrating T cells, which can results in T cell exhaustion and reduced immune responses [14,15]. Consequently, strategies targeting the metabolic state of the TME are increasingly showing potential to enhance the efficacy of immunotherapies [16,17].
Photodynamic therapy (PDT) is a localized treatment that uses photosensitizer (PS) activated by specific light wavelengths to generate reactive oxygen species (ROS), leading to tumor cell death [18]. The excitation wavelength used for PDT must not only efficiently activate the PS but also provide adequate tissue penetration for therapeutic effects in solid tumors. PDT typically uses light in the 600–800 nm range, where biological tissues exhibit minimal light absorption and scattering, allowing for deeper tissue penetration [19].Owing to its minimally invasive nature and repeatability, PDT has attracted increasing attention in solid tumor therapy [20]. For example, PDT has shown promising clinical results in breast cancer, including in early-stage disease as well as recurrent and metastatic lesions. Notably, PDT has demonstrated excellent clinical responses and minimal morbidity in patients with chest wall recurrences following mastectomy [21,22]. Beyond its direct cytotoxic effects, PDT has been shown to trigger immunogenic cell death and stimulate antitumor immune responses, suggesting its potential as a combinatorial therapy with ICIs [23,24].
However, existing research has primarily focused on its known immunological effects, while its regulatory role in tumor metabolic networks and specific immune-metabolic axes remain insufficiently explored. Given that ROS generated during PDT may influence tumor metabolism, understanding these metabolic changes is crucial for elucidating PDT's immunomodulatory properties and its potential to enhance the effectiveness of ICIs.
To investigate the impact of PDT on tumor metabolism, the present study focused on a novel, structurally optimized porphyrin-based PS, DTP. DTP is a tetraphenylporphyrin(TPP)-derived PS synthesized by conjugating diethylenetriamine-pentaacetic acid (DTPA) to mono-aminophenyl triphenylporphyrin (MAPP) through an amide linkage [25]. The porphyrin macrocycle enables strong absorption of light and the generation of ROS upon photoexcitation, which forms the basis of PDT's therapeutic effects [26]. However, TPP-based porphyrins are often hindered by poor aqueous solubility, which may reduce their photoactivity and complicate formulation [27]. To address these limitations, the conjugation of DTPA introduces multiple ionizable groups that significantly enhance hydrophilicity and aqueous compatibility of the PS. Previous studies have demonstrated that DTP efficiently generates ROS upon irradiation and exhibits remarkable phototoxicity with low dark toxicity in multiple gastric cancer cell lines in vitro [28]. Its mechanism of action involves activation of the mitochondrial apoptosis pathway and inhibition of the P38 MAPK signaling pathway [29]. Collectively, these findings establish DTP as an effective PS with strong antitumor properties.
Nevertheless, the photodynamic efficacy of DTP in breast cancer, as well as its influence on tumor metabolic reprogramming, remains unexplored. In this study, we systematically evaluated the antitumor activity of DTP-PDT in breast cancer both in vitro and in vivo. Furthermore, we employed untargeted and targeted metabolomics to characterize PDT-induced metabolic alterations, specifically focusing on immunometabolic pathways involved in remodeling the tumor immune microenvironment. Finally, we investigated the potential synergistic effects of combining DTP-PDT with PD-L1 blockade in suppressing breast cancer metastasis.

2
Materials and methods
2.1
Optical properties of DTP
The photosensitizer DTP was synthesized and provided by Prof.Tianjun Liu (Institute of Biomedical Engineering, Chinese Academy of Medical Sciences and Peking Union Medical College) [25]. The purity of DTP was analyzed using High Performance Liquid Chromatography (HPLC). The UV-visible absorption spectrum of DTP in dimethyl sulfoxide (DMSO) was obtained with a Hitachi UH5700 UV-visible spectrophotometer. A fluorescence spectrophotometer (Fluoromax-4, HORIBA, Japan) was used to measure the emission (Em) spectra.

2.2
ROS detection in solution
To detect the generation of hydroxyl radicals (•OH), singlet oxygen (1O2), and superoxide anions (O2•−) by DTP, 3,3′,5,5′-Tetramethylbenzidine (TMB), 1,3-diphenylisobenzofuran (DPBF), and dihydroethidium (DHE) were used as probes [30]. Briefly, DTP solution (5 μM) was mixed with an equal volume of probe working solution (TMB: 20 μM; DPBF: 10 μM; DHE: 5 μM) and gently stirred in a quartz cuvette. The mixture was then exposed to 650 nm laser (100 mW cm−2) for varying durations using a semiconductor laser (WSLS-650–500 mW cm−2, Wave Spectrum Laser Group Limited, China). The absorbance of TMB (652 nm) or DPBF (420 nm) was immediately measured after irradiation using a UV-Visible spectrophotometer. The fluorescence intensity of DHE (Ex/Em: 403/460 nm) was recorded right after each irradiation using a fluorescence spectrophotometer. ROS generation efficiency was calculated by comparing the signal intensity at each time point to the initial intensity (t = 0).

2.3
Cell culture
The Murine breast cancer cell line 4T1, human breast cancer cell line MDA-MB-231, and human mammary epithelial cell line MCF-10A were obtained from the National Collection of Authenticated Cell Cultures (Chinese Academy of Sciences, Shanghai, China). Cancer cells were cultured in RPMI 1640 or DMEM medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. The MCF-10A cell line was cultured in DMEM/F12 medium supplemented with 5% horse serum, 10 μg/mL insulin, 20 ng/ml epidermal growth factor, 100 ng/mL cholera toxin, and 0.5 μg/mL hydrocortisone. All cells were maintained at 37 °C in a humidified incubator with 5% CO2.

2.4
Subcellular colocalization assay
4T1 cells (2.0 × 105 cells/dish) were seeded and incubated in confocal dishes. Subsequently, DTP (4 μM) was added to the culture medium for 24 h. After incubation, the cells were stained with LysoTracker Green (50 nM) (ab176826, Abcam) and MitoTracker Green (100 nM) (C1048, Beyotime) at 37 °C for 30 min in the dark. After washing with PBS, the cells were then incubated with Hoechst 33342 (C1026, Beyotime) for 10 min. After a further PBS wash, fluorescence of DTP, lysosomes, mitochondria and nucleus was observed using Confocal laser scanning microscopy (CLSM) (APEXVIEW APX100, Olympus, Japan). Colocalization between DTP and organelles was analyzed using ImageJ (NIH, Bethesda, MD, USA).

2.5
Cellular uptake
4T1 cells (1 × 104 cells/well) were seeded in 96-well plates and cultured for 24 h. Cells were incubated with different concentrations (0-8 μM) of DTP for different periods (1-24 h). The cells were harvested after incubation, rinsed three times with PBS, counted using a hemocytometer, and lysed with lysis buffer. A multifunctional plate reader (VARIOSKAN FLASH, Thermo Fisher Scientific, Massachusetts, USA) was then used to measure the fluorescence intensity of the cell lysate. Uptake was expressed as the amount of DTP (pmol) per 2 × 104 cells.

2.6
Intracellular ROS measurement
4T1 cells (1 × 105 cells/well) were planted in 12-well plates and cultured for 24 h. The cells were then pretreated with DTP (0, 100, 200 nM) for 24 h, followed by incubation with 2′,7′-dichlorofluorescin diacetate (DCFH-DA) (1 μM) for 30 min at 37 °C in the dark. After treatment, the cells were exposed to laser (650 nm, 20 mW cm−2, 5 min). ROS signal was observed using a fluorescence microscope and quantified using image J software. For Flow cytometry (FCM) analysis, cells were harvested and analyzed by a flow cytometer (CytoFLEX LX, BECKMAN). FlowJo V10 software (FlowJo LLC, USA) was used to quantify mean fluorescence intensity of ROS signal.

2.7
In vitro cytotoxicity assay
4T1 cells (1 × 104 cells/well) were seeded and cultured in 96-well plates, followed by incubation with varying concentrations of DTP (0–2 μM) for 24 h. The cells were then irradiated with or without a 650 nm laser (20 mW cm−2, 5 min). After an additional 24 h incubation, cell viability was assessed using the MTT assay, following the manufacturer's instructions. With the addition of MTT and another 4 h incubation at 37 °C, the formed purple formazan crystals were dissolved with 150 μL DMSO. Cell viability was calculated using the formula: Viability (%) = (ADTP−Ablank)/(Acontrol−Ablank) × 100 (A: the absorbance measured at 570 nm using a multimode plate reader). Dose-response curves were determined by nonlinear regression analysis using GraphPad Prism 9.0 (GraphPad Software, USA). The half-maximal inhibitory concentration (IC50) value was calculated as the concentrations of DTP that caused 50% cell death relative to untreated controls. Cytotoxicity assays were also conducted in both the human TNBC cell line MDA-MB-231 and the normal human mammary epithelial cell line MCF-10A.

2.8
Live/dead cell staining
4T1 cells (5 × 105 cells/well) were seeded in 6-well plates and incubated for 12 h. The cells were then treated with DTP (0, 100, 200 nM) for 24 h, followed by 650 nm laser irradiation (20 mW cm−2, 5 min). After an additional 24 h incubation, 4T1 cells were stained using a Calcein AM and propidium iodide (PI) double-staining kit (C2015S, Beyotime Biotechnology) to distinguish live and dead cells following the manufacturer's protocol. Fluorescent images were then captured using fluorescence microscope.

2.9
Apoptosis analysis by FCM
4T1 cells (5 × 105 cells/well) were seeded in 6-well plates and incubated for 12 h. The cells were then pretreated with DTP (0, 100, 200 nM) for 24 h, and subsequently exposed to 650 nm laser irradiation (20 mW cm−2, 5 min). After a further incubation of 8 h, the Annexin V-FITC/PI apoptosis detection kit (C1062L, Beyotime Biotechnology) was applied to assess apoptosis according to the manufacturer's instructions. Finally, samples were analyzed on a flow cytometer.

2.10
In vivo animal models
Female BALB/c mice (6-8 weeks old, 18-20 g) were purchased from HFK Bioscience (Beijing, China) and housed in specific pathogen-free facilities conditions, with controlled temperature (22 ± 1 °C), humidity (50 ± 10%), and a 12 h light/dark cycle. All experimental protocols were conducted according to the NIH Guidelines for the Care and Use of Laboratory Anima and approved by the Institutional Animal Care and Use Committee (IACUC) of the Institute of Radiation Medicine, Chinese Academy of Medical Sciences (Ethics Approval No. IRM/2-IACUC-2409-094). To establish breast tumor model, 5 × 105 luciferase-transfected 4T1 (4T1-Luc) cells were suspended in 100 μL of PBS and subcutaneously injected into the right flank.

2.11
In vivo fluorescence imaging
Once 4T1 tumors reached 200-300 mm3 volume, DTP (10 mg/kg body weight) was injected intravenously into mice (n = 5). At 2, 8, 12, 24, and 48 h after injection, in vivo fluorescence imaging was conducted using the Maestro™ in vivo imaging system (CRI, Woburn, MA). For ex vivo imaging, tumor tissues and major organs (heart, kidney, liver, lung, and spleen) were excised and imaged. Fluorescence intensity was quantified as average radiance (photons/sec/cm2/sr).

2.12
In vivo antitumor efficacy
Once the tumor volume reached 80-100 mm3, tumor bearing mice were randomized into three groups (n = 10 per group): saline (Saline), DTP alone (DTP), and DTP with laser irradiation (DTP + L). DTP (10 mg/kg body weight) was intravenously injected every 2 days for 3 cycles. PDT was carried out using a 650 nm laser (100 mW cm−2, 15 min) at 12 and 24 h post-injection during each treatment cycle. Tumor volume and body weight were monitored every other day, and tumor progression was assessed weekly using bioluminescence imaging (IVIS Lumina III, Caliper Life Sciences). During the treatment, mice were anesthetized with a 2.5% isoflurane/oxygen mixture. Tumors were collected, fixed in 4% PFA, paraffin embedded, and sectioned into 4 μm slices. Tumor tissue sections were subjected to Hematoxylin & Eosin (H&E) staining for morphological evaluation. Immunohistochemical (IHC) staining was performed with primary antibodies against Ki67 (1:400, Abcam, AB15580), indoleamine 2,3-dioxygenase 1 (IDO1) (1:2000, Proteintech, 66528-1), and tryptophan 2,3-dioxygenase (TDO2) (1:200, Proteintech, 155880-1). Apoptosis was assessed using the TdT-mediated dUTP nick end labeling (TUNEL) assay. Image-J software was used to quantify the positive staining cells.

2.13
Untargeted metabolomics analysis
Cells were rinsed with PBS, counted, and lysed using a pre-chilled methanol/water solution. Disruption was further facilitated by freeze-thaw cycles, alternating between liquid nitrogen and 37 °C. The lysate was then ultrasonicated and centrifuged, and the supernatant was dried under nitrogen. Metabolites were analyzed using ultra-high-performance liquid chromatography coupled with quadrupole-Orbitrap high-resolution mass spectrometry (UHPLC-Q-OrbitrapMS) in both positive and negative ion modes as previously described [31]. After peak alignment, denoising, batch correction, and normalization, multivariate statistical analyses were conducted, including principal component analysis (PCA), orthogonal partial least squares discriminant analysis (OPLS-DA). Differential metabolites were identified with |log2FC| > 1 and p < 0.05, followed by KEGG pathway enrichment and hierarchical clustering analyses.

2.14
Targeted metabolomics analysis
Metabolite extraction from cells followed the same procedure as in the untargeted analysis. The supernatant was collected after centrifugation and analyzed by UHPLC-MS/MS analysis. Targeted detection used triple quadrupole mass spectrometry (MRM mode) to quantify tryptophan (Trp), kynurenine (Kyn), kynurenic acid (QA), and 5-hydroxytryptamine (5-HT), with separation achieved on a reversed-phase C18 column. Stable isotope-labeled internal standards were used for accurate quantification.
For tumor tissues, samples were harvested immediately after euthanasia, rinsed with PBS, snap-frozen in liquid nitrogen, and stored at −80 °C. Approximately 50 mg of tissue was homogenized in pre-chilled methanol, then centrifuged, and the supernatant was collected for targeted metabolomic analysis.

2.15
Kyn production assay
4T1 cells (5 × 105 cells/well) were seeded in 6-well plates and allowed to adhere for 24 h. To activate IDO1 activity, cells were pretreated with recombinant mouse interferon-γ (IFN-γ) (50 ng/mL, 485-MI, R&D System) for 24 h. After IFN-γ priming, cells were treated with DTP-PDT and/or the IDO1 inhibitor (NLG919, Aladdin). Culture supernatants were then collected for Kyn analysis via UHPLC-MS/MS analysis. The same assay was also performed in IFN-γ-stimulated MDA-MB-231 cells.

2.16
RNA extraction and quantitative real-time PCR (qRT-PCR)
Tumor tissues were dissociated into a single-cell suspension, and CD3+ T cells were isolated using a Mouse CD3+ T Cell Isolation Kit (19851, STEMCELL Technologies) following the manufacturer's instructions. Total RNA was extracted from the isolated CD3+ T cells using an RNA extraction kit (R1200, Solarbio). cDNA synthesis was performed using HiScript III RT SuperMix for qPCR (R323-01, Vazyme). qRT-PCR was conducted with AceQ Universal SYBR qPCR Master Mix (Q511, Vazyme) on a Real Time PCR System (7500, Applied Biosystems). The mRNA expression levels of Cyp1a1, Cyp1b1, and Ahrr were quantified using the 2−ΔΔCt method, with Hprt1 and Rplp0 serving as controls. Primer sequences are provided in Table S1.

2.17
Kyn rescue experiment
Tumor-bearing mice were randomly divided into four groups: Control (100 μL saline, intravenously), PDT (DTP-PDT was performed as described above), Kyn (100 mg/kg Kyn (K8625, Sigma-Aldrich) intraperitoneally every 2 days [32]), and PDT + Kyn (combination of DTP-PDT and Kyn treatment).Tumor tissues were collected after treatment to measure intratumoral Kyn levels and analyze tumor-infiltrating T cells.

2.18
Anti-metastasis effect evaluation
Primary tumors were established by subcutaneously inoculating 5 × 105 4T1-Luc cells into BALB/c mice. The mice were then randomly divided into four experimental groups: Model (100 μL saline, intravenously), α-PD-L1 (10 mg/kg α-PD-L1, intraperitoneally every 3 days), PDT (DTP-PDT was performed as described above), and PDT + α-PD-L1 (combination of DTP-PDT and α-PD-L1 treatment). After receiving treatment, the primary tumors were surgically removed on the 10th day to remove local lesions.
To mimic lung metastasis, secondary tumor challenge was performed via intravenous injection of 1 × 105 4T1-Luc cells 20 days after resection. Pulmonary metastatic progression was monitored by in vivo bioluminescence imaging on days 8, 11, and 14 post-rechallenge, with survival tracked up to day 60. Additionally, spleens were harvested for FCM analysis of central memory T cell (Tcm) populations. Lung tissues were also collected, fixed in Bouin's solution and stained with H&E staining.

2.19
Immune profiling analysis
Tumors and spleens were collected to assess the immune response. Tissues were mechanically dissociated into single cell suspensions. After RBC lysis, cells were filtered, washed, and stained with the Zombie Aqua™ Fixable Viability Kit (423101, Biolegend) to identify live and dead cells. For blocking non-specific binding, the cells were incubated with TruStain FcX™ (anti-mouse CD16/32 antibody, 101319, BioLegend). The following antibodies were used to stain the cells: anti-CD45 FITC (103107, Biolegend), anti-CD3ε Percp-cy5.5 (100327, Biolegend), anti-CD4 PE (100511, Biolegend), anti-CD8a APC (100711, Biolegend), anti-FoxP3 BV421 (126419, Biolegend), anti-CD44 BV421 (103040, Biolegend), and anti-CD62L PE (104407, Biolegend), followed by FCM analysis.

2.20
Statistical analysis
Data are presented as mean ± standard deviation (SD). Comparisons between two groups were performed using a two-tailed unpaired Student's t-test. For comparisons among multiple groups, one-way analysis of variance (ANOVA) followed by Tukey's multiple-comparisons test was applied. Statistical significance was set at ∗p < 0.05, ∗∗p < 0.01, and ∗∗∗p < 0.001.

3
Results
3.1
Optical properties of DTP
The compound DTP, formally known as meso-5-[ρ-diethylenetriaminepentaacetic acid-aminophenyl]-10,15,20-triphenylporphyrin, has the molecular formula C58H52N8O9 and a molecular weight of 1004.39 g/mol. Its chemical structure is shown in Fig. 1A. HPLC assessment confirmed over 98% purity for DTP (Fig. S1A). Fig. 1B shows the spectral profile of DTP, which displays the main peak at 420 nm, along with additional peaks at 516, 551, 592, and 647 nm. For this study, a 650 nm light source was selected to activate DTP, as light in the red spectral range offers superior tissue penetration compared to shorter wavelengths [19]. This allows for more efficient excitation of PSs within tumor tissues, providing an optimal balance between sufficient tissue penetration and effective photodynamic activation. Upon excitation at 420 nm, DTP demonstrates strong fluorescence emission at 654 nm in DMSO (Fig. 1C).
Effective PDT depends on ROS production, including •OH, 1O2, and O2•− [33]. The •OH generation capacity of DTP was assessed utilizing the chromogenic substrate TMB, which reacts with •OH to form ox-TMB with absorbance at 652 nm [34]. Following laser irradiation, the absorbance peak of ox-TMB increased with the irradiation over time, demonstrating a time-dependent •OH generation (Fig. 1D and S1B). In the absence of laser, no such change in absorption was observed. To evaluate 1O2 generation, we used DPBF, a probe exhibiting a 420 nm absorption band [35]. Fig. 1E shows that the absorbance of DPBF decrease after laser irradiation, verifying the production of 1O2 in a time-dependent manner. Additionally, O2•− generation was measured using DHE, a fluorescence probe that emits at 610 nm upon interaction with O2•− [36]. Fig. 1F shows a significant increase in fluorescence intensity with prolonged irradiation, confirming the sustained generation of O2•−. These results collectively demonstrate that DTP is an effective PS capable of generating substantial ROS, including •OH, 1O2, and O2•−, upon 650 nm laser irradiation.

3.2
Cell uptake of DTP
Before evaluating the antitumor effects of DTP-PDT, we first examined the cellular uptake of DTP by CLSM. Fig. 2A demonstrated colocalization between DTP-associated red fluorescence and lysosomal green fluorescence, achieving a Pearson's correlation coefficient (PCC) of 0.98 ± 0.02. In contrast, DTP showed less overlap with Mito-Tracker Green (PCC: 0.45 ± 0.07), indicating limited association with mitochondria. These results suggest that DTP primarily localizes in the lysosomes after internalization, with less distribution in the mitochondria or nucleus.
To further investigate DTP uptake, we assessed its intracellular fluorescence intensity in 4T1 cells under varying conditions of concentration and incubation time. Cells were exposed to varying concentrations of DTP for a fixed incubation time (24 h) and to a fixed concentration of DTP (1 μM) for different incubation period. As shown in Fig. 2B, DTP uptake increased with higher concentrations after 24 h of incubation. Regarding the relationship between incubation time and uptake, the cell number-normalized uptake increased steadily during the first 12 h. However, between 12 and 20 h, there was a transient decrease, followed by a subsequent increase at 24 h (Fig. 2C). The decrease between 12 and 20 h likely corresponds to a phase of higher cell density, during which competition for the extracellular DTP pool reduces the availability per cell. Additionally, changes in the cell state at higher density may decrease the average uptake activity. These factors contributed to the temporary reduction in normalized uptake rather than a true decrease in internalization capacity. Overall, DTP uptake is concentration-dependent and increases over time, supporting the use of 24 h incubation for subsequent experiments.

3.3
Cytotoxicity of DTP-PDT
Next, ROS levels in 4T1 cells were detected via the DCFH-DA probe, which generates green fluorescence upon oxidation by ROS [37]. Cells maintained in PBS or free DTP in the dark showed negligible fluorescence, indicating minimal basal ROS levels (Fig. 2D). After laser irradiation, DTP-treated cells exhibited significant green fluorescence, demonstrating that DTP effectively generates ROS under PDT conditions (Fig. 2D). Quantitative analysis revealed that compared to the PBS group, 100 and 200 nM DTP-PDT caused 29.6- and 187.9-fold increases in integrated fluorescence intensity, respectively (Fig. S2A). FCM analysis confirmed these findings (Fig. S2B and S2C).
Building on ROS production ability under laser, we systematically evaluated the in vitro therapeutic potential of DTP. Firstly, MTT assay demonstrated that DTP with laser irradiation induced a concentration-dependent reduction in cell viability due to ROS production (IC50 = 279.8 nM). In contrast, DTP in dark conditions showed minimal cytotoxicity (IC50 > 2 μM).
To further evaluate DTP's ability to induce tumor cell death, we performed a live-dead assay. As shown in Fig. 2F, negligible red fluorescence (PI-positive cells) was observed in both PBS and DTP-only groups. However, the DTP + L group exhibited significantly more red fluorescence, indicating extensive cell death. Notably, the 200 nM DTP + L group exhibited stronger red fluorescence compared to the 100 nM DTP + L group, demonstrating a dose-dependent photodynamic efficiency of DTP, consistent with the MTT assay results.
Finally, we assessed DTP's potential to induce apoptosis by FCM analysis using Annexin V-FITC/PI staining. No significant apoptosis was observed in the PBS or DTP-only groups (4.96 ± 0.83% and 6.5 ± 0.10%, respectively). In contrast, apoptosis rates significantly increased with higher DTP concentrations. The 100 nM and 200 nM DTP + L groups showed apoptosis rates of 24.4 ± 5.14% and 36.4 ± 3.90%, respectively (Fig. 2G and H). These results confirm that DTP is a highly effective PS, with low dark toxicity and strong photo-activated cytotoxicity.
To assess the generalizability of DTP-PDT's photodynamic effects, we performed preliminary test in the human TNBC cell line MDA-MB-231. Consistent with the results from the 4T1 cell line, DTP-PDT exhibited minimal dark toxicity under non-irradiated conditions and significantly inhibited cell proliferation upon 650 nm irradiation (IC50 = 276.0 nM, Fig. S2D). To further evaluate the biosafety of DTP-PDT, we conducted MTT assays in normal breast epithelial cell line MCF-10A. Compared to cancer cells, MCF-10A cells showed significantly less viability reduction at equivalent drug concentrations and light doses, suggesting that DTP-PDT exhibits low phototoxicity toward normal cells and indicating potential biosafety within the study's dosage range (Fig. S2E).

3.4
In vivo tumor accumulation of DTP in the 4T1 breast tumor model
We used real-time fluorescence imaging to assess DTP biodistribution and optimize laser irradiation timing. After intravenous injection, the DTP fluorescence signal in the tumor was detectable 2 h after injection, reaching peak intensity between 12 and 24 h post-injection, with a sustained signal up to 48 h (Fig. 3A and B). Based on these results, 12 and 24 h post-injection were selected as optimal time points for subsequent PDT treatments.
Ex vivo analysis at 24 h revealed that DTP accumulated predominantly in the liver (2.81 ± 0.47) and spleen (2.43 ± 0.47), which are rich in the reticuloendothelial system (RES), as well as in tumor tissue (2.22 ± 0.44) (Fig. 3C and D). This preferential accumulation in tumors is likely due to passive retention mechanisms, including the enhanced permeability and retention (EPR) effect, which is associated with incomplete tumor vasculature and restricted lymphatic drainage [38,39]. Additionally, some porphyrinoid PSs may bind to serum lipoproteins, and LDL/LDLR-mediated endocytosis could contribute to the observed tumor accumulation [40].

3.5
Antitumor efficacy of DTP in the 4T1 breast tumor model
We then evaluated the anti-tumor activities of DTP-PDT in vivo. Mice were treated with saline, DTP alone, or DTP with laser irradiation (DTP + L) (Fig. 3E). The body weights of these three groups at the experimental endpoint were not significantly different (p > 0.05, Fig. 3F) and no abnormal behavior of the experiment animal was observed during treatment, suggesting that DTP administration did not induce systemic toxicity. Both saline- and DTP-treated groups showed rapid tumor growth, with no statistical difference in tumor volume at the endpoint, indicating that DTP alone did not inhibit tumor progression (p > 0.05, Fig. 3G and H). In contrast, DTP + L group significantly inhibited tumor growth (p < 0.001), achieving remarkable tumor suppression rate of 80.64%.
We further analyzed the treatment response through histopathological evaluation of tumor tissue. Tightly packed tumor cells were seen in H&E staining of the saline and DTP groups, while the DTP + L group displayed apoptotic features, including chromatin condensation and nuclear shrinkage (Fig. 3J) [41]. Cell proliferation was then assessed using Ki-67 IHC [42]. Both the saline (15.06%) and DTP (13.41%) groups exhibited a high number of Ki-67 positive cells, indicating strong proliferative activity. In contrast, DTP + L group showed a significant decrease in Ki-67 expression (1.20%, p < 0.001), confirming a potent antiproliferative effect (Fig. 3K and L). Additionally, TUNEL assay demonstrated significant apoptosis in the DTP + L group with 42.44% TUNEL-positive cells compared to <1% in the saline and DTP-only groups (Fig. 3M and N), which was consistent with the in vitro apoptosis data. These results demonstrate that DTP preferentially accumulates in tumors and triggers tumor growth inhibition through apoptosis induction and cell proliferation suppression under irradiation.

3.6
Untargeted metabolomics reveals metabolic changes induced by DTP-PDT
Untargeted metabolomic profiling was conducted to assess the metabolic alterations induced by DTP-PDT. Representative total ion current (TIC) chromatograms for the control and DTP + L groups are shown in Fig. 4A and B. PCA and OPLS-DA revealed clear separation between the Control and DTP + L groups (Fig. 4C and D). The OPLS-DA model proved to be stable without overfitting, as confirmed by permutation tests (Fig. 4E). The differential metabolites are listed in Table S2 and shown in volcano plots (Fig. 4F). Pathway enrichment analysis indicated that DTP-PDT primarily affected amino acid-related pathways, including phenylalanine-tyrosine-tryptophan metabolism, tryptophan metabolism, and the biosynthesis of valine, leucine, and isoleucine, as well as primary bile acid biosynthesis (Fig. 4G and H).
Hierarchical cluster analysis revealed a distinct separation between the two groups in terms of the relative abundance levels of differential metabolites(Fig. 5A). Among differential metabolites, lysophosphatidylcholine (LysoPCs) and lysophosphatidyl-ethanolamine (LysoPEs) were generally upregulated, while certain glycerophospholipids (PG, PC, and PE) showed significant downregulation. Additionally, multiple amino acids and their derivatives, including tryptophan and certain indole metabolites, phenylalanine-related metabolites, and acetylcarnitine, exhibited significantly increased levels (Table S2). These results suggest that DTP-PDT induces significant changes in both lipid and amino acid metabolism. The metabolic network of potential differential metabolites is shown in Fig. 5B.

3.7
Targeted metabolomics reveals suppression of the Ido-Kyn-AhR axis by DTP-PDT
Given the significant alterations in the Trp metabolism observed in the untargeted analysis, we further validated these findings using targeted metabolomics in cells and tumor tissues. In cell samples, DTP-PDT significantly reduced Trp levels (p < 0.001), along with marked decreases in Kyn, QA, and 5-HT (Fig. 6A–D, p < 0.001), indicating suppressed Trp availability and overall metabolic flux. In tumor tissues, however, only Kyn levels were significantly decreased (p < 0.001), whereas changes in Trp, QA, and 5-HT were not significant (p > 0.05), suggesting a buffering effect in the TME (Fig. 6E and S3A-S3C). These results indicate that DTP-PDT primarily modulates the Trp-Kyn axis by suppressing Kyn production in vivo.
To determine whether the reduction in Kyn was due to changes in upstream enzyme expression or activity, we assessed the expression of IDO1 and TDO2 in tumor tissues through IHC. We observed no significant difference in the positive staining areas for IDO1 or TDO2 between the Control and DTP + L groups (Fig. S3D–S3G), suggesting that DTP-PDT does not markedly alter enzyme expression.
As protein expression does not fully reflect enzyme activity, we further examined Kyn production in an IFN-γ pretreated model to stimulate IDO1 activity in vitro. IFN-γ stimulation increased extracellular Kyn levels, while DTP-PDT significantly reduced IFN-γ-induced Kyn production (Fig. 6F). Additionally, IDO1 inhibitor markedly suppressed Kyn accumulation, and after pharmacological IDO1 inhibition, no further reduction in Kyn levels was observed with DTP-PDT treatment. These results indicate that DTP-PDT suppresses IFN-γ-induced Kyn production in an IDO1-dependent manner, despite unchanged IDO1 expression. We next examined the effect of DTP-PDT on Kyn levels in the human TNBC cell line MDA-MB-231. Similar to the results in 4T1 cells, DTP-PDT significantly reduced extracellular Kyn levels in IFN-γ-stimulated MDA-MB-231 cells (Fig. S3I). These findings support the generalizability of DTP-PDT's regulatory effects on the Trp–Kyn axis across species and cell types.
Kyn, as an endogenous ligand for the AhR, activates AhR signaling [43]. To assess this, we first evaluated AhR expression and its intracellular localization in tumor-infiltrating CD3+ T cells using immunofluorescence. In the DTP+L group, the fluorescence intensity of AhR in tumor-infiltrating CD3+ T cells was notably decreased compared to the Control group. Furthermore, AhR localization shifted markedly following treatment. In Control group, AhR was predominantly localized in the cell nucleus, indicative of its activated state [44]. In contrast, AhR signal in the treatment group exhibits a significantly reduction in nuclear accumulation, with its distribution shifting predominantly to the cytoplasm (Fig. 6G).
To further investigate the functional impact of AhR inhibition, CD3+ T cells were isolated from tumor tissues with purity over 90% as confirmed by FCM (Fig. S3H). qRT-PCR analysis revealed significant downregulation of classic AhR target genes, including Cyp1a1, Cyp1b1, and Ahrr, in the DTP-PDT-treated group compared to controls (Fig. 6H). These transcriptional changes align with the observed reduction in nuclear AhR, supporting the attenuation of AhR signaling.
We then assessed the impact of reduced AhR signaling on the tumor-infiltrating lymphocyte landscape by FCM [45]. Compared to the control group, the DTP + L group demonstrated a notable increase in CD8+ T cells infiltration, along with a substantial decrease in Treg cells (CD4+Foxp3+), indicating a reduction in tumor immunosuppression (p < 0.001, Fig. 6I–L).

3.8
Exogenous Kyn partially restores AhR signaling and attenuates immune remodeling induced by DTP-PDT
To provide direct evidence that DTP-PDT modulates the Kyn-AhR axis, we conducted a rescue experiment using exogenous Kyn. As expected, DTP-PDT markedly reduced intratumoral Kyn levels compared to the control group (p < 0.05). Kyn supplementation, however, partially replenished Kyn levels in the tumors treated with DTP-PDT (p < 0.001, Fig. S3J). Correspondingly, DTP-PDT reduced AhR nuclear localization in tumor-infiltrating T cells, while exogenous Kyn partially restored AhR nuclear translocation (Fig. 6M). Similarly, RT-qPCR analysis revealed that the expression of AhR target genes was reduced following DTP-PDT treatment (p < 0.05), but was partially recovered by Kyn supplementation (p < 0.01 and p < 0.05). Kyn alone increased the expression of cyp1a1 and cyp1b1 compared to controls (p < 0.01 and p < 0.05, Fig. 6N).
Moreover, Kyn supplementation also attenuated the immune remodeling induced by DTP-PDT. Specifically, it led to a reduced proportion of intratumoral CD8+ T cells (p < 0.05) and an increase in Treg cells (p < 0.01) compared to the DTP-PDT group (Fig. 6O–R). Collectively, these findings provide rescue evidence that DTP-PDT modulates the Kyn-AhR axis in the TME, contributing to the relief of tumor-associated immunosuppression.

3.9
Anti-metastasis effect of DTP-PDT in combination with anti-PD-L1 therapy in a breast cancer lung metastasis model
To evaluate the effect of DTP-PDT combined with α-PD-L1 therapy on lung metastasis, we established a metastatic model to mimic postoperative pulmonary metastasis in breast cancer patients (Fig. 7A). As shown in Fig. 7B and C, the model group had a much higher average radiance than the PDT monotherapy and combination therapy groups (p < 0.001). However, there was no significant difference between the model and the α-PD-L1 monotherapy groups (p > 0.05). Radiance in the α-PD-L1 and PDT monotherapy groups was 3.63-fold and 3.46-fold greater than that in the combination therapy group, respectively (p < 0.01). Lung weight assessment revealed that the lung weight ratio was significantly higher in the model, α-PD-L1, and PDT groups compared to the combination group (p < 0.001, Fig. 7D).
Regarding pulmonary metastatic nodules, the model group exhibited up to 60 nodules, which was significantly more than those in the combined therapy group (p < 0.001, Fig. 7E and F). There was no significant difference in the number of nodules between the α-PD-L1 and PDT groups (p > 0.05), although both were notably higher than the combination group (p < 0.001 and p < 0.01). These findings were confirmed by histological analysis. The model, α-PD-L1, and PDT groups showed extensive tumor cell infiltration and disruption of alveolar architecture were observed, while the combination group exhibited intact lung architecture with almost no metastatic lesions (Fig. 7G).
Immune analysis revealed that the combination group exhibited a markedly higher proportion of Tcm (CD3+CD8+CD44+CD62L+) in the spleen (29.10 %) compared to the model (1.50 %), α-PD-L1 (4.65 %), and PDT (10.30 %) groups (p < 0.001). The PDT monotherapy group also showed a significantly higher Tcm proportion than the model group (p < 0.05) (Fig. 7H and I). Survival analysis demonstrated that the combination group achieved a 60% survival rate at day 60, which was significantly superior to all other groups (Fig. 7J).
These results demonstrate that combining DTP-PDT with α-PD-L1 effectively induces a systemic antitumor immune memory, significantly suppresses lung metastasis upon tumor rechallenge, and improves survival.

4
Discussion
This study systematically assessed the antitumor potential of the novel photosensitizer DTP in breast cancer. Our findings show that DTP-PDT exhibits potent cytotoxic properties and reshapes the tumor immune microenvironment through metabolic reprogramming. Metabolomic analysis revealed that DTP-PDT significantly reduced the levels of the immunosuppressive metabolite Kyn within tumor tissues. This reduction was accompanied by impaired nuclear localization of AhR and suppression of AhR transcriptional activity in tumor-infiltrating T cells, indicating the attenuation of AhR signaling. Importantly, exogenous Kyn partially restored AhR signaling and attenuated the associated immune remodeling induced by DTP-PDT. Furthermore, combining DTP-PDT with PD-L1 blockade markedly suppressed lung metastasis and promoted the expansion of Tcm, suggesting a synergistic therapeutic effect and durable immune protection.
Metabolic reprogramming is a major driver of immune evasion in tumors. The Trp-Kyn pathway acts as a central factor in shaping the immunosuppressive TME [17]. Tumor cells upregulate IDO1 and TDO2, which deplete Trp and produce large amounts of Kyn [46,47]. This metabolic alteration creates nutritional stress for effector T cells [48,49] while activating the AhR pathway. AhR activation further promotes Treg differentiation, suppresses effector T cell and NK cell functions, and enhances immune checkpoint expression [[50], [51], [52]]. Therefore, targeting the Kyn-AhR axis is a promising strategy to restore immune activity and enhance the efficacy of immunotherapies [53]. Current strategies mainly rely on IDO1/TDO2 inhibitors or exogenous Kynureninase. However, clinical trials of IDO1 inhibitors have shown limited success and enzyme-based therapies hindered by poor stability and tumor-targeting [[54], [55], [56]].
Our study highlights that DTP-PDT offers a unique approach to regulate the Kyn-AhR metabolic axis. At the cellular level, PDT induces changes in Trp and several of its downstream metabolites. However, in the complex in vivo TME, PDT selectively, reduces Kyn levels with minimal impact on other metabolites. This selectivity suggests that DTP-PDT does not broadly suppress the entire Trp metabolic pathway but specifically targets Kyn, the metabolite with the most significant immunosuppressive activity. The discrepancy between in vitro and in vivo metabolic profiles suggests that PDT may primarily affect upstream regulatory enzymes, such as IDO1 and TDO2.
IHC staining of tumor tissues revealed no significant changes in IDO1 or TDO2 protein expression following PDT, indicating that DTP-PDT does not directly downregulate enzyme abundance. Instead, functional assays in IFN-γ-primed cells showed a marked reduction in Kyn levels after DTP-PDT. Notably, in the presence of an IDO1 inhibitor, DTP-PDT did not further reduce Kyn levels. These results suggest that DTP-PDT suppresses IDO1-dependent Trp-Kyn metabolic flux without affecting IDO1/TDO2 protein levels. This mechanism may offer advantages over direct IDO1 inhibitors by avoiding compensatory adaptations and enabling more tumor-localized modulation of Kyn in the TME.
Importantly, the reduction in Kyn levels was accompanied by impaired AhR signaling in tumor-infiltrating T cells. This was evidenced by decreased AhR nuclear translocation and reduced mRNA expression of Cyp1a1, Cyp1b1, and Ahrr. Cyp1a1 and Cyp1b1 represent canonical downstream targets of the AhR–ARNT transcriptional complex and serve as sensitive markers of AhR transcriptional activity [57]. Whereas Ahrr functions as a feedback inhibitor that is induced upon sustained AhR activation [58]. Therefore, the downregulation of these genes reflects the attenuation of AhR signaling. In the rescue experiment, exogenous Kyn partially restored AhR signaling and attenuated the DTP-PDT-induced increase in CD8+ T cells and decrease in Treg cells. These findings provide further direct evidence that DTP-PDT interferes with the Kyn-AhR axis and contributes to alleviating tumor-associated immunosuppression.
However, Kyn supplementation only partially reversed the effect of DTP-PDT, indicating that the immune-remodeling effects of DTP-PDT are not solely mediated through the Kyn-AhR axis. In addition to metabolic reprogramming, PDT is well known to generate ROS, causing tumor cell damage, promoting the release of tumor-associated antigens, and triggering broader inflammatory and immunogenic responses. These additional effects likely play a role in reshaping the TME [18].
Building on this immunological effect, we explored the combined therapeutic potential of DTP-PDT and PD-L1 blockade. This combination is biologically grounded in the cancer immunity cycle [59]. By reducing Kyn levels and suppressing AhR signaling, DTP-PDT alleviates the metabolic barriers that restrict T-cell activation, creating a more favorable immune environment [60]. Tumor cells often counteract immune activation by upregulating PD-L1, which inhibits T-cell function [61]. In this context, PD-L1 blockade removes this immune checkpoint. Therefore, the combination of DTP-PDT and PD-L1 blockade provide a dual positive effect by removing metabolic suppression and releasing immune brakes. Notably, a pronounced increase in Tcm cells was observed in the combination therapy group. Tcm cells are critical for long-term immune surveillance and preventing tumor recurrence [5,62]. Their expansion suggests the establishment of durable tumor-specific immune memory, which likely accounts for the ability of combination therapy to resist tumor rechallenge and suppress metastatic progression.
From a translational perspective, several challenges and opportunities must be addressed. Efficient light delivery to deeper or less accessible tumors remains a practical challenge. Advances in optical fiber-based illumination, interstitial light delivery, and image-guided PDT may help overcome these limitations and expand the applicability of PDT [63,64]. The immunometabolic mechanism identified in this study also opens opportunities for patient stratification. Tumors with elevated Kyn levels or enhanced AhR pathway activity may be particularly responsive to DTP-PDT-mediated immunomodulation. In this context, baseline activity of the IDO–Kyn–AhR axis could serve as a biomarker for patient selection. Furthermore, the ability of DTP-PDT to reduce Kyn-driven immunosuppression provides a strong rationale for its combination with PD-L1 blockade. These considerations highlight both the translational potential and future clinical optimization of DTP-PDT.
Despite these promising results, several limitations remain. First, the current study primarily utilized murine tumor models to assess immune effects. Although preliminary validation in the human TNBC cell line MDA-MB-231 supports the generalizability of the photodynamic and metabolic effects of DTP-PDT, further studies in additional human breast cancer subtypes and more clinically relevant models are needed to fully establish translational applicability. Second, the preferential tumor accumulation of DTP requires further investigation to validate its mechanism. Third, a systematic comparison with commercially available PSs under rigorously matched optical and dosimetric conditions will be important for positioning DTP among established PDT agents. Finally, while this study focused on the IDO–Kyn–AhR axis, the broader impact of PDT on metabolic networks warrants further investigation.

5
Conclusions
In conclusion, this study demonstrates that DTP-PDT exerts direct cytotoxic effects on breast cancer cells and reprograms the immunosuppressive TME by reducing intratumoral Kyn levels and attenuating AhR signaling, an effect that is partially restored by Kyn supplementation. Furthermore, combiningDTP-PDT with PD-L1 blockade significantly reduces tumor metastasis and promotes long-term immune memory. These findings strongly support the combination of PDT with ICIs as an effective strategy to overcome immunotherapy resistance in breast cancer.

CRediT authorship contribution statement
Yuetong Liu: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft. Ge Hong: Project administration, Software, Validation, Visualization, Writing – original draft. Tianjun Liu: Funding acquisition, Resources, Supervision, Writing – review & editing. Hong Liu: Funding acquisition, Resources, Supervision, Writing – review & editing.

Declaration of competing interest
The authors declare that there is no competing financial interests or personal relationships regarding the publication of this article.