D-peptide engineered hydrogel with dual-enzyme-ALA cascades enables multimodal oxygen modulation for self-sustaining EDT-PDT synergy

Abstract

The transformation of O₂ is primarily facilitated by the catalytic action of redox enzymes, which play a pivotal role in sustaining cellular energy metabolism and redox balance. Bioinspired by O₂ distribution and ROS regulation related to cascade biocatalytic process, a D-peptide NapG^DF^DF^DY engineered hydrogel has been constructed with encapsulated dual-enzyme superoxide (SOD) and chloroperoxidase (CPO) cascade catalytic circuit and co-assembled photosensitizer of 5-aminolevulinic acid molecules (ALA). Multimodal oxygen modulation has been conducted by the concurrent oxygen generation via SOD-catalyzed ⋅O₂⁻ dismutation and oxygen consumption for ¹O₂ production by CPO and ALA, achieving self-sustaining enzymatic dynamic therapy (EDT)-photodynamic therapy (PDT) (EDT-PDT) synergy. The endogenous cascade-amplified EDT not only enhances the ¹O₂ efficacy in exogenous PDT therapy, but the intermediate O₂ can also alleviate local neuropathic pain caused by hypoxia for safe PDT treatment. This work pioneers enzyme-mediated dynamic control of tumor redox homeostasis, establishing a new therapeutic axis between biocatalytic amplification and photodynamic processes.

Graphical abstract

Highlights

• •D-peptide hydrogel was constructed with dual-enzyme cascade catalytic circuit and co-assembled 5-aminolevulinic acid (ALA).
• •Multimodal oxygen modulation was conducted by SOD induced O₂⁻ dismutation and O₂ generation for ¹O₂ production by CPO /ALA.
• •Self-sustaining enzymatic dynamic therapy (EDT)-photodynamic therapy (PDT) synergy was achieved.
• •The endogenous EDT enhances the ¹O₂ efficacy in exogenous PDT, the intermediate O₂also alleviates local neuropathic pain.

1. Introduction

Squamous cell carcinoma (SCC), the second most common form of skin cancer, accounts for approximately 20 % of all cutaneous malignancies. In the worldwide, the incidence of SCCs was rising year by year, posing a major public health challenge [[1], [2], [3]]. Surgical excision is the therapeutic standard of cutaneous SCCs but is not always clinically feasible due to the limitations such as intolerability, difficulty in skin graft repair, appearance and function of damaged organs [[4], [5], [6]]. Photodynamic therapy (PDT), a minimally invasive and spatially targeted treatment modality, presents novel therapeutic opportunities for cancer management [7,8]. Within the predominant Type II PDT reaction pathway, photosensitizer activation by light irradiation produces ¹O₂, which acts as the principal cytotoxic agent driving therapeutic efficacy [9]. In contrast to other types of ROS produced in alternative PDT mechanisms (e.g., Type I reactions), the ¹O₂ exhibits potent oxidative activity [10], enabling efficient damage to adjacent target molecules, yet its reactive influence is restricted to a short diffusion radius (∼10–55 nm), ensuring localized cytotoxicity [11,12]. 5-aminolevulinic acid (ALA) mediated photodynamic therapy (PDT) has been validated as an effective noninvasive therapeutic intervention in the past 40 years, which has drawn great attention for the treatment of cutaneous SCCs [13,14]. ALA is a small hydrophilic molecule widely existing in the mitochondria of animals and plants to form protoporphyrin IX (PpIX) during normal metabolic process [15]. High-dose of exogenous ALA can lead to the accumulation of PpIX, which can react to specific irradiation of light and produces the reactive oxygen species (ROS) to kill cancer cells, namely the photosensitizer in the PDT [16,17]. Topical ALA-PDT is very convenient for superficial skin cancer due to high tumor cell selection and less systemic side effects. However, ALA-PDT has limited efficiency for deeper (e.g., nodular) SCCs due to the inadequate penetration of the drugs and the light penetrating depth, as well as cellular hypoxia and pain during the treatment attributed to PpIX accumulation in peripheral sensory nerves and inflammatory responses. Recent advances in light delivery systems (e.g., fractionated irradiation, two-photon excitation) [[18], [19], [20]] and oxygen-economizing strategies (e.g., nano-oxygen carriers, catalase-like nanomaterials) have partially addressed these limitations [[21], [22], [23], [24]]. Nevertheless, a systematic approach to synergistically enhance the exogenous ALA-PDT efficacy while mitigating intrinsic side effects remains an unmet need.

In living organisms, the transformation of O₂ is primarily facilitated by the catalytic action of redox enzymes, which play a pivotal role in sustaining cellular energy metabolism and redox balance [25,26]. Particularly in the generation and elimination of ROS, enzymes such as superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GPx) operate within a sophisticated multi-enzyme cascade network [27,28]. This intricate regulatory framework not only supports energy metabolism and biosynthesis but also rigorously controls ROS levels to prevent oxidative stress [29,30]. The maintenance of this equilibrium is vital for essential life processes. Inspired by the respiratory burst process in the marvelous neutrophils of innate immune systems [[31], [32], [33], [34]], the ROS-regulated cancer therapeutic strategy by enzyme dynamic therapy (EDT) has been proposed in authors’ group, which can convert the endogenous ROS (⋅O₂⁻ and H₂O₂) into highly biotoxic ¹O₂ by the cascade biocatalytic reaction of peroxidase system in the tumor region [[35], [36], [37], [38]]. Endogenous enzymatic catalysis leverages tumor-specific substrate (e.g., overexpressed ⋅OH, NO⋅, H₂O₂, ⋅O₂⁻) to autonomously upregulate therapeutic species, especially the ¹O₂ within the tumor microenvironment (TME).

The photosensitizer of ALA could be converted into the highly photosensitive protoporphyrin IX (PpIX) under the action of biocatalysts, which may convert adjacent O₂ into ¹O₂ under irradiation. Therefore, the outcome of ¹O₂, the primary type of ROS in PDT, is severely limited by hypoxia within the tumor. In SOD-CPO implemented cascade catalysis, the endogenous ⋅O₂⁻ may be converted into O₂ and H₂O₂, thereby improving the production of ¹O₂ in PDT. Meanwhile, besides the above-mentioned SOD-induced oxygen modulation, the depth-independent cascade circuit also enables the generation of ¹O₂ via the catalyzation of CPO, which may further enhance the overall therapeutic effect, especially in the deeper tumor regions. Herein, a dual-enzyme catalytic circuit of SOD-CPO-ALA co-assembled supramolecular hydrogel has been prepared, to simultaneously resolve hypoxia while converting tumor-associated ROS into cytotoxic ¹O₂, achieving self-sustaining EDT-PDT synergy. D-type peptide of NapG^DF^DF^DY (Pep) was prepared as the gelators of supramolecular hydrogel. Relative to the L-type peptide, D-type peptide shows better aminopeptidase-inhibiting activity which ensures better stability and sustained drug release capability. In dual-enzyme catalytic circuit, the endogenous ⋅O₂⁻ may be converted into O₂ and H₂O₂. In the presence of Cl⁻ and H₂O₂, CPO can catalyze the formation of hypochlorous acid (HClO), which can further react with H₂O₂ to produce ¹O₂. By fully considering the ROS regulation of the SOD and CPO enzymes and their cascading catalytic metabolic regulation, the O₂ and ¹O₂ production can be modulated for endogenous ROS-activated EDT and Near Infrared (NIR) triggered PDT. Therefore, as the pioneering drug delivery system, we envision that the dual enzyme and ALA co-loaded polypeptide hydrogel can potentially regulate the ROS in the microenvironment and achieve efficient bio-catalyzation enhanced PDT for cutaneous SCC therapy.

2. Results and discussion

The characterizations below provide important proof of concept data for the development of a local drug delivery system as a new class of in situ forming long acting injectables. Wang and colleagues performed studies reporting that NapGFFpY (p: H₂PO₄⁻) could produce gels at a low minimum gelation concentration of 0.08 wt% wherein glycine acting as a linker to separate the bulky groups of Nap and FF for stepwise self-assembly into hydrogel. Additionally, D-peptides offer the desirable building blocks for developing peptide formulations with improved biostability. Therefore, we utilized the peptide of NapG^DF^DF^DY as the gelators enable the hydrogel to hold the photosensitive molecule for long-acting drug delivery applications (Fig. 1).

Fig. 1.

Schematic illustration of injectable SOD-CPO-ALA co-assembled hydrogel as the local drug delivery system for EDT enhanced PDT of cancer.

The following results outline the preparation of the enzyme-photosensitizer co-loading hydrogel demonstrating superior mechanical stiffness and biostability for local drug delivery applications.

2.1. Characterizations of peptide synthesis, drug conjugation and hydrogel formulation

D-type peptide of NapG^DF^DF^DY was synthesized via the solid phase synthesis method and the structure characterization was further accomplished through the analysis of the ¹H NMR spectrum (Fig. S1). The supramolecular hydrogel in this work was fabricated based upon the delicate balance between hydrophilicity (water solubility) and hydrophobicity (water insolubility). The prepared peptide is a water-soluble molecular whose hydrophobicity would be enhanced with the pH decrease, and thereafter self-assembled into fibers for fabricating the hydrogel network (Concentration: 0.6 wt%). Enzymes of SOD and CPO (100U/mL) and ALA (10 mg/mL) were pre-dissolved and physically mixed into peptide solution during preparation. The morphology of the as-prepared hydrogel was characterized. As shown in Fig. 2a, in the Pep solution (0.01 wt%, pH = 7.4), the nanofibers of the self-assembled peptide were clearly observed. The manufactured hydrogel was transparent and self-supporting in the vial and its cryoelectron microscopy (cryo-EM) structures were depicted in Fig. 2b, wherein typical porous structure could be observed. As the carrier of ALA, the mechanical attributes of peptides hydrogels both with (Pep-ALA^gel) and without ALA (Pep^gel) were evaluated gauged by Young's modulus. Both the Pep^gel and Pep-ALA^gel exhibited higher storage modulus (G′) relative to their corresponding loss modulus (G″) signifying the solid-like properties inherent in the hydrogels (Fig. 2c). Notably, Pep-ALA^gel had a higher G′ value than that of Pep^gel, demonstrating that ALA enhanced the mechanical performance of the hydrogel. Moreover, the increased strain sweep test suggested that Pep-ALA^gel was more stable than Pep^gel, the conjugation of ALA to peptide sequences resulted in an increase in gel strength measured by the breakage strain/flow point (Fig. 2d). Meanwhile, under the shear mode, the viscosity of the Pep-ALA^gel gradually decreased, indicating that the hydrogel has desirable shear thinning property, which is desirable for injectional materials (Fig. 2e). The shear stress caused size change of the self assembles was also verified by dynamic light scattering (DLS), which indicated that the average length of the fibers was above 1000 nm. However, when the hydrogel was tested with extrusion and shear force, the length decreased to ∼110 nm, indicating that the nanofibers fragmented into smaller sized fragments (Fig. 2f). Normally, the length of fibers is in micron scale, with the shearing stress action, the nanofibers can break into shorter fragments. The morphology change was also verified by SEM, which displayed the original fibers and shorter fragments, namely the clusters clearly (Fig. S2). That is, a morphology change could occur in the inherent hydrogel after injection. Such physical property enables more enzymes and ALA being exposed to cancer cells, which is of great importance for the drug release and cellular uptake.

Fig. 2.

Characterizations of polypeptide assembled hydrogel. (a) TEM image and the photographic image of the Pep self-assembled hydrogel. (b) The freeze cracking SEM image of the hydrogel. (c) (d) The G′ and G″ of the hydrogel under time scan mode and shear mode of rheological measurement, respectively. (e) The viscosity of the hydrogel under the shear mode of rheological measurement. (f) Size distribution of the Pep self-assemblies before and after shear-thinning. (g) Conformational changes in simulation box of Pep with ALA after 200 ns simulation. (h) The dynamics of RMSD within 200 ns simulation. (i) The number of hydrogen bonds between Pep and ALA during 200 ns simulation. (j) The bonding energy between peptides, solution and ALA.

In view of enhanced mechanical property of Pep-ALA^gel relative to Pep^gel, we employed GROMACS-based molecular dynamics (MD) simulations to explore the binding mode between peptides and ALA at a 1:17 ratio which was in accordance with the formulation in subsequent animal experiments. Briefly, Pep, ALA, water molecules (Solution) and ions at a certain proportion were randomly placed into a simulation box, and a 200ns MD simulation was performed. As illustrated in Fig. 2g, visual structural analysis of the simulation system at 0 ns and 200 ns directly displayed this process. Fig. 2h shows the time-dependent RMSD of peptides in the simulation box. In the initial stage, the molecular conformation of the peptides in restricted aggregation state underwent substantial fluctuations due to the interactions with water molecules and ALA in the solvent environment, so significant RMSD oscillations could be observed. As the self-assembly proceeded, another oscillation signal was detected, which was caused by the boundary interference of the peptide-assembly. The fluctuation of RMSD decreased after 110 ns simulation with an average MSD value of 2.94 ± 0.01 nm, indicating that the conformation of peptides reached stability. Meanwhile, the numbers of hydrogen bonds (N_HB) between peptides, water molecules, and ALA were calculated. The N_HB of peptides-ALA was 21.53 ± 4.79, indicating that there was hydrogen bond interaction between hydrogel scaffolds and ALA (Fig. 2i). Amounts of hydrogen bonds existing between Pep and water molecules, and the corresponding N_HB was 266.32 ± 12.61 (Fig. S3a). The N_HB of Pep-Pep was determined as 10.03 ± 3.48 (Fig. S3b). Meanwhile, the binding energy among Pep, water molecules and ALA molecules was calculated using the MMPBSA method, and the results are presented in Fig. 2j, in which the binding energies of Pep-Pe, Pep-Solution, Pep-ALA were determined to be −1413.88 ± 29.59, - 591.04 ± 44.68, - 133.51 ± 25.00 kJ mol⁻¹. In summary, we therefore reasonably inferred that the ALA could interact with the peptides and therefore enhance the mechanical stiffness of the hydrogel.

Enzyme uniformity is of great importance to the catalytic outcome of enzyme-laden material. Therefore, enzyme uniformity was analyzed. we labeled SOD and CPO with CY3 and CY5 respectively via EDC/NHS reaction. The fluorescein-labeled enzyme was mixed into the hydrogel by vortexing. Firstly, we tried to confirm the distribution of enzymes in hydrogel by laser scanning confocal microscopy (LSCM). The enzyme-laden hydrogel was transferred on the confocal microscopy-compatible cell culture dish to obtain the images. As shown in Fig. S4a, the green light (CY3-labeled SOD) and red light (CY5-labeled CPO) were uniformly distributed in the view, proving from the side that the enzyme did not undergo agglomeration or aggregation in the hydrogel. Meanwhile, equal volumes of hydrogel (20 μL) were randomly sampled from various positions of the hydrogel, and their fluorescence spectra were analyzed. As shown in Fig. S4b–c, the fluorescence intensity of CY3-labeled SOD approximated very much. Similarly, the fluorescence intensity of CY5-labeled CPO was numerically close, indicating the uniform distribution of SOD and CPO in the hydrogel.

2.2. Cascade enzymatic reaction enhanced PDT

The schematic illustration of cascade enzymatic reaction enhanced PDT was shown in Fig. 3a. Compared to normal cells, cancer cells overproduce more ROS including ⋅O₂⁻, H₂O₂, OH⋅ etc. due to the oxidative stress. SOD is a kind of iron enzyme that could catalyze the reaction of ⋅O₂⁻ and create O₂ and H₂O₂. CPO is a member of peroxidase family that could produce ¹O₂ using H₂O₂ as the substrate. In view of this, we utilize SOD as the ROS regulator, and the CPO and ALA as the ¹O₂ generator. Specifically, ⋅O₂⁻ can convert into O₂ and H₂O₂ by SOD catalyzation, which will assist in the downstream CPO catalyzation and PDT. That is, the endogenous as well as the as-produced H₂O₂ can be decomposed into H₂O and highly reactive ¹O₂ by CPO, the O₂ also convert into ¹O₂ with the specific irradiation in the presence of ALA, thereby killing the tumor cells.

Fig. 3.

Multimodal oxygen modulation via SOD-CPO cascade catalytic circuit (a) Schematic cascade enzymatic reaction enhanced PDT via ROS regulation of SOD-modulated O₂ and CPO/PDT-modulated ¹O₂ production. (b) The fluorescence changes of O₂ probe (Ru(dpp)₃)Cl₂ over time. (c) The fluorescence changes of H₂O₂ probe ADHP over time. (d) The fluorescence changes of ¹O₂ probe SOSG over time. (e) The intensity of ¹O₂ at 15 min of the X/XO, CPO@Gel and SOD-CPO@Gel reaction system. (f) Activity retention of loaded SOD in 2 weeks. (g) Activity retention of loaded CPO in 2 weeks. (h) The fluorescence spectra of H₂O₂ generated by SOD@Gel over time. (i) The fluorescence spectra of ¹O₂ generated by SOD-CPO@Gel over time. (j) The concentrations of H₂O₂ and ¹O₂ produced by the SOD-CPO cascade circuit.

We utilized xanthine (X) and xanthine oxidase (XO) to generate ⋅O₂⁻ in vitro for simulating the tumor microenvironment (TME). As proof of concept of enzyme-mediated oxygen modulation, the enzyme catalyzed generation of O₂, H₂O₂ and ¹O₂ were primarily confirmed. First, SOD@Gel was mixed into the X/XO solution, the O₂ and H₂O₂ yield by SOD was detected separately. As shown in Fig. 3b, O₂ was detected by fluorescence quenching using [Ru(dpp)₃]Cl₂ as the probe. The fluorescence intensity at 613 nm gradually decreased, indicating that SOD@Gel could catalyze the formation of O₂. In the same way, H₂O₂ was detected using 10-Acetyl-3,7-dihydroxyphenoxazine (ADHP) as the probe. As shown in Fig. 3c, the fluorescence intensity at 585 nm gradually increased in 30 min, indicating that H₂O₂ could be generated by SOD@Gel in the presence of ⋅O₂⁻. Meanwhile, the ¹O₂ generated during the catalyzation of SOD-CPO@Gel was also verified. As shown in Fig. 3d, using SOSG as the fluorescence probe, the fluorescence intensity at 525 nm was significantly enhanced, suggesting the formation of ¹O₂ during the cascade catalytic reactions of SOD and CPO. Meanwhile, the ¹O₂ produced by CPO@Gel, SOD-CPO@Gel were semi quantified by the electron spin resonance (ESR) method with 2,2,6,6-tetramethylpiperidine (TEMP) as a trapping agent. As shown in Fig. S5, no typical signals of ¹O₂ could be observed in the pioneering substrate X/XO solution. With CPO@Gel was added into X/XO solution, wherein H₂O₂ (100 μM) and NaCl (100 mM) was externally added as the substrate of CPO, typical signals (spin height ratio: 1:1:1, triplet) of ¹O₂ (a^N = 1.698 mT, g = 2.002) were observed with gradually increasing as a function of time. Then, SOD-CPO@Gel was incubated in X/XO solution with NaCl (100 mM), typical signals of ¹O₂ were also observed and the signal intensity was much higher than that produced by CPO@Gel. That is, compared to single CPO catalysis, the cascade enzyme catalysis between SOD and CPO may efficiently promote the formation of ¹O₂. The signal intensities of ¹O₂ at 15 min were calculated and summarized in Fig. 3e. With the incorporation of SOD, the capability of CPO@Gel for yielding ¹O₂ could be greatly improved.

As one kind of injectable enzyme-related preparation, the enzymatic activity is unneglectable. Thus, the catalytic activity retention of SOD and CPO were measured continually in 2 weeks using UV–Vis spectrophotometer. The SOD activity was studied using pyrogallol autoxidation method (Fig. S6). The CPO activity was measured using 5,5-Dimethylcyclohexane-1,3-dione (MCD) chlorination method (Fig. S7). As presented in Fig. 3f and g, compared with free enzymes, the catalytic activity of SOD and CPO in hydrogel were determined to 83.36 % and 84.22 %, respectively. In addition, the activity of these two enzymes still maintained at 69.16 % and 74.44 % after 2 weeks, the calculated results were summarized in Table S1.

Considering the influence of cellular proteases on enzyme activity, the catalytic activities of enzyme-laden hydrogel in the presence of Protease K were studied. Free enzyme and enzyme-laden hydrogel were separately incubated with physically mixed Protease K (1 mg/mL), then the activities were analyzed after 24 h by UV–Vis spectra. As shown in Fig. S8a–d, in the presence of Protease K, both the free enzymes of SOD and CPO showed no catalytic activities. In contrast, the SOD and CPO in hydrogels still retained 86.62 % and 80.25 % of its catalytic activity, respectively. The influence of pH on enzyme activity was also assessed (Fig. S8e–h). Free enzymes and enzyme-laden hydrogels were exposed in DI water (pH = 5) separately, the catalytic activities were measured after 24 h. The SOD@Gel and SOD retained 87.23 % and 84.31 % of its catalytic activity. For CPO, the enzyme in gel exhibited a higher activity retention of 73.66 % than the free enzyme. Therefore, the hydrogel offers a protective microenvironment for the enzymes.

Meanwhile, we tried to define the quantitative stoichiometric relationship between SOD-generated H₂O₂ and CPO-mediated ¹O₂ via fluorescence spectroscopy. First, the quantitative standard curves of H₂O₂ and ¹O₂ were established (Fig. S9). Then, the SOD-generated H₂O₂ and SOD-CPO-generated ¹O₂ were separately detected. The reaction system consisting of xanthine (10 μM)/xanthine oxidase (10U/mL)/Cl⁻ (50 mM)/SOD@Gel (10U/mL)/ROS Green ^TMH₂O₂ Probe (2.5 μM) was established to monitor the H₂O₂ generation by fluorescence spectroscopies along with the time (Fig. 3h). Meanwhile, with the same amounts of xanthine, xanthine oxidase, Cl⁻ and SOD of above test, another reaction system consisting of xanthine (10 μM)/xanthine oxidase (10U/mL)/Cl⁻ (50 mM)/SOD-CPO@Gel (10U/mL)/SOSG (2.5 μM) was established to monitor the ¹O₂ generation, in which the SOD-generated H₂O₂ was converted into ¹O₂ by CPO (Fig. 3i). Based on the catalytic mechanism of CPO, using SOD-generated H₂O₂ as the substrate, the theoretical ratio of CPO-produced ¹O₂ and SOD-generated H₂O₂ is 2:3. The concentrations of H₂O₂ and ¹O₂ were calculated according to the standard curves, and the calculated results were shown in Fig. 3j. The dashed line stands for the theoretical yield of ¹O₂ which is deduced by the concentration of SOD-generated H₂O₂. Based on this result, the conversion rate of H₂O₂-¹O₂ is on an overall gradual decline, which was about 60 % at 1 h.

Considering the influence of cellular proteases on the structural stability of polypeptide hydrogels, the sustained-release behavior of the hydrogel encapsulated enzyme and ALA were detected in the presence of Protease K, wherein CPO was set as the model enzyme, Pep^Gel with Protease K was set as blank. First, standard curves of Concentration-Absorbance were established based on the UV–Vis absorbance curves of CPO and ALA (Fig. S10a–d). As shown in Fig. S10e–f, the release rate of ALA reached 37.56 % in the first 2 h, then it increased to 57.23 % at 5 h. Finally, the release rate of ALA was 87.34 % at 24 h, indicating the hydrogel exhibits a certain sustained-release effect on ALA. As summarized in Fig. S10g–h, the release rate of the CPO, which reached 17.42 % in the first 24 h, was significantly slower than that of ALA. The release rate increased to 49.56 % after 7 days, indicating good structural stability and potent interactions between gel network and enzyme. Overall, hydrogels can achieve sustained drug release. The release rate of small molecule drugs is obviously faster than that of the macromolecular enzyme, which may attribute to the different interactions with gel matrix and the steric hindrance.

2.3. Intracellular characterization for EDT-PDT synergy

To assess the cytotoxicity of the EDT-PDT, cutaneous squamous carcinoma cells of XL 50 cells, abbreviated as SCC, and normal fibroblasts of NIH-3T3 cells were utilized as the model cell for CCK-8 assay. First, both SCC and NIH-3T3 were co-incubated with pure hydrogel at different concentrations. As shown in the Supplementary Fig. S11a–b, with the increased concentration, the hydrogel showed no cytotoxicity to both cells at 24 h and 48 h, even promoted the proliferation of NIH-3T3 cells, whose cell viability reached 136 % in the incubation solution with 0.12 wt% hydrogel. Then, the cytotoxicity of SOD-CPO@Gel (EDT) was then examined. SOD and CPO were incubated with both cells and the viability at 12 and 24 h were calculated (Fig. S11c–d). Similarly, EDT also showed almost no detectable cytotoxicity to NIH-3T3 cells, the cell viability was still more than 80 % at 24 h. The EDT exhibited strong cytotoxicity to SCC, wherein the cell viability decreased to 6.92 % at the enzyme concentration of 4U/mL. In the same way, the cytotoxicity of PDT was also systematically analyzed. The cells were incubated in the ALA contained culture medium and then irradiated under the 630 nm red light. As shown in Fig. S11e–f, the viability of NIH-3T3 cells and SCC gradually decreased with the increase of light dosage. At the energy density of 0.25 and 0.5 J/cm², the cell viabilities were ∼95 % and ∼80 % of NIH-3T3 cells and SCC, respectively. PDT may have certain killing effect on both normal and cancer cells at a high dose, therefore a rational light dose matching biosafety and anti-tumor cytotoxicity is of great important for the treatment. Taking this into consideration, enzymes of 1 and 2 U/mL, energy density of 0.25 and 0.5 J/cm² were adopted for the following EDT-PDT tests. As shown in Fig. 4a and b, at the energy density of 0.25 J/cm², the viability of NIH-3T3 cells, which reached 75 % at 24 h, still remained at 71 % at 72 h. When the energy density increased to 0. 5 J/cm², the viability of NIH-3T3 cells slightly declined to 66 % at 72 h. Under identical experimental conditions, the viability of SCC cells exhibited a dramatic decline, dropping to approximately 6 % at 24 h, indicating that EDT-PDT effectively suppresses tumor cell proliferation with minimal cytotoxicity toward normal cells (Fig. 4c and d). The EDT-PDT treatment strategy demonstrated excellent biosafety. Notably, even under extremely low light intensity, EDT markedly potentiated the therapeutic efficacy of PDT in vitro. In view of this, the quantitative synergy analysis was performed using Jin's formula method, where the calculation formula for Q value is:

Q = E_(a+b)/[(E_a + E_b)- E_a* E_b].

where E_a represents the cell death rate of EDT, E_b represents the cell death rate of PDT, E_(a+b) represents the cell death rate of EDT-PDT.

Fig. 4.

Intracellular characterization for EDT enhanced PDT. Cytotoxicity of EDT-PDT to NIH-3T3 cells at different light density of (a) 0.25 J/cm² and (b) 0.5 J/cm² in 72 h. Cytotoxicity of EDT-PDT to XL50 cells at different light density of (c) 0.25 J/cm² and (d) 0.5 J/cm² in 24 h (n = 3). (e) The flow cytometry test of SCC cells treated with EDT, PDT and EDT-PDT. (f) The statistical analysis of flow cytometry test. (n = 3). (g) Western blot analysis of p53 and cleaved Caspase-3 of EDT, PDT and EDT-PDT groups in SCC cells. (h) The relative intensities of cleaved Caspase-3 and p53. (i) The ratio of JC-1 aggregates (Red) and JC-1 monomers (Green) by the flow cytometric assays of JC-1 staining of EDT-PDT treated XL50 cells. (j) The ATP level in XL50 cells of different groups. (k) Intratumoral ROS imaging of O₂, H₂O₂, and ¹O₂. The corresponding fluorescence images of normal tumor tissues were set as control. Scale bar = 200 μm. All the data are presented as mean ± s.d. The data were statistically analyzed by ANOVA method in Fig. 4f, and by two-tailed t-test in Fig. 4c, d, h, i, and j. The significance level was set at*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. All statistical analyses were conducted using GraphPad Prism software.

The detailed survival rates and corresponding Q value were summarized in Table 1. The results demonstrated that EDT and PDT demonstrate notable synergistic effects in all the combined treatment groups. The synergistic index increased significantly when the dose was reduced from 2 U/mL; 0.5 J/cm² to 1 U/mL; 0.25 J/cm².

Table 1.

Q value of EDT-PDT at different dosage.

Dose (Enzyme:U/mL; Light:J/cm2 Time:h)	Survival Rate (%)
	EDT	PDT	EDT-PDT	Q Value
1; 0.25; 12	92.27	95.65	29.19	6.03
1; 0.5; 12	92.27	83.275	25.69	3.20
2; 0.25; 12	65.03	95.65	7.21	2.45
2; 0.5; 12	65.03	83.27	3.21	2.11
1; 0.25; 24	91.15	95.05	27.78	5.40
1; 0.5; 24	91.15	80.95	27.22	2.78
2; 0.25; 24	52.40	95.05	7.90	1.59
2; 0.5; 24	52.40	80.95	5.73	1.88

The Q value: Q < 0.85: Antagonistic effect; 0.85 < Q < 1.15: Addition effect; Q > 1.15: Synergistic effect.

In addition, the cells were stained with annexin V-FITC/PI. The apoptosis of XL50 cells of EDT, PDT and EDT-PDT treatment was accurately analyzed by flow cytometry. As shown in Fig. 4e and f, compared with the EDT group and the PDT group, the EDT-PDT treatment group demonstrated markedly increased apoptosis that was statistically significant. Therefore, the SOD-CPO@Gel can be a potential LDDS for safe and effective EDT enhanced PDT treatment. Apoptotic cells activate the Caspase-dependent pathway to participate in the death process. As shown in Fig. 4g and h, the results of Western blot revealed that after EDT-PDT treatment, the participation of caspase-3 activation in XL50 cells was statistically different from that of the EDT and PDT groups, indicating that mitochondria play a key role in the bioenergetics of most eukaryotic cells, and they regulate metabolism and mediate acute cell death. Flow cytometric assays of JC-1 staining were used to measure the mitochondrial membrane potential of XL50 cells after EDT-PDT treatment. Functional mitochondria containing J-aggregates (JC-1 aggregates) were stained fluorescent red (FL2 channel), whereas damaged mitochondria containing J-monomers (JC-1 monomers) were stained green (FL1 channel). The ratio of FL2/FL1 was summarized in Fig. 4i. The ratio of the EDT-PDT group of remarkably higher the that of the other experimental groups. Subsequently, the intracellular ATP level of XL50 cells were assessed by ATP assay. The ATP concentration of each sample was calculated which were illustrated in Fig. 4j. The ATP level in PDT-EDT treated XL50 cells was significantly lower than that of the other experimental groups.

Additionally, we confirmed the oxygen modulation of EDT and EDT-PDT in vitro via frozen section staining method. [Ru(dpp)₃]Cl₂ and ROS Green ^TMH₂O₂ Probe were utilized as the fluorescence probe, which emit red and green fluorescence in the presence of O₂ and H₂O₂, respectively. As shown in Fig. 4k, compared with normal tumor tissues, the tumor sections exhibited obvious red and green fluorescence, indicating the upregulation of O₂ and H₂O₂ upregulated by SOD. Meanwhile, EDT and EDT-PDT generated ¹O₂ were also detected using SOSG as the fluorescence probe. No green fluorescence signals were detected in tumor sections of control group and SOD@Gel treated group, indicating no ¹O₂ were yielded. However, when the mice were treated by SOD-CPO@Gel (EDT) or SOD-CPO-ALA@Gel (EDT-PDT synergy), distinct green fluorescence signals could be detected in tumor sections, indicating that EDT and EDT-PDT significantly upregulate the ¹O₂ in tumor. As a TME-responsive regulator, SOD can efficiently catalyze the conversion of endogenous ∙O₂⁻ into H₂O₂ and O₂. This reaction not only helps to regulate the redox homeostasis within the TME but also significantly promotes the generation of ¹O₂ by CPO and PDT, thereby effectively enhancing the therapeutic efficacy. By comparison, the ¹O₂ produced by EDT-PDT exhibited a broader spatial distribution, which may attribute to the diffusion of ALA in tumor tissues. Leveraging the in-situ gelation and prolonged retention properties of the gel preparation, the SOD-CPO-ALA co-assembled hydrogel can enhance local drug concentration while extending its dwell time, thereby establishing a sustained in situ “drug depot” that promotes subsequent deep tissue penetration.

2.4. In vivo cancer therapy and biosafety evaluation

First, Nude mice with XL50 cell-derived squamous cell carcinoma model were established to further evaluate the in vivo efficacy of EDT enhanced PDT therapy. SCC mice with similar tumor sizes (80–100 mm³) were divided into 5 groups of 6 mice each, which were the PBS control group, Gel group, EDT group, PDT group and EDT-PDT group. The photo images of the mice at day 0 and day 14 are listed in Fig. 5a. The tumor size was recorded with vernier calipers every 2 days, the average tumor growth of EDT-PDT group was obviously lower than that of the control groups with a statistical difference (Fig. 5b). Notably, there was no dramatical decrease in body weight of the mice during the entire treatment procedure, verifying that dual enzyme-ALA coloaded hydrogel is virtually harmless to the mice's growth (Fig. 5c).

Fig. 5.

Therapeutic effect of EDT enhanced PDT on the XL50 cell-derived squamous cell carcinoma tumor mice model. (a) Photographs of DAY 0 and DAY 14 of tumor-bearing mice after various treatments. (b) The change curves of the relative tumor volume and (c) body weight of the mice during the treatments. (d) Immunohistochemical staining of PAR2, TRPV1, and Substance P in skin tissues and (e) the Statistical analyses. (e) Quantification of hind limb spasms within 1 min post-irradiation (f) TUNEL and HE images of tumor slices, Scale bar = 200 μm. All the data are presented as mean ± s.d. T-test was employed for inter-group comparisons. The significance level was set at*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. All statistical analyses were conducted using GraphPad Prism software.

To determine whether EDT-PDT reduces pain while maintaining therapeutic efficacy, we assessed nociceptive responses at both molecular and behavioral levels. Immunohistochemical staining of skin tissues revealed significantly lower expression of the pain-related markers PAR2, TRPV1, and substance P in the EDT-PDT group compared to the ALA-PDT group (Fig. 5d and e), indicating reduced local nociceptive activation. In line with these findings, behavioral assessments under anesthesia showed that mice treated with ALA-PDT exhibited intense pain-related behaviors, including frequent hind limb spasms. In contrast, mice in the EDT-PDT group showed significantly fewer spasms within the first minute after irradiation, suggesting a lower level of PDT-induced pain (Fig. 5f). These results highlight the potential of EDT-PDT to minimize pain during photodynamic therapy.

To further confirm the therapeutic efficacy of EDT enhanced PDT, both terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay and the hematoxylin and eosin (HE) staining were conducted. As shown in Fig. 5g, all the treatment groups of EDT, PDT, EDT-PDT led to cellular destruction and clear apoptosis in which the EDT-PDT group caused the most significant damage to the tumor tissues, indicating the excellent therapy outcomes of EDT-PDT. Additionally, the major organs of heart, liver, spleen, lung, and kidney were excised and sliced for HE staining (Fig. S12), verifying the desired biosafety and high biocompatibility of SOD-CPO@Gel. Above results comprehensively confirmed the high efficiency and high biocompatibility of the EDT enhance PDT strategy based on the dual-enzyme and ALA coloaded injectable hydrogel.

Meanwhile, we also evaluated the treatment efficiency of EDT-PDT on the mice model of A431 cells for cancer modeling with a longer treatment duration. The treatment process is the same as that of XL50 cells-derived mice model. The photo images of the mice at day 0 and day 28 are listed in Fig. 6a. The average tumor growth curve of the EDT-PDT group was also obviously lower than that of the control group with a statistical difference (Fig. 6b). Similarly, no dramatical decrease in body weight of the mice was detected during the entire treatment procedure (Fig. 6c). TUNEL and HE stain assay were conducted in the same way.

Fig. 6.

Therapeutic effect of EDT enhanced PDT on the A431 cell-derived squamous cell carcinoma tumor mice model. (a) Photographs of DAY 0 and DAY 28 of tumor-bearing mice after various treatments. (b) The change curves of the relative tumor volume and (c) body weight of the mice during the treatments. (d) TUNEL and HE images of tumor slices, scale bar: 100 μm. (e) AST and (f) ALT levels in mouse serum after treatment. The data was presented as mean ± standard deviation. The t-test was employed for inter-group comparisons. The significance level was set at p < 0.05. All statistical analyses were conducted using GraphPad Prism software.

As shown in Fig. 6d, all the treatment groups of EDT, PDT, EDT-PDT led to cellular destruction and clear apoptosis in which the EDT-PDT group caused the most significant damage to the tumor tissues, indicating the excellent therapy outcomes and generalizability of EDT-PDT. Alanine aminotransferase (ALT) and Aspartate aminotransferase (AST) levels in serum are commonly used as biomarkers for liver and heart damage. The amounts of ALT and AST in mouse serum were detected at the end of treatment. As shown in Fig. 6e and f, all the indicators were within the normal range and there were no significant differences in the ALT/AST levels among the treatment groups. This further supported that EDT-PDT synergy exhibits good biosafety.

3. Conclusion

Inspired by the cascade biocatalytic processes of reactive oxygen species regulation and oxygen distribution in the oxidative-antioxidative system, we innovatively designed an injectable dual-enzyme (SOD-CPO) and photosensitizer (ALA)-coloaded hydrogel as a pioneering localized drug delivery system for enhanced photodynamic therapy (PDT). On one hand, SOD-catalyzed O₂ may enhance the generation of ¹O₂ in PDT. On the other hand, SOD-produced H₂O₂ could serve as a substrate to promote CPO-mediated catalyzation, further contributing to ¹O₂ production. Based on these mechanisms, an EDT enhanced PDT strategy was proposed to achieve ROS-regulated SCC treatment. In vitro cytotoxicity assays demonstrated that the SOD-CPO-ALA co-loaded hydrogel exhibited excellent biosafety while demonstrating potent cytotoxicity against tumor cells under irradiation. The cascade enzymatic reactions significantly enhanced the PDT efficacy against cancer cells. In vivo antitumor studies further confirmed that EDT-PDT markedly suppressed tumor growth with negligible systemic side effects. Compared to conventional PDT, EDT-PDT also reduced pain levels and decreased pain-related mediators. Histopathological analyses revealed that EDT-PDT induced substantial apoptosis in tumor tissues without causing evident damage to major organs. Collectively, both in vitro and in vivo results validate the potential of our SOD-CPO-ALA co-loaded hydrogel as an effective injectable localized drug delivery system, significantly enhancing the photodynamic therapeutic efficacy. The synergistic strategy of EDT-PDT represents a promising avenue for the development of ROS-modulated combination cancer therapies, with significant potential for precise drug delivery and multimodal combined therapy, particularly in targeting deep-seated or advanced lesions.

4. Experimental section

4.1. Materials and methods

Superoxide Dismutase (SOD, ≥6000 U/mg, EC 1.15.1.1) and Chloroperoxidase (CPO, ≥18000 U/mL, EC 1.11.1.10) were purchased from Sigma-Aldrich. Amino acids were purchased from GL Biochem. Tris·HCl buffer was purchased from Beijing Solarbio Science & Technology. Monochlorodimedon, 10-Acetyl-3,7-dihydroxyphenoxazine and [(Ru(dpp)₃)]Cl₂, xanthine and xanthine oxidase were purchased from Shanghai Macklin. Singlet Oxygen Sensor Green (SOSG) was purchased from Thermofisher. Monochlorodimedon was purchased from Shanghai Aladdin. ALA hydrochloride powder was from Shanghai Fudan-Zhangjiang Bio-Pharmaceutical Co, Ltd. (Shanghai, China). Murine cSCC cell line XL50, established from UV-induced cSCC in SKH-1 hairless mice, is stored at the China Center for Type Culture Collection (CCTCC No. C201827, Wuhan, China).The cell line A431 was purchased from American Type Culture Collection(Manassas, VA, USA).

4.2. Preparation of the peptide-based hydrogels

4.2.1. Preparation of Pep^Gel and Pep-ALA^gel

Briefly, the 6 mg peptide was dissolved in the DI water under alkaline conditions wherein 100 μL NaOH solution (1M) was added. Then, the pH value was adjusted to 7 slowly using HCl solution (0.5 M) to obtain supramolecular hydrogel. The volume of the preparation was adjusted to 1 mL to obtain the final hydrogel at the concentration of 0.6 wt%.

4.2.2. Preparation of Pep-ALA^gel

For Pep-ALA^gel, 100 μL ALA(100 μg/mL) was mixed with the peptide solution before HCl. Then, the pH value was adjusted to 7 gradually using HCl solution (0.1 M) to obtain supramolecular hydrogel. Next, 10 μL solutions of SOD and CPO (1000 U/mL) were physically mixed into the hydrogel, separately. Finally, the volume of the preparation was adjusted to 1 mL.

4.2.3. Preparation of SOD-CPO@ gel and SOD-CPO-ALA co-assembled hydrogel

Briefly, hydrogelation of the peptide occurred along with the addition of quantitative HCl. Then, 10 μL solutions of SOD and CPO (1000 U/mL) were physically mixed into the hydrogel, separately. Finally, the volume of the preparation was adjusted to 1 mL to obtain the final SOD-CPO@ Gel.

4.2.4. Preparation of SOD-CPO-ALA co-assembled hydrogel

To be specific, the peptides (6 mg) were dissolved in dissolved alkaline water adjusted by NaOH (100 μL, 1 M). Then, 100 μL ALA solution (100 mg/mL) was added in small amounts multiple times with thorough vortex. After that, the pH was lowered to 7 with the addition of HCl (0.1 M). In this process, the hydrophobicity of the polypeptide molecule is enhanced, thereby assembling into the hydrogel. After that, SOD and CPO (1000 U/mL, 10 μL) were added into the hydrogel via thorough vortex. Finally, DI water was added to fix the volume at 1 mL.

4.3. Morphology characterizations of peptides assemblies and Pep^gel

The morphology of peptides assemblies was characterized by TEM. The prepared solution of peptides (0.01 wt%, pH = 7.4) was dropped onto the copper mesh and dried for further observation. For SEM characterization, the Pep-ALA^gel was extruded on the silicon slice and free-dried for the subsequent observation.

4.4. Rheological measurements

The kinetics of Pep-gel and Pep-ALA gel were evaluated using a RS6000 rheometer with parallel plate geometry (25 mm diameter, 0.3 mm gap). The strain amplitude sweeps of gels were taken at 37 °C in the dynamic oscillatory mode with a constant deformation of 1 % and frequency of 1 Hz. The measurements of frequency-dependent sweep were measured as a function of angular frequency at strain of 0.03 %. The viscosity of Pep-ALA gel was measured under shear mode with a shear stress range of 0.1–1000 Pa.

4.5. Preparation of CY3-labeled SOD and CY5-labeled CPO

Briefly, 10 mg of enzyme was incubated with 50 μg of EDC in 10 mM PBS buffer (pH = 8.5) for 24 h. N-hydroxysuccinimide ester dyes of CY3-NHS ester or CY5-NHS ester were added in the dark for another 5 h. Finally, the unreacted free dye was removed by ultrafiltration using a centrifugal filter tube.

4.6. Activity test of SOD and CPO

The activity of SOD within the hydrogel was determined based on the inhibition ability of SOD on pyrogallol autoxidation. The inhibition rate of pyrogallol autoxidation can be calculated according to the equation: Inhibition (%) = [(A-B)/A] × 100, where A is the autoxidation rates of pyrogallol in the absence of SOD, B is the autoxidation rates of pyrogallol in the presence of free SOD or SOD-CPO co-loaded hydrogel. The autoxidation rate of pyrogallol can be calculated by the slope of absorbance curve during the process of autoxidation at the first minute. The activity of the immobilized SOD in hydrogel is defined as the ratio between the inhibition of sample and free SOD: Activity (%) = I_gel/I_free, Where I_gel and I_free represent the inhibition rates of pyrogallol autoxidation by SOD-CPO@Gel and free SOD, respectively.

The activity of CPO was evaluated based on the ability of CPO to catalyze the conversion of monochlorodimedon (MCD) to dichlorodimedon (DCD) at pH = 2.75 in the presence of potassium chloride (KCl) and H₂O₂. The activity assay was performed in 0.1M phosphate buffer (pH = 2.75), containing 20 mM KCl, 2 mM H₂O₂, 0.1 mM MCD and standard free CPO (2 U) or SOD-CPO@Gel (including 2 U CPO) in supernatant. The reaction progress was monitored by recording the absorbance changes at 278 nm. The specific activity of CPO@Gel was expressed as the reaction rate comparison of CPO@Gel relative to free CPO: Activity (%) = R_Gel/R_free, where R_Gel and R_free represent the reaction rate of SOD-CPO@Gel and free CPO, respectively.

4.7. CCK8 test

The cells were inoculated in 96-well plates at the density of 5 × 10³ cells/well for 24 h, and 3 repetitive wells were prepared for each group. For the PBS and EDT groups, the cells were first cultured after the addition of PBS or SOD-CPO@Gel (5–20 μL, 20U/mL). Then, refresh the PBS or SOD-CPO@Gel at the time point of 4 h to maintain parallel experimental conditions with PDT. After further incubation for 8 and 20 h, 10 μl Cell Counting Kit-8 was added. 1 h later, the absorption values at 450 nm were measured by a microplate reader to calculate the cell survival rate (%).

For the PDT groups, the cells were precultured with ALA (10 μL, 1 mg/mL) in serum-free medium for 4 h in the dark, then irradiated by a LED light (630 nm, 10 mW/cm²). After medium renewal, the cells were further cultured for another 8 and 20 h. Then, 10 μl Cell Counting Kit-8 was added for CCK8 test. For EDT-PDT groups, the cells were precultured with ALA (10 μL, 1 mg/mL) and SOD-CPO@Gel (5–20 μL, 20 U/mL) in serum-free medium for 4 h in the dark, then irradiated by LED light (630 nm, 10 mW/cm²). After medium renewal, the cells were further cultured with SOD-CPO@Gel (5–20 μL, 20 U/mL) for another 8 and 20 h. Then, 10 μl Cell Counting Kit-8 was added for the CCK8 test.

4.8. Apoptosis analysis

For cell apoptosis analysis, the XL50 cells were firstly co-cultured with PBS, Gel, SOD-CPO@Gel, ALA and SOD-CPO-ALA@Gel, separately. 4 h later, the cells of the PDT group were irradiated by LED light (630 nm, Philips, Netherlands) at a power density of 10 mW/cm². And then, the cells were preincubated with 5 μL annexin V (5 μL annexin V-FITC was dissolved in 50 μL buffer, Bio Basic Inc., Markham, ON, Canada) in dark at room temperature for 15 min, followed by adding 10 μL PI (10 μL PI was dissolved in 250 μL buffer, Sigma). After staining, the percentage of apoptotic cells was analyzed by flow cytometry (BD, Franklin Lakes, NJ). The Q2 region represents late apoptotic cells, and Q4 region represents early apoptotic cells.

4.9. Animal experiments

BALB/c nude mice were purchased from shanghai leigen company. For the SCC tumor model, the right hind legs of BALB/c mice were subcutaneously implanted with XL50 cells/A431 cells suspended in saline. Nude mice with squamous cell carcinoma models were established to further evaluate the in vivo efficacy of EDT enhanced PDT therapy. 80–100 mm³ tumor-bearing mice with similar tumor size were divided into 5 groups of 6 mice each, which were the PBS control group, Gel group, EDT group, PDT group and EDT-PDT group. PBS, Gel, and EDT group were injected with 50ul. PDT group was to apply a newly prepared 8 % ALA cream to the tumor-bearing mouse back skin lesions. The thickness of ALA cream was about 1 mm. The mice were placed in a dark room and protected from light for 3 h. The skin lesions of the mice were irradiated with Omega LED laser with a wavelength of 632.8 nm, a power density of 80 mw/cm², an energy density of 30 J/cm², and an irradiation time of 375 s. Nude mice were euthanized by breaking their cervical vertebra after 2 weeks. Tumors were separated and weighed. The growth inhibition ratio of tumor was calculated.

4.10. Histopathology analysis

The tumour tissues, livers, lungs, hearts, spleens and kidneys were excised and fixed in 10 % neutral formaldehyde, conventionally paraffin embedded, sectioned, and placed on slides. For analysis, 4 μm sections from each sample were stained with hematoxylin and eosin (H&E, Sigma-Aldrich) for the histopathological evaluation using a standard procedure. The 4 μm tumour sections from all groups were further immunohistochemically stained for TUNEL (terminal transferase UTP nick-end labelling) to analyse the cell death. Sections were evaluated from six randomly selected fields by two separate pathologists in a blinded manner.

4.11. Western blot

RIPA buffer (Sigma, USA), which contains a protease inhibitor cocktail and a phosphatase inhibitor cocktail, was used to extract the total protein (Epizyme Biomedical Technology Company, China). The manufacturer's instructions used BCA Protein Assay Kit (Thermo Electron Corporation, USA) to normalize the lysates. BCA (Thermo Scientific, Rockford, IL, USA) was used to measure protein concentration. The total protein extract was separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to 0.45 μm polyvinylidene fluoride membrane (Amersham Biosciences, Amersham, UK). The membrane was then blocked with TBST containing 5 % skim milk, and incubated with mouse anti p53, caspase 3 and β-actin in TBST with gentle agitation overnight at 4 °C. The membrane was washed with TBST and hybridized with HRP-conjugated secondary antibody according to the primary antibody. After washing with TBST, protein bands specific for the antibody were visualized by enhanced chemiluminescence (ECL) (Thermo Scientific, Rockford, IL, USA) and images were captured using ChemiDoc XRS (BioRad, Hercules, CA, USA). The intensities of the bands were quantified using Gel-Pro Analyzer (Version 4.0) (Media Cybernetics, Silver Spring, MD, USA).

4.12. JC-1 staining

Cells that had been treated with Adjudin at varying concentrations for overnight were washed with PBS and collected by trypsinization. The suspended cells were washed with PBS, and added with 10 ml of 200 mM JC-1 (2 mM final concentration Enzo, Farmingdale, NY USA) and incubated at 37 °C, 5 % CO₂, for 30 min. The suspend cells were pelleted by centrifugation at 2000g for 5 min, then resuspended by adding 500 ml PBS. Finally, the cells were analyzed on a BD FACSAria II flow cytometer (San Jose, CA, USA) using 488 nm excitation with 530 nm and 585 nm bandpass emission filters.

4.13. ATP assay

ATP was quantified using the Roche ATP Bioluminescence Assay Kit following the standard protocol provided by the vendor. In brief, cells were washed once with PBS and lysed with the Cell Lysis Reagent for 20min. Then the homogenates (50 ml each) were mixed with the Luciferase Reagents (150 ml per sample), and the luminescence was detected using a plate reader (Synergy2, BioTek, Winooski, VT, USA). The protein concentrations of the samples were determined using the BCA assay. The ATP concentration of the sample was calculated using an ATP standard and normalized against the total protein quantity of the same sample.

4.14. Frozen section staining

A431 mice models were established for the characterization of intracellular ROS, mice without any treatment were set as the control. After pre-mixing the sample and fluorescence probe, the tumor tissue was collected 2 h post-injection, frozen-sectioned into 4 μm-thick glass slides. Each tissue section was observed via the fluorescence microscopy (Leica DMI6000). Two independent pathologists evaluated the sections blindly across six randomly selected fields of view.

4.15. Immunohistochemical staining of PAR2, TRPV1, and substance P

Freshly excised skin tissues were collected from mice post-PDT and fixed in formalin. The tissues were embedded in paraffin, sectioned into 5 μm slices, and deparaffinized (30 min at 56 °C, followed by two 10-min washes in xylene). The sections were rehydrated, subjected to antigen retrieval, and blocked with a blocking solution. Subsequently, the sections were incubated at 37 °C for 30 min with primary antibodies against PAR2, TRPV1, and Substance P (1 μg/mL in blocking solution). After washing with PBS, the slides were incubated with a goat anti-rabbit IgG secondary antibody (Boster, China) diluted in blocking solution for 30 min. Streptavidin-biotin complex (Boster, China) was applied to the slides for 30 min, followed by PBS washing. Staining was performed using DAB chromogen, and sections were counterstained with hematoxylin. The stained slides were observed under a light microscope. Negative control sections were treated with PBS instead of the primary antibody.

4.16. Pain responses

Pain responses were recorded during the first minute after treatment initiation, and the frequency of hind limb spasms was counted to evaluate pain behavior.

CRediT authorship contribution statement

Qi Zhang: Writing – original draft, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation. Zhe Zheng: Writing – original draft, Software, Methodology, Investigation. Yiting Zhao: Methodology, Investigation, Formal analysis, Data curation. Qing Wu: Visualization, Validation, Supervision, Project administration, Funding acquisition. Chu Wu: Software, Investigation, Data curation. Xiuli Wang: Writing – review & editing, Supervision, Resources, Project administration. Xia Wang: Writing – review & editing, Validation, Resources, Project administration, Funding acquisition, Conceptualization. Qigang Wang: Writing – review & editing, Validation, Resources, Project administration, Conceptualization. Peiru Wang: Writing – review & editing, Visualization, Validation, Supervision, Resources, Project administration.

Ethics approval and consent to participate

This study was performed in strict accordance with the Guidelines for the Care and Use of Laboratory Animals of the National Institutes of Health and ethically approved by the Tongji University Science and Technology Ethics Committee (No. TJBH15523201).

Declaration of competing interest

There are no conflicts to declare.

Appendix A. Supplementary data

The following is the Supplementary data to this article.