Self-Assembled Cationic Phthalocyanine Nanophotosensitizer Drives Hypoxia-Tolerant Photodynamic Immunotherapy for HCC Authors: Fang-Cheng Jiang, Yue Qin, Li Gao, Su-Ning Li, Wei Su, Xue-Mei Sun, Xin Chen, Ying Wang, Zhen-Bo Feng, Ze-Feng Lai Published: 2026 ABSTRACT BackgroundPhotodynamic therapy (PDT) eliminates malignancies through spatially controlled reactive oxygen species (ROS) generation and has achieved clinical success in localized tumors. Nevertheless, its therapeutic potential is severely restricted by the poor aqueous solubility, suboptimal ROS generation efficiency, and oxygen dependency of conventional photosensitizers. To address these limitations, we developed cationic silicon phthalocyanine nanoparticles (c-SPNPs) by self-assembling morpholine-modified silicon phthalocyanine, which significantly improves aqueous solubility.ResultsUnlike conventional oxygen-dependent Type II PDT, c-SPNPs work primarily through Type I photochemical mechanism, generating ROS with reduced oxygen consumption. Mechanistically, c-SPNP-mediated PDT induces immunogenic cell death (ICD) in hepatocellular carcinoma (HCC) cells, as evidenced by the release of damage-associated molecular patterns (DAMPs), including surface-exposed calreticulin (CRT), extracellular ATP and HMGB1. These DAMPs promote dendritic cell maturation and enhance CD8+ T-cell infiltration within the tumor microenvironment. In murine HCC models, c-SPNP-mediated PDT effectively suppressed tumor growth while eliciting systemic antitumor immunity.ConclusionOur study establishes c-SPNPs as a multifunctional nanoplatform that synergizes oxygen-efficient PDT and immunotherapy, offering a promising strategy for hypoxic solid tumor treatment. Introduction Hepatocellular carcinoma (HCC), the sixth most common malignancy and third leading cause of cancer-related deaths globally, poses significant therapeutic challenges.1 While radical surgical resection remains the gold standard for HCC treatment, postoperative recurrence rates remain alarmingly high. Recent advances in systemic therapies, such as combining immune checkpoint inhibitors with molecular-targeted agents, have marginally improved 5-year survival rates to 22%.2,3 However, persistent limitations, including chemotherapy resistance, insufficient immunotherapy efficacy, and adverse effects associated with multimodal treatments, hinder their clinical utility.4,5 Recent advances underscore the use of integrative bioinformatics in dissecting molecular networks to identify hub genes as therapeutic targets in HCC.6–8 And artificially designed nanomedicines targeting the biological characteristics of malignant tumors have shown great success. These offer hope for humanity in defeating tumors.9 Photodynamic therapy (PDT), a non-invasive therapeutic strategy, employs photosensitizers (PSs) to generate cytotoxic reactive oxygen species (ROS) under targeted light irradiation.10,11 While PDT has demonstrated efficacy against drug-resistant bacterial infections,12 its application extends to superficial tumors, including skin, head and neck, early-stage obstructive lung, and non-muscle-invasive bladder cancers.13 The US Food and Drug Administration (FDA) has approved several PS agents, such as 5-aminolevulinic acid (ALA) and porfimer sodium, for dermatological oncology applications. The translational potential of PDT is anchored in a constellation of well-defined advantages. ROS production is confined to illuminated tumor sites, minimizing systemic toxicity.14 In addition, PSs in the bloodstream bind preferentially to low-density lipoproteins (LDLs), which tumor cells internalize actively due to their heightened metabolic demands, enhancing tumor selectivity.15 Emerging research highlights improved therapeutic outcomes when PDT is combined with modalities like immunotherapy, molecular-targeted agents, or photothermal therapy, creating synergistic effects that disrupt multiple cancer pathways simultaneously.16–18 However, the hypoxic tumor microenvironment significantly limits the efficacy and clinical application of PDT.19 PDT is broadly categorized into two types based on the ROS generation mechanisms: Type I and Type II. Type I PDT generates superoxide radicals (O2•−) through electron-transfer (ET) reactions, whereas Type II PDT produces singlet oxygen (1O2) via electron energy transfer (EET) processes. Crucially, Type I PDT is less dependent on oxygen, making it better suited for hypoxic tumor microenvironments, while Type II PDT relies on high oxygen levels.20,21 Most existing PSs utilize the Type II mechanism, because ET reaction is more difficult to attain than EET process.22 Developing Type I PSs for deep solid tumors remains hindered by multiple factors, including poor aqueous solubility, limited tissue penetration, and suboptimal ET efficiency.23 Addressing these limitations is critical to advance PDT applications in oxygen-deficient tumors. Consequently, researchers are increasingly exploring type I reactive oxygen species (ROS) for targeting cancers within hypoxic tumor microenvironments (TMEs). For instance, Feng et al shifted the photosensitizer’s mechanism from type II to type I by modifying its molecular structure to enhance ET capability, thereby improving its efficacy under hypoxic conditions.24 Concurrently, advances in addressing photosensitizer’s poor water solubility such as nanoparticle encapsulation or chemical conjugation, have optimized their bioavailability. Despite these advancements, the therapeutic efficacy of oxygen-efficient and water-soluble photosensitizers in hepatocellular carcinoma (HCC) has been rarely reported so far. Herein, we introduce a novel second-generation nano-photosensitizer based on our previous work.25 The synthesis involves the self-assembly of bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) diiodide into stable nanoparticles (c-SPNPs). Modified with morpholine, the N element in morpholine structure forms cationic groups with iodide, thus improving its water solubility. In addition, the cationized 3-(N-methyl-N-morpholinyl) propoxy substituent makes it a typical amphiphilic molecule, which is beneficial to assemble into spherical nanoparticles. In vitro, c-SPNP-mediated PDT efficiently generated type I ROS to induce HCC cell death. Consequently, dying cells released damage-associated molecular patterns (DAMPs) including adenosine triphosphate (ATP), high-mobility group box 1 protein (HMGB1), and calreticulin (CRT). This process facilitated activation of antigen-presenting cells (APCs). In vivo validation in immunocompetent HCC models confirmed the multilayered efficacy of c-SPNP-mediated PDT, and primary tumors were durably controlled. Meanwhile, the therapy triggered ICD, primed APCs, and expanded tumor-infiltrating CD8⁺ T cells. Most strikingly, this localized intervention abrogated the outgrowth of distant metastasis, providing direct evidence of systemic immune activation.26 Collectively, these findings establish c-SPNP-mediated PDT as a promising therapeutic strategy for advanced HCC. Materials and Methods Synthesis and Characterization of Silicon Phthalocyanine Self-Assembled Nanoparticles (c-SPNPs) The c-SPNPs were prepared as reported previously.27 Briefly, silicon (IV) phthalocyanine dichloride (100 mg), 3-(4- morpholinyl) −1- propylamine (70 mg), and NaH (35 mg) were dissolved in 15 mL of toluene and refluxed for 2 d under the protection of N2. The cooled solution was vacuum distilled to remove the solvent. The residue was purified by silica gel column chromatography, and chloroform was used as eluent to obtain bis [3-(morpholinyl) propoxy] phtalocaninato silicon. Subsequently, 18 mg of bis [3-(morpholinyl) propoxy] phtalocaninato silicon and 8.0 mL of CH3I were mixed in 7.0 mL of chloroform and stirred at 45 °C for 12 h. Then, the dark blue powder was filtered and washed with CHCl3, ether and acetone, respectively. The obtained bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) diiodide was dissolved in dimethyl sulfoxide (DMSO) and then injected into ultrapure water to prepare c-SPNPs. The morphology and size of the as-prepared c-SPNPs were determined by transmission electron microscopy (TEM) using a H7500 transmission electron microscope (Hitachi, Tokyo, Japan). The zeta potential of c-SPNPs dispersed in pure water was measured using a Zetasizer Nano ZS (Malvern Instruments, Malvern, UK). And the UV-Vis absorption spectrum was measured using a Cary 3500 UV-Vis spectrophotometer (Agilent, CA, USA) over a wavelength range of 300–800 nm. 1H NMR spectra and 13C NMR spectra were detected using a 600 MHz AVANCE III HD600 spectrometer (Bruker, Germany) in dimethyl sulfoxide (DMSO). Mass spectra were determined using a Xevo G3 QTof mass spectrometer (Waters, MA, USA). O2•− and Singlet Oxygen Sensor Green (SOSG) Detection Dihydroethidium (DHE) was employed to quantify superoxide anion (O2•−) generation.28 c-SPNPs were mixed with a DHE solution containing circulating tumor DNA (ctDNA, 250 μg/mL) and irradiated with a 660 nm laser (0.5 W/cm2) for 20 min. Fluorescence intensity (λex = 490 nm) was measured using an F-4600 fluorescence spectrometer (Hitachi, Tokyo, Japan). The generation of singlet oxygen (1O2) was assessed using SOSG reagent. c-SPNPs were mixed with an SOSG solution containing circulating tumor DNA (ctDNA, 250 μg/mL) and protected from light. The mixture was then irradiated with a 660 nm laser (0.5 W/cm2) for 20 min. The resulting fluorescence intensity (λex = 504 nm, λem = 525 nm) was measured using the F-4600 fluorescence spectrometer. Cell Lines The human hepatocellular carcinoma (HCC) cell lines Huh7 and MHCC97H, along with the C57BL/6J mouse-derived HCC cell line Hepa 1–6, were purchased from the Cell Bank of the Chinese Academy of Sciences. Cells were maintained at 37 °C in a humidified 5% CO2 incubator. All cell lines were cultured in DMEM (Gibco, 8123102) supplemented with 10% fetal bovine serum (FBS; Biological Industries, 04–001-1A) and 1% penicillin-streptomycin (Solarbio, P1400). Medium was replaced every 24 h. Detection of ROS Generation Approximately 2×105 Huh7 and MHCC97H cells were seeded in 35-mm confocal dishes and incubated overnight, respectively. The medium was removed and the cells were incubated with or without 500 nM c-SPNPs solution (diluted in DMEM basal medium) for 4 h. Washed with phosphate buffer solution (PBS) for three times, the cells was treated with DCFH-DA (1 mM) and incubated at 37 °C for 30 min. After three times’ washing with PBS, the cells in the c-SPNPs+Laser and Laser (without c-SPNPs) groups were irradiated with a 660 nm laser (0.5 W/cm2) for 5 min. Intracellular ROS levels were examined using an Olympus FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan). Subsequently, the cells were trypsinized and harvested, and their DCF fluorescence intensity (λex/λem = 488/525 nm) were determined using a BD Accuri™ C6 Plus flow cytometer (BD, NJ, USA). Dark Toxicity and Phototoxicity Assessment of c-SPNPs Huh7 and MHCC97H cells in logarithmic growth phase were seeded into 96-well plates at 1×104 cells/well. After 24 h incubation, c-SPNPs were added to achieve gradient concentrations of 0, 200, 400, 600, 800 and 1000 nM. For dark toxicity assessment, plates were returned to the incubator for 24 h, followed by addition of 10 μL CCK-8 reagent (Vazyme, A311-01) per well. After 1 h incubation, absorbance at 450 nm was measured using a Synergy H1 microplate reader (BioTek, VT, USA). For phototoxicity assessment, cells were treated with gradient concentrations of c-SPNPs for 4 h, washed with PBS, refilled with 100 μL of complete medium, and irradiated with a 660 nm laser (0.5 W/cm2) for 5 min. After 24 h incubation, cell viability was quantified via CCK-8 assay as above. Scratch Wound-Healing Migration Assay Huh7 and MHCC97H cells in logarithmic growth phase were separately seeded into 6-well plates until reaching 80–90% confluency. The medium was removed and the cells were incubated with or without c-SPNPs (500 nM) for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. Mechanical scratches were generated perpendicular to the plate bottom using a sterile 100 μL pipette tip. Debris was aspirated, and cells were washed three times with PBS before replacing the medium with serum-free medium. Initial scratch images were captured using a Leica DMi8 inverted fluorescence microscope (Leica, Wetzlar, Germany). Wound closure rates were quantified by re-imaging scratches at 48 h and 72 h post-treatment, followed by analysis of relative wound area reduction using ImageJ (v1.52a, Java 1.8.0). Transwell Assay Huh7 and MHCC97H cells were separately seeded into 6-well plates until reaching 80–90% confluency. The medium was removed and the cells were incubated with or without c-SPNPs (500 nM) for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. Cells were then harvested for subsequent assays. A mixture (1:8) of Matrigel matrix (Corning, 356234) and DMEM medium was prepared, and 60 μL was added to each transwell chamber (JET, TCS003024). Chambers were transferred to a 24-well plate (JET, TCS003024) and incubated at 37 °C for 30 min to solidify. Harvested cells were resuspended in serum-free medium, and 4 × 104 Huh7 and 8 × 104 MHCC97H cells were separately seeded into each transwell chamber. The lower chamber was filled with 500 μL of medium containing 10% FBS. After 48 h (Huh7) or 96 h (MHCC97H), invaded cells were stained and quantified while non-invaded cells were wiped off. Chambers were washed with PBS, fixed with methanol (30 min), air-dried, and stained with 1% crystal violet (Solarbio, G1062) for 12 h. Excess stain was removed by PBS washes, and images were captured. Invaded cells were quantified using ImageJ (v1.52a, Java 1.8.0). Cell Proliferation Assay Huh7 and MHCC97H cells were separately seeded into 6-well plates until reaching ~80% confluency. After aspirating the medium and washing twice with PBS, the cells were incubated with or without c-SPNPs (500 nM) for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. Subsequently, cells were trypsinized, resuspended at 3 × 104 cells/mL, and seeded into 96-well plates (100 μL/well, 6 replicates per group). After 24, 48, 72, and 96 h incubation, 10 μL CCK-8 reagent was added to each well and incubated for 1 h. Absorbance at 450 nm was measured using the Synergy™ H1 microplate reader (BioTek, VT, USA). Proliferation curves were plotted using mean absorbance values. Annexin V-APC/7-AAD Staining Annexin V-APC/7-AAD apoptosis kit (Multi Sciences, AP105-100) was employed to assess the apoptosis following the manufacturer’s protocol. Approximately 2×105 Huh7 cells or MHCC97H cells were seeded into 6-well plates and cultured overnight. Washed twice with PBS, the cells were incubated in the medium with or without c-SPNPs (500 nM) for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. After 8 h incubation, the cells were harvested, washed with ice-cold PBS, and resuspended in binding buffer (1 × 106 cells in 500 µL binding buffer). The cells were stained with Annexin V-APC (5 µL) and 7-AAD (10 µL) for 5 min at room temperature in the dark. Subsequently, each sample was washed and resuspended with PBS, filtered through a 400-mesh sieve, and analyzed by flow cytometry using a BD Accuri C6 Plus flow cytometer (BD, NJ, USA). Extracellular ATP Measurements An ATP detection assay kit (Servicebio, G4309-48T) was used to measure extracellular ATP content. Briefly, approximately 1×104 Huh7 cells or MHCC97H cells per well were seeded in 96-well plates and incubated for 24 h. After removing the medium, the cells were treated with or without c-SPNPs (500 nM) for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. After 24 h, the cell supernatant of each well was collected after centrifugation, and the ATP detection enzyme reagent and detection buffer were added to the working solution at a ratio of 1:50 and kept at room temperature for 5 min. 100 μL ATP detection working solution and 20μL band detection sample were added to each well of 96-well plate. The fluorescence value was detected using a BioTek Synergy™ H1 microplate reader (BioTek, VT, USA). Immunofluorescence Imaging of CRT Approximately 2×104 Huh7 cells or MHCC97H cells were seeded into 35-mm confocal dishes and cultured for 24 h. After discarding the medium, the cells were exposed to 2 mL DMEM medium with or without 500 nM c-SPNPs for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. After 6 h incubation, the medium was discarded and the cells were washed twice with PBS, fixed with 4% paraformaldehyde for 30 min, washed twice with PBS, and permeabilized in 0.3% Triton X-100 (Servicebio, GC204003-100mL) for 20 min. The fixed cells were incubated with the rabbit anti-human calreticulin (CRT) antibody (1:200 dilution; Proteintech, 27298-1-AP) overnight at 4 °C. Washed with PBS, the cells were incubated with the secondary antibody CoraLite®488-conjugated goat anti-rabbit IgG (H+L) (1:200 dilution; Proteintech, SA00013-2) for 2 h. Subsequently, the cells were washed thrice with PBS, and incubated with 1 mL Hoechst 33342 solution (1 mg/mL; Solarbio, C0031) at room temperature for 10 min. Washed with PBS, the cells were imaged with an Olympus FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan). The CoraLite®488 was excited at 488 nm and its fluorescence was monitored at 515 nm. The Hoechst 33342 was excited at 346 nm and its fluorescence was monitored at 460 nm. Extracellular HMGB1 Measurements Extracellular levels of HMGB1 were measured by an Enzyme-linked immunosorbent assay (ELISA). ELISA kits for HMGB1 were used according to manufacturer instructions (JONLNBIO, JL13693). Approximately 1×104 Huh7 cells or MHCC97H cells were seeded into 35-mm confocal dishes and cultured for 24 h. After removing the medium, the cells were exposed to 2 mL DMEM medium with or without 500 nM c-SPNPs for 4 h, followed by irradiation with a 660 nm laser (0.5 W/cm2) for 5 min. Incubated for 24 h, the supernatants were collected for the measurement of extracellular HMGB1 content. Briefly, 50 µL of samples (a mixture of supernatant (10 µL) and diluent (40 µL)), and control (diluent only) were separately added into different wells of the plate. Except for the control wells, 100 µL of HRP-antibody solution was added into each well of the standard and sample wells. Incubated at 37 °C for 60 min, the liquids were removed, and the wells were washed with a washing buffer for 5 times. Substrate A (50 μL) and Substrate B (50 μL) were added into each well (including the control) and the plate was incubated at 37 °C in the dark for 15 min. Finally, 50 μL of stop solution was added to each well, and the optical density (OD) of each well was measured at 450 nm using a BioTek Synergy™ H1 microplate reader (BioTek, VT, USA). We detected the standards and generated a standard curve. The R2 value of the standard curve was 0.998, which could be used for subsequent calculation of protein concentrations. Animal Experiments The care and treatment of the animals were in accordance with the guidelines established by the Ministry of Science and Technology of China on the Humane Care and Use of Laboratory Animals, and were approved by the Institutional Animal Care and Use Committee (IACUC) of Guangxi Medical University (Guangxi, People’s Republic of China; approval No.202310203; 202310004). Wild-type AB strain zebrafish were purchased from the China Zebrafish Resource Center (CZRC) and housed in the zebrafish facility of School of Pharmacy, Guangxi Medical University. Adults were maintained at 28 ± 0.5 °C in system water (conductivity: 480–510 μS/cm; pH: 6.9–7.2) under a 14-h light:10-h dark cycle, with males and females reared separately. For breeding, pairs (one male and one female) were placed in a mating tank equibbed with a divider on the evening prior to embryo collection. Dividers were removed the following morning, and embryos were collected at 1 hour post-fertilization (hpf), transferred to a temperature-controlled incubator (28 °C) with identical light cycles, and monitored for viability. Nonviable embryos were promptly discarded. Zebrafish larvae were sustained by yolk sac nutrients until 9 days post-fertilization (dpf). C57BL/6J mice (20–30 g) were obtained from the Guangxi Medical University Animal Experiment Center and housed in a Specific Pathogen Free (SPF) facility with ad libitum access to food and water. All protocols strictly followed ethical standards approved by Guangxi Medical University’s Institutional Animal Care and Use Committee (IACUC), incorporating environmental enrichment and stress-minimization practices. Post-experimentation, mice were humanely euthanized via isoflurane overdose, with carcasses disposed of in compliance with institutional biosafety guidelines. Isolation and Differentiation of Bone Marrow-Derived Dendritic Cells (BMDCs) from C57BL/6J Mice Eight-week-old female C57BL/6J mice were euthanized via terminal isoflurane anesthesia, and tibias and femurs were aseptically excised. Muscle tissue was dissected from the bones, which were then sterilized in 75% ethanol (5 min), rinsed twice with PBS, and transected at both epiphyses. Marrow was flushed from the medullary cavities using PBS-filled syringes until the cavities turned translucent. The collected marrow suspension was filtered through a 70-μm strainer (Biologix, 15–1070) to remove debris and residual bone fragments. Red blood cells (RBCs) were lysed with RBC lysis buffer (Solarbio, R1010). The cells were resuspended in RPMI 1640 medium (Gibco, C11875500BT) at a density of 1×106 cells/mL and seeded into 24-well plates. The cells were cultured with recombinant murine GM-CSF (20 ng/mL; Servicebio, GB300026-M-10ug) and IL-4 (10 ng/mL; Abways, PY2575) in a humidified 5% CO2 incubator at 37°C. Half-medium changes were performed every 48 h, with cytokine ementation to maintain original concentrations. Mature bone marrow-derived dendritic cells (BMDCs) were harvested on day 5. Assessment of Murine Bone Marrow-Derived Dendritic Cell Activation in vitro Hepa 1–6 cells (C57BL/6J-derived HCC line) were exposed to the DMEM medium with or without 500 nM c-SPNPs for 4 h, and transferred to transwell inserts (JET, TCS003024). Inserts were placed into 24-well plates containing pre-cultured BMDCs in the lower compartments, and Hepa1-6 cells in the upper inserts were irradiated with a 660 nm laser (0.5 W/cm2) for 5 min. After 24 h incubation, BMDCs were harvested, washed with PBS, and resuspended in 96.5 μL PBS. Cells were stained with 2.5 μL CD80-PE (BioLegend, PE anti-mouse CD80 antibody, 104707/50μg) and 1 μL CD86-FITC (BioLegend, FITC anti-mouse CD86 antibody, 105005/50μg) at 4 °C in the dark for 30 min. Following PBS washes, cells were filtered through a 40-μm cell strainer and analyzed by flow cytometry using a BD Accuri C6 Plus flow cytometer (BD, NJ, USA). Anticancer Efficacy of c-SPNP-Mediated PDT in Zebrafish Xenograft Tumor Models Five hundred Huh7 cells labeled with Dil dye (Servicebio, G1705) were microinjected into the yolk sac of each zebrafish larva (3 dpf) as reported.29–31 At 24 hours post injection (hpi), tumor-bearing zebrafish larvae were identified using a Leica M205FA fluorescence stereo microscope (Leica, Vizsla, Germany) and selected for subsequent experiments, with baseline fluorescence intensity recorded. The selected larvae were randomized into four groups: control (no treatment), c-SPNPs (intratumoral injection of 10 nL 1.0 μM c-SPNPs), Laser (660 nm laser irradiation 0.5 W/cm2, 5 min); c-SPNPs+Laser (c-SPNPs injection followed by 660 nm laser irradiation 0.5 W/cm2 for 5 min) groups. Laser irradiation was performed 4 h post-injection. At 24 h post-treatment, fluorescence intensity at the tumor site was re-evaluated to quantify therapeutic efficacy. Immune Response Induced by c-SPNP-Mediated PDT in vivo Hepa 1–6 cells (C57BL/6J-derived HCC line) in logarithmic growth phase were incubated with c-SPNPs for 4 h, followed by 10 min of 660 nm laser irradiation (0.5 W/cm2). Cells were then trypsinized, resuspended at 1×107 cells/mL, and prepared for vaccination. Untreated Hepa 1–6 cells were parallelly harvested, adjusted to the same concentration, and snap-frozen in liquid nitrogen as controls. Twelve 4- to 6-week-old C57BL/6J mice were randomly assigned into two groups (n = 6 per group). Mice in the vaccine group were subcutaneously injected with 200 μL of c-SPNP-laser-treated cells (2×106 cells) into the left flank, whereas control mice received an equal volume of untreated cells (2×106 cells) via identical administration. One week later, all mice were challenged with 2×106 Hepa 1–6 cells in the right flank. Tumor volumes were measured every 48 h using calipers. On day 12, mice were euthanized under terminal isoflurane anesthesia, and tumors were excised for macroscopic and histopathological analysis. Anticancer Efficacy of c-SPNP-Mediated PDT in Mouse Tumor-Bearing Models Twenty 6 to 8-week-old C57BL/6J mice received subcutaneous injections of 2×106 untreated Hepa 1–6 cells in the right flank. At 72 h post-transplantion (hpt), 100 μL c-SPNPs (1.0 μM) were intratumorally injected into the tumor in both c-SPNPs and c-SPNPs+Laser groups while the laser group received no nanoparticle administration. After 4 h, tumors in the c-SPNPs+Laser and Laser groups were irradiated with a 660 nm laser (0.5 W/cm2) for 10 min. Tumor volume was measured every 48 h using calipers. Mice were euthanized via terminal isoflurane anesthesia on day 8 post-treatment, and tumors were excised for histopathological analysis. Quantification of Tumor-Infiltrating DCs and CD8+ T Cells in C57BL/6J Xenografts Tumor tissues from mice were mechanically dissociated into single-cell suspensions using a tissue homogenizer. Cells were washed twice with ice-cold PBS (centrifugation: 1000 ×g, 5 min) and resuspended in 100 μL of PBS. Cell suspensions were stained with 2.5 μL of PE-conjugated anti-mouse CD80 (BioLegend, 104707) and FITC-conjugated anti-mouse CD86 (BioLegend, 105005) antibodies per tube. After gentle vortexing, samples were incubated at 4°C for 30 min protected from light. Unbound antibodies were removed by centrifugation (3000 ×g, 5 min) and two PBS washes. Cells were resuspended in 400 μL of ice-cold PBS, filtered through a 40-μm cell strainer, and analyzed on a BD Accuri™ C6 Plus flow cytometer. Parallel samples were processed identically using PE-conjugated anti-mouse CD3 (Thermo Fisher, 12–0032-82) and FITC-conjugated anti-mouse CD8 (Multi Sciences, F2100801) antibodies to assess tumor-infiltrating lymphocytes. ALT and AST Detection We measured alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels in stored serum samples collected from individual mice, using commercially available kits purchased from JONLNBIO (Cat# JL-T0873-96, JL-T0875-96). The procedures were strictly performed according to the manufacturer’s instructions. Enzyme activity of ALT and AST was finally measured using a microplate reader (BioTek, VT, USA), and enzymatic activity differences between groups were compared. Hematoxylin and Eosin (HE) Staining and Immunohistochemical (IHC) Staining Tumor tissues, heart, liver, spleen, lung, and kidney tissues were separated from mice and fixed with 4% paraformaldehyde. Subsequently, these fixed tissues were dehydrated, embedded in paraffin, sliced (thickness, 4 μm) and routine staining (HE). Slices for IHC analyses were deparaffinized with xylene. Endogenous peroxidase was blocked with 0.3% H2O2 for 10 min. Sections were treated with blocking goat serum for 20 min at 37°C. Then, sections were incubated overnight at 4°C with antibodies against CD8, GZMB, CD86, and CD206 (1:100 dilution; Hunan Aifang Biotechnology, Changsha, People’s Republic of China), followed by incubation with biotin-linked secondary antibody (Hunan Aifang Biotechnology, Changsha, People’s Republic of China) for 10 min at 37 °C and then with peroxidase-labeled streptavidin for 10 min at 37 °C. Finally, sections were incubated with diaminobenzidine substrate chromogen solution (ZSGB-BIO) for 3 min and counterstained with hematoxylin. Slices were analyzed under a microscope (ECLIPSE CI; NIKON). Statistical Analysis For each set of results, independent determinations were repeated at least thrice, and data are described as the mean ± standard deviation. The statistical analysis and visualization of this study are completed using GraphPad Prism v8.0. Normality test was carried out on multiple independent sample data, and nonparametric test (Kruskal Wallis test) was used when the data did not obey the normal distribution. When multiple independent sample data obey normal distribution, variance homogeneity test was carried out. When the variance between multiple independent samples was homogeneous, analysis of variance (ANOVA) was used, otherwise Welch test was conducted. All data were expressed as the mean ± standard deviation (SD, error bars). P-value < 0.05 was considered to be statistically significant (ns: no significance, *P < 0.05, **P < 0.01, ***P < 0.001, and ***P < 0.0001). Results and Discussion Characterization of Self-Assembled c-SPNPs The c-SPNPs were prepared as described in our previous study.27 Briefly, bis[3-(N-morpholinyl) propoxy] phthalocyaninato silicon (IV) is synthesized and subsequently cationized with methyl iodide to yield amphiphilic bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) diiodide (structural formula: Figure 1A). The cation-substituted compound is dissolved in DMSO and injected into pure water, where it assembles into c-SPNPs. The 1H NMR spectra, 13C NMR spectra, and mass spectra of amphiphilic bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) diiodide were shown in Figures S1–S3. Figure 1B displayed the UV-Vis spectrum of c-SPNPs. The absorption peak at 680 nm, which is the characteristic Q-band of the cation-substituted phthalocyanine, is weakening when forming self-assembly particles, as previously reported.27 Based on the UV-Vis results, a wavelength of 660 nm which lies within the effective absorption range was selected for PDT. Figure 1Characteristic of bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) (SiPc) diiodide and cationic silicon phthalocyanine nanoparticles (c-SPNPs). (A) Chemical structure of SiPc diiodide. (B) UV‑Vis absorption spectrum of c-SPNPs. (C) Superoxide anion (O2•−) detection using dihydroethidium fluorescence probe, comparing c-SPNPs with H2O control. (D) Representative transmission electron microscopy (TEM) image of c-SPNPs (scale bar: 500 nm). (E) Size distribution of c-SPNPs based on TEM analysis. (F) Zeta potential of c-SPNPs; Stability of c-SPNPs at 1 μM (G) and 10 μM (H) in water and serum, monitored over time.An eight-panel scientific figure labeled A to H summarizes properties of a silicon phthalocyanine formulation and its nanoparticles. Panel A is a line drawing of the molecule with a central phthalocyanine ring and two charged side chains. Panel B plots absorbance versus wavelength (about 300 to 800 nm) with three curves; the nanoparticle sample in water shows a prominent peak near the red/near IR region compared with the molecular sample in DMSO and a DMSO baseline. Panel C is a time-course line graph of fluorescence intensity for a superoxide-sensitive probe: the red curve (nanoparticles) rises quickly and plateaus high, while the blue water control stays nearly flat. Panel D is a TEM micrograph with scattered dark dots representing particles; a 500 nm scale bar appears in the corner. Panel E is a histogram of particle diameters (roughly 16 to 40 nm) with the tallest bars around the mid 20 nm range and a reported mean near 25.8 nm. Panel F is a bar chart of zeta potential across three measurements, clustered around +39 mV. Panels G and H are photo series of microcentrifuge tubes at 0, 3, 6, 12, 24 and 48 hours for two concentrations (1 microM and 10 microM), showing persistent coloration and minimal visible precipitation over time in different media.Multi-panel figure showing spectra, ROS signal, TEM, size/zeta plots and vial stability photos.Note: (1) SiPc in serum; (2) SiPc in water; (3) c-SPNPs in serum; (4) c-SPNPs in water. Characteristic of bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) (SiPc) diiodide and cationic silicon phthalocyanine nanoparticles (c-SPNPs). (A) Chemical structure of SiPc diiodide. (B) UV‑Vis absorption spectrum of c-SPNPs. (C) Superoxide anion (O2•−) detection using dihydroethidium fluorescence probe, comparing c-SPNPs with H2O control. (D) Representative transmission electron microscopy (TEM) image of c-SPNPs (scale bar: 500 nm). (E) Size distribution of c-SPNPs based on TEM analysis. (F) Zeta potential of c-SPNPs; Stability of c-SPNPs at 1 μM (G) and 10 μM (H) in water and serum, monitored over time. Evidenced by O2•− detection using DHE as the fluorescent probe, c-SPNPs exhibited efficient ROS generation through a hypoxia-tolerant type I mechanism (Figure 1C). Consistent with efficient photodynamic activation, the ROS signal attained a plateau at 5 min following 660 nm laser irradiation. Mechanistic studies using SOSG probes indicated that c-SPNPs primarily rely on a Type I photodynamic pathway in hypoxic environments, although distinct signals for Type II singlet oxygen were also observed (Figure S4). Comparative analysis of fluorescence intensities confirmed that superoxide anion generation was predominant, underscoring the Type I-dominant mechanism. Further investigation using DHE probes demonstrated that c-SPNPs maintained robust ROS generation capability under hypoxic conditions, although it was slightly lower than that under normoxic conditions (Figure S5). This property makes c-SPNPs particularly suitable for the treatment of hypoxic deep solid tumors. TEM observation revealed the well-defined morphology and uniform size distribution of c-SPNPs, with an average diameter of approximately 25 nm (Figure 1D and 1E). DLS measurements indicate an average hydrodynamic diameter of approximately 400 nm in aqueous dispersion (Figure S6). Notably, the c-SPNPs aqueous dispersion exhibited a highly positive zeta potential of +39.20 mV (Figure 1F), which can be ascribed to the presence of morpholinyl-derived cationic groups on the nanoparticle surface. Finally, stability assessment confirmed that c-SPNPs retained their structural integrity in both water and serum (Figure 1G and 1H). These results suggest that c-SPNPs possess favorable colloidal stability under physiological conditions, which is crucial for reliable in vivo delivery and consistent PDT efficacy. c-SPNPs Induce HCC Cell Death via ROS Production in vitro We first evaluated the cytotoxicity of c-SPNPs against HCC cells under both dark and laser-irradiated conditions using the CCK8 assay. Building on the our previous observation of significant antimicrobial activity at 100 nM,27 we extended the concentration range to 0–1000 nM to account for the higher ROS tolerance of cancer cells. As shown in Figure 2A, c-SPNPs exhibited negligible dark cytotoxicity to Huh7 cells at concentrations up to 1000 nM. However, 660 nm laser irradiation induced obvious cytotoxicity to Huh7 cells dose-dependently (Figure 2B). A similar trend was observed in MHCC97H cells, which remained viable under dark conditions (Figure 2C) but underwent significant cell death upon irradiation in a dose-dependent manner (Figure 2D). To further elucidate the mode of cell death, flow cytometry analysis was performed. As shown in Figure 2E–2G, treatment with 500 nM c-SPNPs combined with laser irradiation significantly increased the apoptosis rates of both Huh7 and MHCC97H cells (P < 0.05). Figure 2Cell death induced by c-SPNPs‑based PDT in HCC cells. (A) Viability of Huh7 cells treated by c-SPNPs at 0–1000 nM without laser irradiation. (B) Viability of Huh7 cells treated by c-SPNPs at 0–1000 nM with laser irradiation. (C) Viability of MHCC97H cells treated by c-SPNPs at 0–1000 nM without laser irradiation. (D) Viability of MHCC97H cells treated by c-SPNPs at 0–1000 nM with laser irradiation. (E) Apoptosis detection of Huh7 and MHCC97H cells using flow cytometry. Statistical analysis of apoptosis ratio of Huh7 (F) and MHCC97H (G) cells. *P < 0.05, **P < 0.01, and ****P < 0.0001. P values are determined by Kruskal–Wallis test (F) and Welch’s ANOVA (G) test for multiple comparisons, n≥3.The image A showing a bar chart labeled, Huh7, P equals 0.2016. The x axis label is c-SPNPs with tick labels 0, 200, 400, 600, 800, 1000 nM. The y axis label is Cell viability rate with range 0.0 to 1.5. Bars at 0, 200, 400, 600, 800, 1000 nM are near 1.0. The image B showing a bar chart labeled, IC50 Huh7 equals 138.0 nM. The x axis label is c-SPNPs with tick labels 0, 200, 400, 600, 800, 1000 nM. The y axis label is Cell viability rate with range 0.0 to 1.5. Bars decrease from about 1.0 at 0 nM to about 0.4 at 200 nM, about 0.2 at 400 nM, about 0.1 at 600 nM and near 0.0 at 800 and 1000 nM. The image C showing a bar chart labeled, MHCC97H, P equals 0.0688. The x axis label is c-SPNPs with tick labels 0, 200, 400, 600, 800, 1000 nM. The y axis label is Cell viability rate with range 0.0 to 1.5. Bars at 0, 200, 400, 600, 800, 1000 nM are near 1.0. The image D showing a bar chart labeled, IC50 MHCC97H equals 266.7 nM. The x axis label is c-SPNPs with tick labels 0, 200, 400, 600, 800, 1000 nM. The y axis label is Cell viability rate with range 0.0 to 1.5. Bars decrease from about 1.0 at 0 nM to about 0.6 at 200 nM, about 0.3 at 400 nM, about 0.2 at 600 nM and near 0.1 at 800 and 1000 nM. The image E showing eight flow cytometry scatter plots with axes labeled 7AAD on the vertical axis and Annexin V APC on the horizontal axis, each with quadrant labels Q1, Q2, Q3, Q4 and percentages. Columns are labeled Control, Laser, c-SPNPs, 250 nM plus Laser, 500 nM plus Laser. Rows are labeled Huh7 and MHCC97H. Huh7 quadrants: Control Q1 5.34, Q2 2.10, Q3 1.79, Q4 89.8; Laser Q1 7.27, Q2 1.81, Q3 1.59, Q4 89.3; c-SPNPs Q1 7.05, Q2 1.52, Q3 1.51, Q4 89.9; 250 nM plus Laser Q1 0.12, Q2 2.34, Q3 10.9, Q4 86.6; 500 nM plus Laser Q1 4.94, Q2 12.4, Q3 58.3, Q4 24.4. MHCC97H quadrants: Control Q1 0.33, Q2 1.32, Q3 1.25, Q4 97.1; Laser Q1 0.32, Q2 1.28, Q3 1.12, Q4 97.3; c- SPNPs Q1 0.48, Q2 1.65, Q3 1.57, Q4 96.3; 250 nM plus Laser Q1 0.18, Q2 1.50, Q3 4.61, Q4 93.7; 500 nM plus Laser Q1 0.26, Q2 3.11, Q3 26.3, Q4 70.3. The image F showing a scatter plus bar chart labeled Huh7. The x axis has categories Control, Laser, c-SPNPs, 250 nM plus Laser, 500 nM plus Laser. The y axis label is Percentage of apoptotic cells with range 0 to 80. Points cluster near 0 for Control, Laser and c-SPNPs; around 10 to 15 for 250 nM plus Laser; around 55 to 65 for 500 nM plus Laser. Significance marks include asterisk and four asterisks. The image G showing a scatter plus bar chart labeled MHCC97H. The x axis has categories Control, Laser, c- SPNPs, 250 nM plus Laser, 500 nM plus Laser. The y axis label is Percentage of apoptotic cells with range 0 to 40. Points cluster near 0 for Control, Laser and c-SPNPs; around 5 to 10 for 250 nM plus Laser; around 25 to 30 for 500 nM plus Laser. Significance marks include ns, two asterisks and four asterisks.Multiple plots showing cell viability and apoptosis for Huh7 and MHCC97H under c-SPNPs and laser. Cell death induced by c-SPNPs‑based PDT in HCC cells. (A) Viability of Huh7 cells treated by c-SPNPs at 0–1000 nM without laser irradiation. (B) Viability of Huh7 cells treated by c-SPNPs at 0–1000 nM with laser irradiation. (C) Viability of MHCC97H cells treated by c-SPNPs at 0–1000 nM without laser irradiation. (D) Viability of MHCC97H cells treated by c-SPNPs at 0–1000 nM with laser irradiation. (E) Apoptosis detection of Huh7 and MHCC97H cells using flow cytometry. Statistical analysis of apoptosis ratio of Huh7 (F) and MHCC97H (G) cells. *P < 0.05, **P < 0.01, and ****P < 0.0001. P values are determined by Kruskal–Wallis test (F) and Welch’s ANOVA (G) test for multiple comparisons, n≥3. To visualize cellular ROS-generating capability, Huh7 and MHCC97H cells were incubated with c-SPNPs and subsequently stained with DCFH-DA probe. Following 660 nm laser irradiation, cells pretreated with c-SPNPs for 4 h exhibited an significant elevation in intracellular ROS accumulation compared to untreated cells or cells treated with lower doses of c-SPNPs (Figure 3A). Flow cytometric analysis further quantified this effect revealing a dose-dependent increase in ROS levels Huh7 cells (Figures 3B and C). A similar trend was observed in MHCC97H cells, where increasing concentrations of c-SPNPs led to significantly enhanced ROS production (Figure 3D and E). These findings highlight the correlation between c-SPNPs concentration and ROS generation in both Huh7 and MHCC97H cells and demonstrated excellent biocompatibility and robust ROS generation capacity of c-SPNPs under laser irradiation, leading to pronounced cytotoxicity in liver cancer cell lines. Figure 3c-SPNPs induce HCC cells death by generating reactive oxygen species (ROS). (A) Representative fluorescence micrographs (magnification: 200×) showing intracellular ROS levels detected by the DCFH‑DA probe under different treatment conditions. (B) Representative flow cytometric analysis of ROS fluorescence intensity in Huh7 cells. (C) Quantitative analysis of ROS fluorescence intensity in Huh7 cells. (D) Representative flow cytometric analysis of ROS fluorescence intensity in MHCC97H cells. (E) Quantitative analysis of ROS fluorescence intensity in MHCC97H cells. Data are presented as mean ± SD from at least three independent experiments. *P < 0.05, **P < 0.01, and ****P < 0.0001. P values are determined by Welch’s ANOVA test for multiple comparisons, n≥3.The image A shows fluorescence micrographs of Huh7 and MHCC97H cells. Each set includes ROS, Bright-Field and Merged images. The image B shows flow cytometry histograms of Huh7 cells with 0, 500 and 1000 nM c-SPNPs plus laser, measuring FITC fluorescence. The image C shows a bar graph of average fluorescence intensity in Huh7 cells, indicating significant increases with higher c-SPNPs concentrations. The image D shows flow cytometry histograms of MHCC97H cells with 0, 500 and 1000 nM c-SPNPs plus laser, measuring FITC fluorescence. The image E shows a bar graph of average fluorescence intensity in MHCC97H cells, showing significant increases with higher c-SPNPs concentrations. Statistical significance is marked with asterisks: p less than 0.05, p less than 0.01, * p less than 0.0001.Fluorescence micrographs and flow cytometry of ROS in Huh7 and MHCC97H cells with c-SPNPs. c-SPNPs induce HCC cells death by generating reactive oxygen species (ROS). (A) Representative fluorescence micrographs (magnification: 200×) showing intracellular ROS levels detected by the DCFH‑DA probe under different treatment conditions. (B) Representative flow cytometric analysis of ROS fluorescence intensity in Huh7 cells. (C) Quantitative analysis of ROS fluorescence intensity in Huh7 cells. (D) Representative flow cytometric analysis of ROS fluorescence intensity in MHCC97H cells. (E) Quantitative analysis of ROS fluorescence intensity in MHCC97H cells. Data are presented as mean ± SD from at least three independent experiments. *P < 0.05, **P < 0.01, and ****P < 0.0001. P values are determined by Welch’s ANOVA test for multiple comparisons, n≥3. The clinical application of PDT in the treatment of deep solid tumors is often hampered by limited light tissue penetration oxygen dependence. However, emerging evidence suggests a distinct advantage for HCC treatment over conventional therapies, due to its minimally non-invasive nature and spatial selectivity.4,32 Indeed, clinical research has demonstrated that PDT can not only extend survival rate of incompletely treatable cancer patients but also improve quality of life.33,34 Our findings proposed c-SPNPs as a promising candidate to address the limitations due to their proficiency in producing Type I ROS thatenables effective cytotoxicity against HCC cells in vitro even at a low irradiation and under hypoxic conditions. Combined with the established efficacy of phthalocyanine derivatives in PDT,35,36 these attributes highlight the therapeutic superiority of c-SPNPs for developing a next-generation photosensitizer for oncology. c-SPNPs Suppress Proliferation, Migration, and Invasion of HCC Cells in vitro To elucidate the impact of c-SPNP-mediated on the metastatic and invasive behaviors of HCC, we conducted a series of functional assays on Huh7 and MHCC97H cells. Scratch wound healing assays demonstrated that treatment with c-SPNPs followed by laser irradiation significantly reduced the migration ability of Huh7 (Figures 4A and C) and MHCC97H (Figures 4B and D) cells (P < 0.0001). Consistently, transwell invasion assay using matrigel to mimic the endothelial basement membrane indicated significant reduction (P < 0.05) in the number of invading Huh7 (Figures 4E and G) and MHCC97H (Figures 4F and H) cells upon irradiation. Furthermore, CCK-8 assays revealed that c-SPNP-mediated PDT significantly suppressed the proliferation of Huh7 (Figure 4I) and MHCC97H (Figure 4J) cells (P < 0.05). Collectively, these results indicate that c-SPNP-mediated PDT effectively abrogate the migration, invasion, and proliferation of HCC cells under laser irradiation, thereby reinforcing their pro-apoptotic effects. Figure 4c-SPNPs inhibit the malignant biological behavior of HCC cells. (A) Representative images of wound‑healing assays showing the migration of Huh7 cells after the indicated treatments. (B) Representative images of wound‑healing assays showing the migration of MHCC97H cells after the indicated treatments. (C) Statistical analysis of healing rate in Huh7 cells. (D) Statistical analysis of healing rate in MHCC97H cells. (E) Representative images of Transwell invasion assays for Huh7 cells after the indicated treatments. (F) Representative images of Transwell invasion assays for MHCC97H cells after the indicated treatments. (G) Quantitative analysis of invaded cell numbers in Huh7 cells. (H) Quantitative analysis of invaded cell numbers in MHCC97H cells. (I) Proliferation curves of Huh7 cells under the indicated treatments. (J) Proliferation curves of MHCC97H cells under the indicated treatments. *P < 0.05, and ****P < 0.0001. P values are determined by ANOVA test followed by Tukey methods (C, D, H, and (I) or Welch’s ANOVA test followed by Dunnett3 methods (G) or Kruskal Wallis test followed by Dunn’s methods (J) for multiple comparisons, n≥3 (C and D), n≥5 (G, H, I and J). Scale bar: 50 μm for (A), (B), (E), and (F).Images A and B show wound healing assays for Huh7 and MHCC97H at 0h and 48h/72h, respectively, with treatments: c-SPNPs and Laser combinations. Wound edges are marked. Image C′s bar chart for Huh7 shows healing rates: Ctrl, Laser, c-SPNPs and c-SPNPs plus Laser, with the latter near 0%. Image D′s bar chart for MHCC97H shows similar treatment effects, with c-SPNPs plus Laser below 0%. Images E and F display Transwell invasion assays for Huh7 and MHCC97H at 48h/96h, with treatments matching Ctrl, Laser, c-SPNPs and c-SPNPs plus Laser. Image G′s bar chart for Huh7 shows invaded cell counts, with c-SPNPs plus Laser near 0. Image H′s bar chart for MHCC97H shows similar trends, with c-SPNPs plus Laser around 20-40. Image I′s line graph for Huh7 shows OD at 450 nm over days 0-5, with c-SPNPs plus Laser lowest. Image J′s line graph for MHCC97H shows similar patterns, with c-SPNPs plus Laser ending around 6. Microscopy images are on the left, quantification graphs on the right.Multi-panel figure: c-SPNPs plus laser lowers healing, invasion and proliferation in Huh7 and MHCC97H. c-SPNPs inhibit the malignant biological behavior of HCC cells. (A) Representative images of wound‑healing assays showing the migration of Huh7 cells after the indicated treatments. (B) Representative images of wound‑healing assays showing the migration of MHCC97H cells after the indicated treatments. (C) Statistical analysis of healing rate in Huh7 cells. (D) Statistical analysis of healing rate in MHCC97H cells. (E) Representative images of Transwell invasion assays for Huh7 cells after the indicated treatments. (F) Representative images of Transwell invasion assays for MHCC97H cells after the indicated treatments. (G) Quantitative analysis of invaded cell numbers in Huh7 cells. (H) Quantitative analysis of invaded cell numbers in MHCC97H cells. (I) Proliferation curves of Huh7 cells under the indicated treatments. (J) Proliferation curves of MHCC97H cells under the indicated treatments. *P < 0.05, and ****P < 0.0001. P values are determined by ANOVA test followed by Tukey methods (C, D, H, and (I) or Welch’s ANOVA test followed by Dunnett3 methods (G) or Kruskal Wallis test followed by Dunn’s methods (J) for multiple comparisons, n≥3 (C and D), n≥5 (G, H, I and J). Scale bar: 50 μm for (A), (B), (E), and (F). It is reported that PDT-induced hypoxia can theoretically promote invasion via upregulating the expression of receptor tyrosine kinase (RTK) and vascular endothelial growth factor receptor (VEGFR).37 The significant suppression of invasion and migration in vitro observed here suggests that the cytotoxic dominance of c-SPNPs effectively neutralizes these adaptive resistance mechanisms, highlighting a distinct advantage over traditional oxygen-dependent photosensitizers. c-SPNP-Mediated PDT Induces Immunogenic Cell Death (ICD) and Promoted DCs Maturation in vitro PDT is demonstrated to induce ICD.38,39 To determine whether c-SPNP-mediated PDT induces ICD in HCC cells, we examined the release of DAMPs from Huh7 and MHCC97H cells. Following a 4 h incubation with c-SPNPs and subsequent 660 nm laser irradiation for 5 min, culture supernatants were harvested after 24 h. ATP levels were markedly elevated in the PDT-treated groups compared to all controls (P < 0.0001; Figures 5A and B). Similarly, HMGB1 release was significantly increased in both Huh7 (P < 0.05, Figure 5C) and MHCC97H (P < 0.001, Figure 5D) cells after PDT. Figure 5c-SPNP-mediated PDT induces dying cancer cells to release damage associated molecular patterns (DAMPs) and promoted DCs maturation. (A) Extracellular ATP levels in Huh7 cells following the indicated treatments. (B) Extracellular ATP levels in MHCC97H cells following the indicated treatments. (C) Extracellular HMGB1 levels in Huh7 cells following the indicated treatments. (D) Extracellular HMGB1 levels in MHCC97H cells following the indicated treatments. (E) Immunofluorescence staining of calreticulin (CRT) in Huh7 and MHCC97H cells. Images were captured using a 100× oil-immersion objective (scale bar: 20 μm) to visualize cellular distribution. (F) Schematic of bone marrow‑derived DC (BMDC) isolation and subsequent co‑culture with Hepa1‑6 cells in vitro. (G) Flow cytometric analysis of CD80+ and CD86+ BMDCs after co‑culture. (H) Quantification of CD80+ and CD86+ BMDCs. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, ns: no significance. P values are determined by Welch’s ANOVA test followed by Dunnett3 methods (A) or ANOVA test followed by Tukey methods (B) or Kruskal Wallis test followed by Dunn’s methods (C, D, and (H) for multiple comparisons, n≥3.A series of images and graphs depict the effects of c-SPNP-mediated PDT on Huh7 and MHCC97H cells. A and B show chemiluminescence intensity for Huh7-ATP and MHCC97H-ATP, respectively, with significant increases in the c-SPNPs plus laser group. C and D display HMGB1 levels in Huh7 and MHCC97H cells, showing significant increases in the c-SPNPs plus laser group. E presents immunofluorescence staining of calreticulin (CRT) in Huh7 and MHCC97H cells, with different conditions of c-SPNPs and laser treatment. F illustrates a schematic of bone marrow-derived DC isolation and co-culture with Hepa1-6 cells, showing the process from bone marrow cells to proliferating and differentiated BMDCs. G shows flow cytometric analysis of CD80 and CD86 BMDCs after co-culture, with different treatment conditions. H quantifies the percentage of matured DCs, highlighting a significant increase in the c-SPNPs plus laser group.Graphs and diagrams on c-SPNP-mediated PDT effects on Huh7 and MHCC97H cells. c-SPNP-mediated PDT induces dying cancer cells to release damage associated molecular patterns (DAMPs) and promoted DCs maturation. (A) Extracellular ATP levels in Huh7 cells following the indicated treatments. (B) Extracellular ATP levels in MHCC97H cells following the indicated treatments. (C) Extracellular HMGB1 levels in Huh7 cells following the indicated treatments. (D) Extracellular HMGB1 levels in MHCC97H cells following the indicated treatments. (E) Immunofluorescence staining of calreticulin (CRT) in Huh7 and MHCC97H cells. Images were captured using a 100× oil-immersion objective (scale bar: 20 μm) to visualize cellular distribution. (F) Schematic of bone marrow‑derived DC (BMDC) isolation and subsequent co‑culture with Hepa1‑6 cells in vitro. (G) Flow cytometric analysis of CD80+ and CD86+ BMDCs after co‑culture. (H) Quantification of CD80+ and CD86+ BMDCs. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, ns: no significance. P values are determined by Welch’s ANOVA test followed by Dunnett3 methods (A) or ANOVA test followed by Tukey methods (B) or Kruskal Wallis test followed by Dunn’s methods (C, D, and (H) for multiple comparisons, n≥3. Calreticulin (CRT) is a hallmark of endoplasmic reticulum (ER) stress during ICD.40 Immunofluorescence staining revealed a distinct translocation of CRT from the cytosol to the plasma membrane in PDT-treated cells, characterized by punctate or linear staining patterns compared with the control, laser only, and c-SPNPs groups (Figure 5E). We further investigated whether c-SPNPs and laser-treated HCC cells could induce the maturation of MBDCs by co-culturing mouse HCC cells (Hepa 1–6) with DCs in vitro (Figure 5F). Flow cytometry analysis revealed that Hepa 1–6 cells treated with c-SPNPs and laser irradiation induced a significant increase in the number of CD80+ and CD86+ MBDCs compared to other groups (P < 0.05, Figures 5G and H). The results indicate that c-SPNPs PDT triggers DAMPs in HCC cells, which subsequently drive MBDC maturation. The induction of ICD is pivotal for converting the immunosuppressive tumor microenvironment into an immunogenic one, thereby enhancing PDT efficacy.41,42 Our data confirmed that c-SPNP-mediated PDT triggers DAMPs including ATP, CRT, and HMGB1, which collectively drive the phenotypic maturation of DCs. Beyond DAMPs, PDT initiates a localized inflammatory cascade involving neutrophils, macrophages, and monocytes.43 For instance, tumor necrosis factor-α (TNF-α), a key inflammatory mediator produced during PDT, rapidly recruits neutrophils to the tumor microenvironment.44 And increased infiltration of monocytes, macrophages, and mast cells was observed shortly after PDT.45 Infiltrating macrophages release lysophosphatidylcholine, a substrate metabolized by phospholipase D into macrophage-activating factor (MAF), enhancing tumor-specific cytotoxic activity in macrophages.44 PDT further augments macrophage function by enhancing their phagocytic activity, enabling efficient clearance of apoptotic and necrotic cellular debris at the treatment site.46 This debris clearance facilitates antigen cross-presentation, thereby directly activating tumor-specific CD8 positive T cells.47 The efficacy of PDT in eliciting antitumor immune responses relies on T cell-mediated cytotoxicity.48 During this process, neutrophils, macrophages, and DCs present tumor-associated antigens (TAAs) to T cell receptors via major histocompatibility complex (MHC) class I proteins, activating CD8+ T cells and triggering tumor-specific immune destruction.49–51 Insufficient activation of endogenous or adoptive transferred immune cells can reduce the immune response.52 Among these immune cells, DCs are uniquely critical: they are considered the most potent antigen-presenting cells (APCs) due to their unparalleled ability to bridge innate and adaptive immunity.53,54 Specifically, DC maturation, marked by phenotypic activation, functional specialization, and upregulated expression of costimulatory molecules, is essential for processing TAAs and priming cytotoxic T lymphocytes (CTLs).55 In antitumor immunity, DCs capture TAAs through multiple mechanisms, including phagocytosis of apoptotic or necrotic tumor cells, uptake of soluble tumor-derived antigens, and internalization of heat shock protein (HSP)-bound antigens, which enhance immunogenic presentation.56 Upon maturation, DCs display MHC-antigen complexes alongside costimulatory ligands (eg, CD80, CD86) on their surface. These signals drive CD4+ T helper cell activation and CD8+ T cell proliferation, ultimately inducing effector T cell differentiation and sustained adaptive immune responses.57 In summary, c-SPNP-mediated PDT exerts its therapeutic effect through a dual mechanism: direct tumor cell ablation via ROS generation and the initiation of a robust, systemic antitumor immune response.58 However, given the complexity of the tumor microenvironment, the release of DAMPs mediated by c-SPNPs under hypoxic conditions and their anti-tumor immune activation effects still need further verification.59 c-SPNP-Mediated PDT Inhibits Tumor Growth and Induces ICD in vivo To investigate the antitumor efficacy of c-SPNP-mediated PDT in vivo, we employed three complementary models: (1) zebrafish larvae xenografted with fluorescently labeled Huh7 cells, (2) C57BL/6J mice vaccinated with PDT-treated rodent HCC cells and implanted with syngeneic Hepa 1–6 cells, and (3) C57BL/6J mice bearing human HCC tumors. This multi-model approach ensures robust validation of tumor growth inhibition, immune activation, and systemic safety. Zebrafish tumor cell xenotransplantation has emerged as a complementary model to mouse xenografts, owing to the high reproductivity via external fertilization, rapid embryonic development, and optical transparency during the larval stage.60,61 These attributes, particularly the transparency of larvae, enable real-time, in vivo live-imaging of tumor cell dynamics such as proliferation, migration, and vascular interactions.62 To visualize the therapeutic efficacy of c-SPNP-mediated PDT, we established a zebrafish xenograft model by microinjecting fluorescently labeled Huh7 human HCC cells into the yolk sac of zebrafish larvae (Figure 6A). Within 24 h post-implantation, Huh7 cells proliferated to form well-defined fluorescent tumor clusters localized specifically within the yolk sac, a common site for tumor cell engraftment (Figure 6B). Fluorescence imaging analysis revealed a significant decrease in tumor cluster fluorescence intensity 24 h after yolk sac injection of c-SPNPs combined with targeted laser irradiation, compared to untreated controls (P < 0.05, Figure 6C). Zebrafish larvae lack a functional adaptive immune system during the first 12–14 d of development, as T-cells and other adaptive components mature only after 4–6 weeks post-fertilization (wpf).63 This immunological gap eliminates the need for immunosuppression during human tumor cell engraftment in the larvae, allowing unimpeded tumor proliferation and metastasis.64 The zebrafish xenograft model thus offers a rapid, high-throughput platform to evaluate the direct antitumor efficacy of PDT, leveraging the absence of adaptive immunity to distinguish PDT’s effects on tumor cells from innate immune interactions. Here, the zebrafish experiment provided direct visual evidence that c-SPNPs PDT induces tumor cell death through selective photodamage in vivo. Figure 6The effect of c-SPNP-mediated PDT on the growth of HCC xenograft in vivo. (A) The schematic diagram of the Huh7 tumor‑bearing zebrafish larvae and the treatment regimen. (B) Representative mages showing Huh7 cells (labeled with Dil) in zebrafish larvae under different treatments (magnification: 30×). (C) Tumor growth curves of mice receiving different treatments (mean ± SD, n ≥ 5). (D) The schematic diagram of the Huh7 tumor‑bearing C57BL/6J mouse xenograft model and the treatment regimen. (E) Tumor growth curves of mice receiving different treatments (mean ± SD, n ≥ 5). (F) Representative photographs of excised tumors from each treatment group. (G) Tumor weights at the endpoint (mean ± SD). (H) Representative hematoxylin and eosin (H&E) staining of tumor tissues (scale bar: 200 μm). *P < 0.05, ns: no significance. P values are determined by Welch’s ANOVA test followed by Dunnett3 methods (C) or ANOVA test followed by Tukey methods (G) or Kruskal Wallis test followed by Dunn’s methods (E) for multiple comparisons, n≥3.The image A shows a timeline of zebrafish larvae from fertilization to 5 days post-fertilization (dpf), illustrating cell-derived xenograft, c-SPNPs treatment and laser application at 660 nm. The image B displays zebrafish larvae under different treatments: Control, Laser, c-SPNPs and c-SPNPs plus Laser at 4 dpf and 5 dpf, with merged and bright field (BF) images. The image C is a graph showing the increase ratio of fluorescence intensity at 4 dpf, comparing Control, Laser, c-SPNPs and c-SPNPs plus Laser treatments, with significance marked as ns and. The image D illustrates a timeline of mouse xenograft model treatments from Day -3 to Day 8, including c-SPNPs injection and laser application at 660 nm. The image E is a line graph showing tumor volume in mm superscript 3 over time for different treatments: c-SPNPs plus Laser, c-SPNPs, Laser and Control, with significance marked as. The image F shows photographs of excised tumors from each treatment group with a ruler for scale. The image G is a bar graph showing tumor weight in grams for different treatments: Control, Laser, c-SPNPs and c-SPNPs plus Laser, with significance marked as *. The image H displays representative hematoxylin and eosin (H&E) stained tumor tissues for Control, Laser, c-SPNPs and c-SPNPs plus Laser treatments.A composite image showing zebrafish and mouse xenograft models with c-SPNPs and laser treatments. The effect of c-SPNP-mediated PDT on the growth of HCC xenograft in vivo. (A) The schematic diagram of the Huh7 tumor‑bearing zebrafish larvae and the treatment regimen. (B) Representative mages showing Huh7 cells (labeled with Dil) in zebrafish larvae under different treatments (magnification: 30×). (C) Tumor growth curves of mice receiving different treatments (mean ± SD, n ≥ 5). (D) The schematic diagram of the Huh7 tumor‑bearing C57BL/6J mouse xenograft model and the treatment regimen. (E) Tumor growth curves of mice receiving different treatments (mean ± SD, n ≥ 5). (F) Representative photographs of excised tumors from each treatment group. (G) Tumor weights at the endpoint (mean ± SD). (H) Representative hematoxylin and eosin (H&E) staining of tumor tissues (scale bar: 200 μm). *P < 0.05, ns: no significance. P values are determined by Welch’s ANOVA test followed by Dunnett3 methods (C) or ANOVA test followed by Tukey methods (G) or Kruskal Wallis test followed by Dunn’s methods (E) for multiple comparisons, n≥3. To further validate the in vivo therapeutic efficacy of c-SPNP-mediated PDT, we administrated the treatment to human HCC-bearing C57BL/6J mice (Figure 6D). Tumor growth kinetics revealed significantly slower progression (P < 0.05) and reduced final tumor volume in the c-SPNP-PDT group compared to the controls (Figure 6E). Resected tumors from the c-SPNP-PDT group exhibited smaller dimensions, necrotic regions (evidenced by dark discoloration), and lower mass (P < 0.05; Figures 6F and G). H&E analysis further demonstrated extensive tumor necrosis with enhanced lymphocyte infiltration in treated mice, whereas the control groups retained intact tumor architecture (Figure 6H). However, the pharmacokinetics and biodistribution of c-SPNPs in mice still need further investigation. Mature DCs (mDCs) in tumor-draining lymph nodes serve as critical intermediaries between tumor antigen presentation and cytotoxic T-cell activation, which is a cornerstone of adaptive antitumor immunity.65,66 To verify the immune activation following c-SPNP-PDT treatment, tumor tissues from mice were analyzed via flow cytometry. The c-SPNP-PDT group exhibited significantly elevated proportions of CD80+ CD86+ DCs (P < 0.05; Figure 7A and B), markers of APC maturation, alongside increased infiltration of CD3+ CD8+ cytotoxic T lymphocyte subsets (P < 0.01; Figure 7C and 7D). CD3 serves as a pan-T cell marker expressed across all T lymphocytes, while CD8 specifically identifies cytotoxic T lymphocytes, a subset critical for targeted tumor cell elimination.67 The results suggest enhanced DCs activation and robust cytotoxic T cell recruitment, hallmarks of potent antitumor immunity.68 Finally, histopathological evaluation of major organs (heart, liver, spleen, lungs, kidneys) via H&E staining revealed no obvious toxicity in any treatment group, confirming the systemic biosafety of c-SPNP-PDT (Figure S7). We also tested the ALT and AST levels in mouse serum, and the results showed no significant differences between the groups, indicating that c-SPNPs and its mediated PDT did not cause hepatotoxicity (Figures S8 and S9). In addition, we selected tumor tissues from different mouse groups for immunohistochemical staining. The results showed that the PDT group had more CD8+ T cell infiltration. Moreover, the deeper staining of CD86 suggested possible better immune activation (Figure S10). Figure 7c‑SPNPs promote immune cell infiltration and exert cytotoxic effects in vivo. (A and B) Flow cytometric analysis (A) and quantification (B) of CD80+ and CD86+ DCs within the tumor microenvironment of C57BL/6J xenografts under indicated treatments. (C and D) Flow cytometric analysis (C) and quantification (D) of CD3+ and CD8+ T cells infiltrating C57BL/6J xenografts under indicated treatments. (E) The schematic diagram of the prophylactic tumor vaccination experiment in C57BL/6J mice. (F) Tumor growth curves in the vaccinated and control groups (mean ± SD). (G) Representative photographs of excised tumors from the vaccinated and control groups. *P < 0.05, **P < 0.01, ns: no significance. P values are determined by ANOVA test followed by Tukey methods (D) or Kruskal Wallis test followed by Dunn’s methods (C) for multiple comparisons, and Mann–Whitney test (F) n≥5.Image A displays flow cytometry dot plots with CD80 and CD86 axes for Control, Laser, c-SPNPs and c-SPNPs plus Laser. Quadrant values: Control Q4 98.0; Laser Q4 98.3; c-SPNPs Q4 93.5; c-SPNPs plus Laser Q4 72.0. Image B is a bar chart showing the percentage of matured dendritic cells, with c-SPNPs plus Laser being the highest. Image C shows dot plots with CD3 and CD8 axes. Quadrant values: Control Q4 92.6; Laser Q4 95.9; c-SPNPs Q4 95.7; c-SPNPs plus Laser Q4 75.8. Image D is a bar chart of CD3+CD8+ T cells percentage, with c-SPNPs plus Laser highest. Image E is a schematic comparing Vaccine and Control Groups, showing tumor cells after PDT or freezing/thawing. Image F is a line graph of tumor volume over days, with the Vaccine Group remaining low and the Control Group increasing. Image G shows excised tumors, with Control Group tumors larger than Vaccine Group tumors.Multi-panel scientific infographic on c-SPNPs plus laser immune markers and tumor growth outcomes. c‑SPNPs promote immune cell infiltration and exert cytotoxic effects in vivo. (A and B) Flow cytometric analysis (A) and quantification (B) of CD80+ and CD86+ DCs within the tumor microenvironment of C57BL/6J xenografts under indicated treatments. (C and D) Flow cytometric analysis (C) and quantification (D) of CD3+ and CD8+ T cells infiltrating C57BL/6J xenografts under indicated treatments. (E) The schematic diagram of the prophylactic tumor vaccination experiment in C57BL/6J mice. (F) Tumor growth curves in the vaccinated and control groups (mean ± SD). (G) Representative photographs of excised tumors from the vaccinated and control groups. *P < 0.05, **P < 0.01, ns: no significance. P values are determined by ANOVA test followed by Tukey methods (D) or Kruskal Wallis test followed by Dunn’s methods (C) for multiple comparisons, and Mann–Whitney test (F) n≥5. Tumor recurrence and metastasis remain primary drivers of post-treatment mortality, directly undermining therapeutic efficacy and long-term patient prognosis.69 Establishing durable adaptive antitumor immunity is therefore critical for suppressing metastatic tumor and enhancing survival outcomes.70 In order to verify the immune activation of c-SPNP-mediated PDT in vivo, Hepa 1–6 murine HCC cells pretreated with c-SPNP-PDT in vitro were inoculated into the left flank of mice. Immunogenicity was subsequently assessed by measuring tumor growth inhibition in HCC cells implanted contralaterally (right armpit, Figure 7E). Compared to the control group which received untreated cells, the vaccination group exhibited significantly reduced contralateral tumor volumes (P < 0.01, Figure 7F). Tumor resection after 12 d post-implantation revealed no detectable tumors in five vaccinated mice, and the only tumor detected in this group was markedly smaller than those in the control mice (Figure 7G). The remarkable regression of the untreated distant tumors in the vaccination group strongly suggests a vigorous systemic activation of adaptive antitumor immunity, which is consistent with the findings of ICD induced by c-SPNP-mediated PDT in vitro. However, how long this anti-tumor effect will last still requires more research and observation. While our previous work established the cation-substituted c-SPNPs as potent agents for oxygen-efficient antimicrobial phototherapy in preclinical models,27 their capacity to engage adaptive antitumor immunity remained unexplored. Here, we demonstrate that c-SPNP-mediated PDT not only directly kills HCC cells via Type I ROS but also induces ICD in vitro, which is evidenced by CRT surface exposure and extracellular release of ATP and HMGB1. Critically, in vivo studies across multiple tumor models reveal that PDT-generated tumor antigens and DAMPs expand CD80+ CD86+ mDCs populations, which prime CD3+ CD8+ cytotoxic T-cell responses and drive systemic antitumor immunity. This aligns with the rationale behind cancer vaccines, which leverage tumor cell death to activate immune defenses, similar with traditional vaccines that introduce pathogens to stimulate protective antibodies.71,72 As one of the most important APCs, DCs are pivotal in this process.73,74 For instance, Ce6-based PDT-generated tumor residues functioned as cancer vaccines in breast cancer models, enhancing DCs activation, inflammatory cytokine release, and antitumor immunity.75 Collectively, c-SPNP-mediated PDT provokes durable antitumor immunity by synergizing direct photodynamic tumor ablation with sustained antitumor immune response. Conclusions In this study, we developed a type I nano-photosensitizer, c-SPNPs, self-assembled from bis[(3-(N-methyl-N-morpholinyl) propoxy] phthalocyaninato silicon (IV) diiodide, specifically for treating hypoxic deep solid tumors. This nanomaterial exhibits excellent water solubility and efficiently generates ROS via a Type I mechanism, making it a highly promising candidate for PDT. We demonstrated that PDT mediated by c-SPNPs effectively induces cell death in HCC cells in vitro, significantly impairing their proliferative, migratory, and invasive capacities. Furthermore, c-SPNP-mediated PDT exhibits superior tumor growth-inhibiting activity in vivo. Our findings also reveal that PDT with c-SPNPs enhances the release of DAMPs, including ATP, HMGB1 and CRT, from dying HCC cells. This process promotes the maturation of DCs and increases the infiltration of cytotoxic CD8+ T cells into tumor tissues, suggesting a potential immunostimulatory effect. Collectively, our custom-synthesized c-SPNPs represent a promising therapeutic agent for the treatment of deep solid tumors, such as HCC.