Mn²⁺–PPa nanocomplexes were built by one nanoprecipitation method. In this study, we demonstrated that Mn²⁺ could serve as an assembly promoter to drive PPa to assemble into uniform nanostructures. To potentiate the stability of nanocomplexes, albumin was chosen as a stabilizer to contain hydrophilic and spherical nanostructures with a particle size of approximately 90 nm (Figures 1A,B; Supplementary Figure S1). Molecular dynamics (MD) simulation studies were conducted to investigate the formation mechanisms of nanocomplexes. As presented in Figures 1C,D, the simulation system tended to stabilize over 50 ns. While the simulation system was stabilized, one single large Mn²⁺–PPa assembly emerged at 100 ns. The snapshots from MD simulations showed the assembly process, transitioning from the original disordered state to the final stable assembly at the molecular level. The results suggested that manganese ions played a crucial role in facilitating the coordination formation and maintaining the stability of Mn²⁺–PPa complexes.

Photosensitizers encapsulated in the conventional NDDSs result in ACQ effect-mediated fluorescence quenching. For Mn²⁺–PPa nanocomplexes, PPa from nanocomplexes would be activated and released by the tumor-reductive environment, leading to fluorescence recovery. In this section, the fluorescence spectra of nanocomplexes were evaluated upon the addition of dithiothreitol (DTT, a reductive GSH simulant) at an excitation wavelength of 415 nm and emission wavelength of 675 nm at different times. As presented in Figures 1E,F, the excitation fluorescence intensity of nanocomplexes was evidently potentiated when incubated with DTT in a time-dependent manner. The enhanced fluorescence intensity could be attributed to the activation of nanocomplexes, following the release of PPa triggered by DTT. Similar trends were observed in the emission fluorescence spectra. Moreover, the large hydrodynamic size of nanocomplexes observed upon incubation with DTT could be attributed to the alleviated ACQ effect, followed by the collapse of nanostructures induced by H₂O₂ (Figure 1G). These results greatly demonstrated that metal ion-mediated photosensitizer (Mn²⁺–PPa) nanocomplexes could serve as an effective approach to solve the dilemma from ACQ effects. Next, we investigated the stability of the nanocomplexes directly in serum, as shown in Supplementary Figure S2. The particle size did not change significantly within 24 h, demonstrating that the nanoparticle formulation has a certain degree of stability in serum. In order to investigate the release efficiency of nanocomplexes, we placed them in different concentrations of GSH (Supplementary Figure S3). The release efficiency of nanomaterials in high concentrations of GSH was much higher than that in low concentrations of GSH, indicating that nanomaterials have good release efficiency in the tumor microenvironment.

We first evaluated the cellular uptake of Mn²⁺–PPa nanocomplexes for 1, 2, and 4 h with K7M2-WT cells. The fluorescence intensity of PPa from nanocomplexes in cancer cells was found through confocal laser scanning microscopy (CLSM). As shown in Figure 2A, the nanocomplexes exhibited a time-dependent elevation in intracellular uptake, and higher intracellular contents of nanocomplexes were found. The cytotoxicity of nanocomplexes was next assessed using the MTT assay (Figure 2B). Following laser irradiation, the nanocomplexes and PPa solution showed strong in vitro anticancer effects, while manganese ions had minimal cytotoxicity on 143B cells. These results were consistent with those of live/dead staining (Figure 2C). We then investigated the intracellular ROS generation under laser irradiation using a DCFH-DA staining approach, both qualitatively and quantitatively, following the previous instructions. As shown in Figures 2D,E, strong intracellular fluorescence intensities were simultaneously observed in the nanocomplexes and PPa solution, suggesting the activated release of PPa from nanocomplexes and alleviated ACQ effects. These findings confirmed the efficient cellular uptake of nanocomplexes and on-demand activation of PPa, which resulted in excellent in vitro anticancer activity.

It is well known that the immunostimulatory Mn²⁺ ion has DC-activating capacities. We investigated the activation effects of nanocomplexes on DCs in the following transwell plate experiments (Figure 2F). Flow cytometric analysis demonstrated that the laser irradiation-triggered disintegration of nanocomplexes promoted the expression of co-stimulatory molecules CD80 and CD86, enhancing the maturation ability of DCs (Figures 2G,H). In addition, we also explored the in vitro tumor penetration of nanocomplexes after exposure to the laser irradiation using CLSM. It was observed that nanocomplexes can penetrate throughout the multicellular tumor spheroids and deliver to the interior under the laser irradiation (Figures 2I,J). Thus, favorable immunoactivating effects and in vitro tumor penetration ability make nanocomplexes potential candidates for further in vivo research.

The tumor-specific accumulation of drugs has a significant effect on the therapeutic outcomes of osteosarcoma. 143B osteosarcoma tumor-bearing BALB/c mice were used to evaluate the biodistribution of nanocomplexes, which could be observed by monitoring the fluorescence intensity of PPa in various organs and tumors. As shown in Figures 3A,B, PPa solutions were instantly eliminated from the body after intravenous injection, whereas nanocomplexes and PPA-BSA-NPs exhibited high intratumoral accumulation fluorescent intensity than PPa solutions within 24 h. After 24 h, we extracted multiple main organs (heart, liver, spleen, lung, and kidney) and tumor tissues for qualitative observation analysis and quantitative investigation, respectively. As shown in Figures 3C,D, nanocomplexes and PPA-BSA-NPs showed significant advantages over PPa solutions in terms of biodistribution, with higher intratumoral accumulation than in the main organs.

Given great in vitro antitumor activity and in vivo biodistribution, in vivo antitumor efficacy of nanocomplexes was evaluated in BALB/c mice bearing primary 143B osteosarcoma tumors that were i.v. injected with PBS, Mn solution, PPa solution (irradiated with a 660 nm laser), and nanocomplexes (irradiated with a 660 nm laser), respectively (Figure 4A). As presented in Figure 4B, the tumor volume in the PBS group-/Mn solution group exhibited a rapid increase. The PPa solution group showed moderate tumor growth suppression, which was due to poor intratumoral accumulation characteristics (Figure 4C). The nanocomplexes showed superior tumor inhibitory capacity compared to the other groups. On Day 10, some of the mice were euthanized, and the tumor tissues were collected and weighed. The average tumor weight in the nanocomplex-treated group was the lowest, as shown in Figure 4D, agreement with tumor growth curve (Figure 4B). The survival rate of the treated mice demonstrated that the nanocomplexes had the strongest in vivo antitumor therapeutic efficiency. In addition, we also investigated the in vivo safety of various formulations during the treatment. A significant change in the body weight of mice was not found among all treated groups. After the treatment, the hematological parameters (aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels, blood urea nitrogen (BUN), and creatinine (CREA)) showed negligible indication of toxicity (Figure 4F). Furthermore, no evident histological changes were found in H&E-stained organs’ sections of the heart, liver, spleen, lung, and kidney (Figure 4G). These results revealed that nanocomplexes had a potent antitumor effect but no apparent toxic effect.

To evaluate the mechanisms underlying the strong oncolytic effects triggered by nanocomplexes, immune responses in different treatment groups were evaluated. The mice were euthanized, and tumors and blood were extracted and collected after the last treatment for flow cytometric analysis. As presented in Figures 5B–D, Mn²⁺–PPa nanocomplex treatment could facilitate the highest level of intratumoral DC maturation compared to other groups. Furthermore, the Mn²⁺–PPa nanocomplex-treated group elevated the infiltration of CD8⁺ T cells and CD4⁺ T cells in the tumors, consistent with the observation shown in Figure 5A. In addition to the increased percentages of DCs and CD8⁺/CD4⁺ T cells, solid malignant tumors feature an immunosuppressive tumor microenvironment (TME) that has been regarded as one of the main factors of therapeutic resistance. The nanocomplex treatment contributed to evidently decreased regulatory T cells (Treg cells: CD45⁺CD3⁺CD4⁺FOXP3⁺ T cells) in the tumor, demonstrating that immunosuppressive conditions in the tumor were effectively alleviated. Moreover, the percentage of CD8⁺ in blood was upregulated, as shown in Figure 5E, while the immunosuppressive Treg cells in blood exhibited a remarkable reduction, as shown in Figure 5F, suggesting that the nanocomplexes could activate robust systemic immune responses. Interestingly, the proportion of memory T cells (CD45⁺CD3⁺CD8⁺CD62L⁻CD44⁺) in the nanocomplex-treated group was higher than in other groups (Figure 5G). This supramolecular nanocomplex treatment not only triggered the strongest immune effect but also induced immune memory protection.

PPa was obtained from Ruixi Biotechnology Co. Ltd. (Xian, China). αLA, DTT, and N, N-dimethylformamide (DMF) were purchased from Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). FBS was provided by Gibco, Invitrogen Corp., USA. ROS probe DCFH-DA, DAPI, Annexin V-FITC/PI Apoptosis Kit, Live–Dead Cell Staining Kit, penicillin–streptomycin, and Dulbecco’s modified Eagle’s medium (DMEM) were supplied by Dalian Meilun Biotechnology Co., Ltd. (Dalian, China). Hoechst 33342 and MTT were provided by Beijing Solarbio Technology Co., Ltd. (Beijing, China). Anti-CD 3, anti-CD 4, anti-CD 8, anti-Foxp3, anti-CD 80, and anti-CD 86 antibodies were obtained from BioLegend.

The mouse osteosarcoma cell line K7M2 was acquired from COBIOER Biotechnology Co., Ltd. (Nanjing, China). The K7M2 cell line was incubated in DMEM with 10% FBS. The female BALB/c mice were supplied by Beijing HFK Bioscience Co. Ltd. (China). Animal experiments were performed according to the Guide for the Management and Use of Laboratory Animals and were authorized by the Institutional Animal Ethics Committee (IAEC) of Shenyang Pharmaceutical University.

Methanol was utilized as the solvent of PPa and MnCl₂ (w:w, 1:10), and the complex solution was prepared at a concentration of 10 mg/mL. BSA was prepared into a solution of 10 mg/mL using ddH₂O as the solvent. After that, 200 µL of the complex solution was added dropwise to a vial containing 1 mL of BSA-containing ddH₂O and stirred using a magnetic stirrer for 10 min. The Mn²⁺–PPa nanocomplexes were formed with uniform particle size through a one-step nano-precipitation approach.

The size and zeta potential of Mn²⁺–PPa nanocomplexes were measured using a Zetasizer instrument (Malvern, United Kingdom). Then, the appearance of Mn²⁺–PPa nanocomplexes was observed with transmission electron microscopy (TEM, JEOL, Japan).

The composite structure of PPa and Mn²⁺ ions was drawn using MarvinSketch 24.1.0. Gaussian 09 was used to optimize the structure using the B3LYP functional and def2SVP basis set, and the single-point energy was calculated using the B3LYP functional and def2TZVP basis set. In Gaussian calculations, the SCRF implicit solvent model was used to simulate the water environment, and Mn radius was set to 2.0 Å. The RESP charge was calculated in Multiwfn based on the wave function computed by Gaussian, the GAFF2 force field parameters of the complex were generated by ACPYPE, and the parameters related to Mn²⁺ were computed and completed by Sobtop. PACKMOL was used to evenly place five PPa–Mn²⁺ complexes into a rectangular box with a side length of 50 Å, and the box was filled with TIP3P water molecules. A 100 ns molecular dynamics simulation was performed using GROMACS 2023 under the amber 14 force field, and the radius of gyration was calculated.

The Mn²⁺–PPa nanocomplexes were incubated with DTT, and the fluorescence spectra were detected at an excitation wavelength of 415 nm and emission wavelength of 675 nm under different time conditions. At the same time, the hydrodynamic size of the Mn²⁺–PPa nanocomplexes was measured with or without DTT.

The K7M2 cells were inoculated in the six-well plate. When the cells reached approximately 80% confluence, the original medium was discarded, and the cells were washed twice with PBS. PBS solution containing Mn²⁺–PPa nanocomplexes (10 μg/mL, PPA equivalent) was added for 1 h, 2 h, and 4 h, after which the cells were observed via CLSM. DAPI was used to stain the K7M2 cells.

K7M2 cells were cultured in 96-well plates to reach approximately 80% confluence. The cells were then treated with 150 μL of a fresh culture medium containing Mn-BSA, PPA-BSA, and nanocomplexes and cultured for 24 h under dark conditions. The PPA-BSA and Mn²⁺–PPa nanocomplex groups were exposed to laser light (660 nm 20 mW/cm² 5 min) after 4 h of incubation. Then, the liquid in the 96-well plates was aspirated, and 150 μL of the MTT solution (1 mg/mL) was added to each well and placed in an incubator for 4 h. Finally, the MTT solution was replaced with dimethyl sulfoxide (100 μL), and the optical density value (490 nm) was recorded using a microplate reader.

In addition, the cytotoxicity of Mn²⁺–PPa nanocomplexes was evaluated using the Live–Dead Cell Staining Kit. K7M2 cells were seeded onto 12-well plates and treated as described above. After treatment, the cells were harvested, washed with PBS, stained with Annexin V-FITC/PI, and finally observed via CLSM.

The K7M2 cells were inoculated into a six-well plate. When the cells reached approximately 80% confluence, the original medium was discarded, and the cells were washed twice with PBS. The cells were incubated with DCFH-DA ROS probe for 1 h. They were then treated with PBS, Mn-BSA, PPA-BSA, and Mn²⁺–PPa nanocomplexes for 4 h, respectively. Subsequently, the liquid was discarded; the cells were washed with PBS, incubated with DIPA for 20 min, and finally observed via CLSM.

The K7M2 cells were cultured to the upper layer of the transwell plate, and the DC cells were cultured to the lower layer of the transwell plate. After overnight incubation, PBS, Mn-BSA, PPA-BSA, and Mn²⁺–PPa nanocomplexes were added for 24 h, respectively. Then, the cells were collected, labeled with antibodies, and subsequently measured and quantified by flow cytometry.

We investigated in vitro tumor penetration of nanocomplexes in three-dimensional (3D) tumor spheroids. The K7M2 cells were cultured into 3D tumor spheroids. After that, the fresh culture solution was replaced regularly. When 3D tumor spheroids reached approximately 350 µm diameter, they were incubated with PPa solutions and Mn²⁺–PPa nanocomplexes for 4 h, respectively. Fluorescence was detected with Z-stack imaging by CLSM.

K7M2 osteosarcoma tumor-bearing BALB/c mice were used to evaluate the biodistribution of nanocomplexes. K7M2 tumors were inoculated into the tibia cavity of BALB/C mice to form osteosarcoma in situ. When the tumor volume reached 350 mm³, the tumor-bearing mice were injected with PPa solutions (n = 3) or Mn²⁺–PPa nanocomplexes (n = 3) intravenously. The images were captured at 6, 12, and 24 h after injection using an IIVIS Lumina Series III. The major organs and tumor-bearing legs were separated for ex vivo luminescence imaging after sacrifice.

K7M2 osteosarcoma tumor-bearing BALB/c mice were used in this study. The mice were i.v. injected with PBS, Mn-BSA, PPa solution (irradiated with a 660 nm laser), and Mn²⁺–PPa nanocomplexes (irradiated with a 660 nm laser), respectively. Meanwhile, the changes in the tumor volume and body weight were recorded during treatment. The death of the mice was also recorded, and the survival curve was drawn.

After the experiment, the mice in each group were euthanized, and their main organs, including the heart, liver, spleen, lung, and kidney, were taken. Then, an H&E staining experiment was performed. At the same time, serum was collected, and AST, ALT, BUN, and CREA levels were analyzed.

For tumor-infiltrated lymphocyte analysis, tumor samples were separated and digested using collagenase IV and hyaluronidase for 90 min to acquire single-cell suspensions. Then, cells were washed with RBC lysis buffer twice to remove red blood cells. The cells were counted and incubated with flow cytometric antibodies on ice for 30 min before testing. After that, the tumor cells were prepared as single-cell suspensions using a filter screen. CD8⁺T cells in tumor cell suspensions were stained with FITC-CD3, Percp-CD4, and APC-CD8a. Regulatory T cells (CD3⁺CD4⁺Foxp3⁺) in suspensions were stained with FITC-CD3, PerCp-CD4, and PE-Foxp3. Mature DCs (CD80⁺CD86⁺) were stained with APC-CD86 and FITC-CD80. The population of T cells, DC cells, and FOXP3⁺ cells in the tumor and the population of T cells, FOXP3⁺ cells, and memory T cells in blood were detected by flow cytometry.

Data are shown as the mean ± SD, and statistical significance was conducted using a two-tailed Student’s t-test (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001; ns denotes not significant). All data were analyzed using GraphPad Prism 8 and Microsoft Excel software version 2019.