Intelligent catalase-coated MnO₂ nanoparticles with programmed oxygen supply and glutathione depletion for enhanced photodynamic therapy

Abstract

Manganese dioxide (MnO₂) nanoparticles have been reported to deliver drugs, supply oxygen and consume glutathione (GSH) to promote cancer photodynamic therapy (PDT). However, most of them suffer from low drug loading capacity and conflicting oxygen/GSH tuning, which restricts their therapeutic potential. In this study, a high capacity MnO₂-derived multifunctional nanocarrier was designed to alleviate tumor hypoxia, one of the most critical conditions for effective PDT, by systematically modulating local oxygen supply and GSH depletion. The prepared MnO₂ (MH) nanoaggregates were coated with catalase (CAT) through molecular assembly and chemical crosslinking, yielding the MH@CAT nanocomposite. In the presence of hydrogen peroxide (H₂O₂), the CAT coating facilitates oxygen generation, while the MnO₂ core remains intact until encountering intracellular GSH, resulting in MnO₂ decomposition and GSH draining. This programmed regulation of oxygen supply and GSH consumption is a key design to optimize the tumor microenvironment for enhanced PDT. After loading chlorin e6 (Ce6), the as-prepared MH@CAT-Ce6 demonstrates improved cellular uptake, oxygen self-supply, and GSH depletion – all of which contribute to the superior PDT effects observed against breast cancer cells both in vitro and in vivo. Notably, the MH@CAT-Ce6 nanoparticles exhibit excellent tumor accumulation and retention, leading to potent anti-tumor efficacy with minimal systemic toxicity.

---

Introduction

Photodynamic therapy (PDT), an emerging cancer treatment approach, has garnered increasing attention in recent years for its application in tumor ablation.^1–3 The core mechanism underlying its tumor-killing effect lies in the generation of reactive oxygen species, which is facilitated by the interplay of photosensitizers (PSs), oxygen, and light at a specific wavelength.^4–6 However, the clinical efficacy of PDT remains constrained by multifaceted challenges, including (1) suboptimal accumulation of small-molecule PSs at the tumor site^7,8 and (2) the hypoxic tumor microenvironment coupled with high glutathione (GSH) levels in solid tumors, which collectively limit reactive oxygen species availability.^6,9–11 Mechanistically, PDT relies on PSs to transfer absorbed light energy to adjacent oxygen to generate highly cytotoxic singlet oxygen (¹O₂), rendering tumor hypoxia a critical barrier to PDT efficacy – a limitation further exacerbated by oxygen consumption during photodynamic reactions.¹² Additionally, excessive GSH in the tumor microenvironment exhibits antioxidant properties, neutralizing PDT-generated ¹O₂ and conferring therapeutic resistance.¹³

Manganese dioxide (MnO₂) nanocarrier platforms have recently emerged as promising delivery solutions, significantly enhancing drug accumulation in tumor tissues or cells.^14,15 For example, albumin-protected manganese dioxide nanoparticles (MnO₂–HSA) possess excellent biocompatibility, biodegradability, and multifunctionality. Many studies have confirmed that MnO₂–HSA can effectively encapsulate PSs as a carrier and interact with tumor microenvironmental factors (e.g., H⁺, H₂O₂, and GSH) to generate oxygen or deplete GSH, thereby enhancing the sensitivity of PDT.^16–19 However, loading PSs onto these MnO₂–HSA nanoparticles necessitates complex chemical modification.²⁰ Additionally, their relatively small particle size and the loose dispersion of HSA may limit the encapsulation efficiency for small-molecule drugs, rendering the drug-loading strategy incompatible with all therapeutic agents.^16,20–24 Therefore, further assembly of MnO₂–HSA into denser nanostructures is required to enhance its drug-loading capacity.

At the same time, MnO₂ nanocarriers integrating oxygen delivery/generation with intracellular GSH depletion capabilities are proposed to simultaneously address both hypoxic and antioxidant challenges.^13,25–28 However, the implementation of these functions in the tumor microenvironment is constrained by two key factors: the abundant H₂O₂ in the acidic extracellular matrix of solid tumors and the high concentration of GSH within tumor cells.^29–32 The premature consumption of MnO₂ nanoparticles by extracellular H₂O₂ results in ineffective intracellular GSH depletion and compromised PDT. Therefore, strategies to program oxygen generation and GSH depletion in the tumor microenvironment are desired.

To simultaneously achieve these two objectives, ingenious design is integrated into the MnO₂-derived nanoparticles. Firstly, to prevent the premature degradation of the MnO₂ component by abundant extracellular H₂O₂, a potential strategy is to minimize the direct contact probability between MnO₂ and H₂O₂. Using catalase (CAT) as a protective shield is a rational approach to accomplish this goal.^33–35 However, simply mixing catalase and MnO₂ is insufficient, as the resulting complex often lacks the desired structural order, which could compromise the protective effect in the H₂O₂-rich tumor microenvironment. Inspired by the layer-by-layer assembly method,^36–39 a more feasible approach is to encapsulate the MnO₂ particles in an ordered structure. The key design consideration is that the outermost layer should allow the passage of GSH while effectively blocking the penetration of H₂O₂. This selective permeability would enable the MnO₂ core to selectively engage with the intracellular GSH, without being prematurely consumed by extracellular H₂O₂.

In this study, MnO₂–HSA nanoparticles, firstly, were synthesized and further aggregated into a denser MH nanoparticulate system via reverse solvent precipitation. CAT was subsequently deposited and crosslinked onto the surface of MH using the same reverse solvent precipitation method, yielding a CAT-shielded MnO₂-based nanocarrier (MH@CAT), which exhibited enhanced physical adsorption capacity for various small molecules, including dyes and PSs. Importantly, the proposed mechanism of MH@CAT nanoparticles lies in their ability to generate oxygen in the presence of H₂O₂ without substantial consumption of the MnO₂ core, thus preserving most MnO₂ to consume intracellular GSH. Upon entering the tumor microenvironment, CAT in the outer layer of MH@CAT preferentially reacts with H₂O₂ to produce oxygen, relieving tumor microenvironmental hypoxia and supplying oxygen for PDT. Concurrently, the protected MnO₂ within the nanoparticles depletes intracellular GSH, which synergistically enhances the efficacy of PDT. Both in vitro and in vivo studies have revealed that the MH@CAT-Ce6 nanomedicine could effectively produce oxygen within tumor cells and tissues while reducing intracellular GSH levels. Following near-infrared light irradiation, this nanodrug exhibited significantly enhanced tumor cell killing efficacy compared to free molecular Ce6 and the MH–Ce6 nanodrug formulation lacking the CAT component (Fig. 1). Overall, we have developed a catalase-coated, MnO₂-based nanoparticle platform that increases drug loading via physical adsorption, while optimizing the regulation of tumor hypoxia and GSH levels to achieve potent photodynamic therapy against solid tumors.

Fig. 1 Schematic illustration of the preparation of MH@CAT–Ce6 as well as its ability to sequentially generate oxygen and reduce GSH for the enhancement of PDT.

Experimental

Synthesis of various MnO₂ nanoparticles

MH nanoparticles. HSA (20 mg) was dissolved in ddH₂O (1.8 mL). After rotation for 40 min, MnCl₂·4H₂O (10 mg mL⁻¹, 200 µL) and NaOH (1 M, 40 µL) were sequentially added. After rotating for overnight, a MnO₂–HSA complex was obtained. The MnO₂–HSA complex (2 mL) was mixed with acetone (7 mL) for 15 min and then glutaraldehyde (50%, 40 µL) was added and rotated for 1 h. The reaction solution was centrifuged (15 000 rpm, 20 min, twice) to collect MH nanoparticles and then resuspended in phosphate buffer (PB) (pH 8.5, 10 mM, 2 mL).

MH@CAT nanoparticles. MH (500 µL), PB (pH 8.5, 10 mM, 1 mL), CAT (20 mg mL⁻¹, 600 µL), acetone (1 mL) and glutaraldehyde (50%, 3 µL) were mixed and rotated for 1 h. ddH₂O (3 mL) was added and the MH@CAT nanoparticles were purified by centrifugation (15 000 rpm, 20 min, twice) with wash.

MH@CAT–Ce6 nanoparticles. Ce6 (600 µg mL⁻¹) was pre-dispersed in PB (pH 8.5, 10 mM). Thereafter, Ce6 (600 µg mL⁻¹, 1 mL) and MH or MH@CAT nanoparticles (500 µg mL⁻¹, 3 mL) were mixed overnight. The Ce6 loaded particles were purified by centrifugation (15 000 rpm, 20 min, twice) with wash to remove the free Ce6. The collected MH–Ce6 and MH@CAT–Ce6 nanoparticles were resuspended in PB (pH 8.5, 10 mM).

Investigation for O₂ generation of nanoparticles

To investigate the ability of MH@CAT–Ce6 to catalyze H₂O₂ to O₂, different concentrations of MH@CAT–Ce6 (0, 50, 100, 200, and 400 µg mL⁻¹) were incubated with 1 mM H₂O₂ solution. The amount of O₂ generated was recorded at 10-second intervals over a 10-minute period.

Investigation for nanoparticles to reduce GSH

To evaluate the H₂O₂ resistance of MH and MH@CAT to resist H₂O₂, as well as their ability to protect MnO₂ for intracellular GSH depletion, a programmed reaction process was designed. First, 20 µL nanoparticles (10 mg mL⁻¹) was added in 175 µL PB buffer (pH 7.4) with 2 µL H₂O₂ (5 mM). After co-incubation for 30 min, 2 µL GSH solutions with different concentrations of (1 mM and 500 mM) were added into the reaction system and incubated for another 30 min. The absorbance of MnO₂ was measured at 370 nm.

Cellular uptake

Human breast cancer MDA–MB-231 cells and murine breast cancer 4T1 cells were cultured in DMEM containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin at 37 °C under 5% CO₂.

MDA–MB-231 cells or 4T1 cells (1 × 10⁴ cells) in DMEM were seeded into confocal dishes and incubated for 24 h. The medium was replaced with fresh DMEM containing free Ce6 or MH@CAT–Ce6 (1 µM Ce6). The medium was discarded at the given time points (0, 2, 4, 8 and 12 h), and then the cellular uptake of free Ce6 and MH@CAT–Ce6 was examined by confocal laser scanning microscopy (FV3000, Olympus).

In vitro O₂ generation

The intracellular O₂ generation was monitored by confocal fluorescence imaging with an O₂ sensing probe, the tris(4,7-diphenyl-1,10-phenanthroline)ruthenium(II) dichloride complex ([Ru(dpp)₃]Cl₂, RDPP).^40,41 First, MDA–MB-231 cells or 4T1 cells were inoculated at a density of 1 × 10⁴ per well. After 24 h, RDPP (4 × 10⁻⁶ M) with or without H₂O₂ (50 µM) was added and incubated for 4 h and washed using PBS. Then, the cells were co-incubated with free Ce6 or MH@CAT (1 µM Ce6, 100 µg mL⁻¹ MH@CAT) for another 4 h, respectively.

Detection of intracellular GSH consumption

To detect the content of GSH, MDA–MB-231 cells or 4T1 cells were incubated with H₂O₂ (50 µM) for 4 h. Then, the medium was discarded and replaced with fresh DMEM containing MH or MH@CAT (100 µg mL⁻¹). After incubating for another 4 h, the level of GSH was measured using the GSH detection kit.

Cytotoxicity tests

To study the cytotoxicity of MH@CAT and MH@CAT–Ce6, MDA–MB-231 cells or 4T1 cells were inoculated in 96-well plates at a density of 1 × 10⁴ per well and incubated for 24 h. Afterward, the medium was discarded and replaced with 100 µL fresh medium containing different concentrations of MH@CAT and MH@CAT–Ce6 to each well. After 24 h, the standard MTT method was applied to detect the cell viabilities.

To detect the PDT efficiency of MH–Ce6 and MH@CAT–Ce6, MDA–MB-231 or 4T1 cells were inoculated in 96-well plates at a density of 1 × 10⁴ per well and incubated for 24 h. Subsequently, different concentrations of MH–Ce6 or MH@CAT–Ce6 (1 µM Ce6, 100 µg mL⁻¹ MH or 100 µg mL⁻¹ MH@CAT) were added. After 8 h of incubation, the medium was discarded and replaced with 100 µL fresh medium with or without 50 µM H₂O₂ for 15 min. Then, the cells were irradiated with a 650 nm laser (20 mW cm⁻², 15 min). The standard MTT method was applied to detect the cell viabilities after further incubation for 16 h.

Hemolysis assay

Briefly, MH@CAT nanoparticles at various concentrations (10–500 µg mL⁻¹) were mixed with erythrocyte suspension (2% v/v). Deionized water was used as a positive control. The mixture was incubated at 37 °C for 2 h. After centrifugation at 3000 rpm for 10 min, the supernatant of each sample was collected. The absorbance at 540 nm was measured, and the hemolysis rate was calculated using the following formula: hemolysis rate = (A_s − A_n)/(A_p − A_n) × 100%, where A_s, A_n and A_p represent the absorbance value of the sample, negative control and positive control, respectively.

Detection of intracellular ¹O₂ generation

1,3-Diphenylisobenzofuran (DPBF) was used for monitoring the generation of ¹O₂ by nanoparticles in the presence of H₂O₂ under laser irradiation at 650 nm.⁴² First, MDA–MB-231 cells or 4T1 cells were inoculated in 96-well plates at a density of 1 × 10⁴ per well and incubated for 24 h. Then, DPBF (100 µg mL⁻¹) was added to 200 µL DMEM containing different concentrations of MH@CAT–Ce6 (1 µM Ce6, 100 µg mL⁻¹ MH@CAT) with H₂O₂ (50 µM). Subsequently, the mixture was irradiated under 650 nm laser (20 mW cm⁻²) for 20 min under a N₂ atmosphere. The absorption of DPBF at 410 nm reported the generation of ¹O₂.

In vivo PDT effect

All the animal experiments were approved by the ethics committee of Xi’an Jiaotong University and conducted with the “guide for the care and use of laboratory animals” of the institute of laboratory animal resources. BALB/c mice bearing subcutaneous 4T1 tumors (∼100 mm³) were divided into six groups (n = 4 per group): (1) untreated, (2) free Ce6 + 650-nm laser irradiation (+), (3) MH–Ce6 with irradiation (+), (4) MH@CAT without irradiation, (5) MH@CAT–Ce6 without irradiation, and (6) MH@CAT–Ce6 with irradiation (+). For the laser groups, the tumor sites of mice were irradiated with 650 nm laser (10 min, 100 mW cm⁻²) at 2, 12, and 24 h after nanoparticle injection (doses: 5 mg kg⁻¹ Ce6, 5 mg kg⁻¹ MnO₂, peri-tumoral injection). The tumor sizes and weights of mice were recorded every 2 days. The tumor volume was calculated using the following formula: length × width²/2. At the end of the treatment period (18 days), the tumor tissues were collected for weighing, and the main organs (liver, heart, lungs, spleen, and kidneys) were also collected and stained with H&E staining.

In vivo fluorescence imaging

For in vivo fluorescence imaging, the BALB/c mice bearing subcutaneous 4T1 tumors (∼100 mm³) were randomly assigned to the following three groups: (1) free Ce6, (2) MH–Ce6 and (3) MH@CAT–Ce6. All nanoparticles were injected into each 4T1 tumor bearing mouse by peri-tumoral injection (doses: 5 mg kg⁻¹ Ce6 and 5 mg kg⁻¹ MnO₂). The fluorescence images were obtained at the given time points (0, 2, 4, 8, 12 and 24 h). The mice were sacrificed 24 h after injection, and their major organs (liver, heart, lungs, spleen, and kidneys) and tumors were collected for ex vivo FL imaging. All the animal experiments were approved by the ethics committee of Xi’an Jiaotong University and conducted with the “guide for the care and use of laboratory animals” of the institute of laboratory animal resources.

Test the ability to alleviate hypoxia of MH@CAT–Ce6

HIF-1α is a typical marker of hypoxia. To test the ability to alleviate hypoxia of nanoparticles, the 4T1 tumor bearing mice were injected with MH and MH@CAT (doses: 5 mg kg⁻¹ Ce6 and 5 mg kg⁻¹ MnO₂), respectively. The mice were sacrificed 24 h after injection and the tumor tissues were collected and stained with antibodies of HIF-1α (Affinity Biosciences, AF1009, 1 : 200).

Blood biochemical analysis

To further evaluate the biosafety of nanoparticles, the serum of mice with different treatments was collected to measure the blood biochemical indexes, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), creatinine (CREA).

Results

Preparation and characterization of MH@CAT

The MH@CAT nanoparticles were synthesized via a two-step reverse solvent precipitation and cross-linking approach, starting from the MnO₂–HSA precursor. Briefly, MnCl₂ and HSA were first mixed, and then NaOH was added dropwise to adjust the solution pH to 12, yielding the MnO₂–HSA intermediate.¹⁶ This MnO₂–HSA was subsequently stirred with glutaraldehyde and acetone to induce cross-linking and precipitation, affording the core MH nanoparticles. The size of the resulting MH was found to be inversely correlated with the MnO₂–HSA to acetone volume ratio, with a 1 : 3.5 ratio producing the smallest MH particles around 105 nm (Fig. S1A). The amount of glutaraldehyde cross-linker was also optimized to 4/900 (v/v) of the total reaction volume, ensuring complete cross-linking while maintaining a favorable size range (Fig. S1B). To prepare the final MH@CAT, the pre-formed MH nanoparticles were mixed with CAT, followed by the addition of a small amount of acetone and glutaraldehyde. The acetone facilitated the deposition of CAT onto the MH surface, while the glutaraldehyde cross-linking stably immobilized the CAT coating.

Next, the drug loading capability of MH@CAT was evaluated by mixing it with various therapeutic agents, including cisplatin (CisPt), chlorin e6 (Ce6), doxorubicin (DOX), methylene blue (MB), and rose bengal (RB). As shown in Fig. 2(A), the color of the loaded MH@CAT nanoparticles matched the colors of original drug molecules (the CisPt molecule is colorless, so MH@CAT–CisPt and MH@CAT have no obvious color difference), indicating successful encapsulation. Quantitative analysis revealed that MH@CAT possesses excellent drug loading capabilities for the aforementioned drugs. Specifically, the drug loading efficiencies of MH@CAT toward CisPt, Ce6, DOX, MB, and RB were determined to be 47.6 ± 2.6%, 47.0 ± 2.3%, 98.0 ± 3.1%, 97.9 ± 2.2%, and 89.8 ± 2.3%, respectively. This demonstrated the versatile physical adsorption-based drug loading capability of the MH@CAT platform, and the differences in drug loading capacities among various drugs may be attributed to the combined effects of multiple factors, such as the charge properties and molecular weights of the drug molecules. In this study, we focused on further evaluating the performance of the MH@CAT–Ce6 construct for tumor photodynamic therapy.

Fig. 2 (A) Photographs of MH@CAT loaded with different drugs. (B) TEM images of MH, MH@CAT and MH@CAT–Ce6. (C) Hydrodynamic sizes of MH, MH@CAT and MH@CAT–Ce6. (D) Elemental mapping images of MH@CAT. (E) XPS spectrum of the Mn element of MH@CAT. (F) Zeta potential of MH, MH@CAT and MH@CAT–Ce6. (G) UV-vis absorption spectra of MH, MH@CAT and MH@CAT–Ce6.

Extensive characterization was carried out on the intermediates and final products during the MH@CAT–Ce6 preparation process. As shown in Fig. 2(B), the prepared MH, MH@CAT, and MH@CAT–Ce6 were all monodisperse spheres, and the CAT coating increased the size of MH, but the loading of Ce6 did not affect the size of the nanocarrier. The DLS results also exhibited the same trend (Fig. 2(C)). Quantitative analysis showed that the size of MH@CAT was about 30 nm larger than that of MH. It is worth noting that the DLS sizes were consistently larger than the TEM sizes due to the presence of a hydration layer in the aqueous environment. Elemental mapping (Fig. 2(D)) confirmed the distribution of manganese within the nanoparticles, and X-ray photoelectron spectroscopy (XPS) further validated that manganese existed in the form of MnO₂ (Fig. 2(E) and S2). Importantly, all nanoparticle formulations exhibited a negatively charged surface (Fig. 2(F)), which facilitated their stable suspension in physiological solutions without aggregation, as evidenced by the unchanged DLS size and lack of sedimentation over 24 hours (Fig. S3). Finally, the characteristic absorption peaks of Ce6 at 405 nm and 650 nm confirmed the successful loading of this photosensitizer onto the MH@CAT platform (Fig. 2(G)).

Oxygen generation and GSH consumption of MH@CAT–Ce6

Nowadays, nanocarriers' functions have expanded beyond drug delivery. The regulation of the tumor microenvironment is often incorporated into the functionality of nanocarriers.^43,44 The core design of the MnO₂ nanoparticles in this study lies in enabling MH@CAT to exhibit programmed responses to the tumor microenvironment. First, the incorporated CAT converts tumor microenvironment associated high-concentration H₂O₂ into oxygen to alleviate tumor hypoxia while simultaneously provides essential components for PDT. Additionally, the CAT coating effectively protects MnO₂ from H₂O₂ caused degradation, preserving more MnO₂ to suppress intracellular GSH (Fig. 3(A)). These functions were systematically verified by simulating the states of MH@CAT in contact with H₂O₂ and GSH in the solution.

Fig. 3 (A) Schematic illustration of MH@CAT's ability to generate O₂ and reduce GSH levels in the tumor microenvironment. (B) The O₂ bubbles produced by MH@CAT and MH@CAT–Ce6 in the presence of H₂O₂. (C) O₂ generation curves of MH@CAT and MH@CAT–Ce6 incubated with H₂O₂. (D) Photographs of MH and MH@CAT solutions before and after treatment with H₂O₂. (E) The ability of MH and MH@CAT to resist varying concentrations of H₂O₂. The ability of MH and MH@CAT to resist (F) low-concentration GSH and (G) high-concentration GSH. (H) Schematic illustration of the experimental design and corresponding reaction mechanism for the programmed response of MH@CAT to H₂O₂ and GSH. (I–K) The remained MnO₂ of MH and MH@CAT after programmed degradation in different H₂O₂/GSH solutions mimicking the tumor microenvironment (CAT final concentrations: 4.0 mg mL⁻¹, 3.3 mg mL⁻¹, and 5.7 mg mL⁻¹, respectively).

Initially, we explored the oxygen production ability based on the interaction between MH@CAT and H₂O₂. As shown in Fig. 3(B), when MH@CAT with or without Ce6 was incubated with 1 mM H₂O₂ in a tube, abundant bubbles appeared on the tube wall, suggesting decomposition of H₂O₂ and generation of oxygen. To further confirm and quantify the production of oxygen, the change of oxygen concentration in the solution after mixing the nanoparticles with H₂O₂ was tested over time. Notably, both MH@CAT and MH@CAT–Ce6 increased the oxygen concentration in the original solution by about 8 mg L⁻¹ within 10 minutes of reaction with 1 mM H₂O₂ (Fig. 3(C)). Moreover, the oxygen production rate of MH@CAT and MH@CAT–Ce6 systems was basically the same, which means that the loading of Ce6 does not adversely affect the inherent oxygen-generating capacity of MH@CAT. Also, the O₂ rate gradually increased with the increasing MH@CAT–Ce6 concentration (Fig. S4). Overall, we have verified the hypothesis that MH@CAT and MH@CAT–Ce6 can interact with H₂O₂ to produce oxygen.

Although conventional MnO₂ nanoparticles also possessed capability to produce oxygen, bare MnO₂ degraded quickly in microenvironments with high concentrations of H₂O₂, thereby losing the ability to regulate GSH. Our ingeniously designed MH@CAT whose MnO₂ core was protected by the crosslinked HSA/CAT layer solves these problems. As shown in Fig. 3(D), after mixing MH and MH@CAT with 50 µM H₂O₂ for 30 minutes, the color of the MH group became significantly lighter than that of the MH@CAT group. The color depth of the solution represented the actual amount of MnO₂, which means that catalase in MH@CAT has the ability to resist degradation of MnO₂. Furthermore, the resistance of MH and MH@CAT to varying H₂O₂ concentrations was investigated. It is found that, in the presence of 50 µM H₂O₂, MH and MH@CAT retained 20% and 80% of the initial MnO₂ content, respectively. Under 100 µM H₂O₂ conditions, MH@CAT still preserved 60% of MnO₂, whereas only 5% remained in MH (Fig. 3(E)). These results collectively demonstrated that MH@CAT possess the properties of resisting the degradation of MnO₂, highlighting the critical role of the CAT coating in maintaining the structural integrity of MnO₂ nanoparticles within H₂O₂-rich microenvironments.

In the tumor microenvironment, extracellular and intracellular GSH levels range from micromolar (∼20 µM) to millimolar (∼10 mM), respectively.⁴⁵ To evaluate its response characteristics to GSH, the MnO₂ retention of MH@CAT under varying GSH concentrations was therefore measured. As shown in Fig. 3(F), MnO₂ degradation remained negligible at GSH levels below 200 µM. However, at GSH concentration exceeding 5 mM, approximately 80% of MnO₂ was degraded (Fig. 3(G)). These findings implied that MnO₂ remained stable against extracellular low-concentration GSH before being delivered into tumor cells.

The aforementioned results support the idea that MnO₂ in MH@CAT can be mostly reserved prior to entering tumor cells for intracellular GSH modulation before accessing tumor cells. Considering the complex microenvironment that MH@CAT may encounter during delivery, we investigated its GSH-depleting capacity by simulating the tumor microenvironment with 50 µM H₂O₂ and 5 mM GSH, and potential normal cellular/tissue microenvironments with 10 µM GSH (Fig. 3(H)). The corresponding results are shown in Fig. 3(I). First, once entering the tumor microenvironment, MH@CAT was in contact with 50 µM H₂O₂. In the first 5 minutes after exposure, MnO₂ was obviously degraded, which may be related to the defects in the MH packaging caused by CAT. In the next 25 minutes, the MnO₂ content essentially remains the same. In contrast, the MH system exhibited a pattern of continual decline as the incubation time increases. GSH (5 mM) was added at the 30-minute mark to simulate the intracellular condition. The results demonstrated that the MnO₂ content of MH@CAT steadily decreased over the course of the following half-hour, suggesting a slow decomposition. In contrast, MnO₂ was not considerably degraded by GSH (10 µM). It should be mentioned that the leftover MnO₂ in the MH system, following its reaction with H₂O₂, also shown some decrease upon meeting with GSH; however, this reducing level was far weaker than it in the MH@CAT system, suggesting that the MH@CAT system had superior GSH consumption capabilities. Furthermore, the amount of CAT addition was correlated with the GSH depletion capability of MH@CAT. The less amount of CAT resulted in limited protection of MnO₂ (Fig. 3(J)), while excessive CAT protected MH@CAT from GSH induced decomposition (Fig. 3(K)), most likely due to the dense coating layer. Overall, MH@CAT obtained by moderate CAT modification (CAT concentration: 4 mg mL⁻¹) can effectively maximize the consumption of intracellular GSH by MnO₂, together with its oxygen production performance, making it a multifunctional nanocarrier capable of altering the tumor microenvironment.

In vitro cell uptake and PDT effects of MH@CAT–Ce6

As a core functionality, the drug-delivery capacity of MH@CAT was evaluated using the photosensitizer Ce6 as the payload. This was verified in human breast cancer (MDA–MB-231) and mouse breast cancer (4T1) cell lines. Cells were incubated with 2 µM Ce6, either as the free drug or encapsulated in MH@CAT, for 2, 4, 8, and 12 hours. Fluorescence microscopy revealed a negligible Ce6 signal in the cells treated with the free drug, even after 12 hours (Fig. 4(A) and (B) and Fig. S5). In contrast, cells incubated with MH@CAT-Ce6 exhibited a strong fluorescence signal after just 2 hours. These results demonstrated that using MH@CAT as a nanocarrier significantly enhances the cellular uptake efficiency of Ce6 compared to the free drug.

Fig. 4 (A) Fluorescence images of MDA–MB-231 cells after incubation with free Ce6 and MH@CAT–Ce6 for 2, 4, 8 and 12 h, and stained with Hoechst (blue channel: Hoechst, red channel: Ce6). (B) Ce6 fluorescence intensities of cells at different time points. (C) Intracellular distribution of Ce6 in MDA–MB-231 cells after incubation with MH@CAT–Ce6. (D) Pearson correlation coefficients of Ce6 with mitochondria and lysosomes, respectively.

The intracellular distribution of Ce6s, particularly their co-localization with lysosomes and mitochondria, is crucial for PDT efficacy. We investigated whether Ce6 delivered by MH@CAT would co-localize with these organelles. After incubating MH@CAT–Ce6 with MDA–MB-231 and 4T1 cells, the corresponding fluorescent probe of mitochondria or lysosomes was added to each group, respectively. Confocal images showed that there was a significant overlap between the fluorescence signals of Ce6 and mitochondria, as well as between Ce6 and lysosomes, indicating a certain degree of co-localization between Ce6 and mitochondria and lysosomes (Fig. 4(c) and Fig. S6). In order to quantitatively measure the degree of co-localization, the Pearson correlation coefficients (PCC) between the Ce6 and organelle fluorescence signals was calculated. For MDA–MB-231 (4T1) cells, the PCC of Ce6/mitochondria and Ce6/lysosome fluorescence signals were 0.6 (0.5) and 0.85 (0.75), respectively (Fig. 4(D)). This indicates a moderate correlation with mitochondria and a strong correlation with lysosomes. Given the short-lived nature of PDT-generated reactive oxygen species,⁴⁶ this ability of MH@CAT-delivering Ce6 to associate with key cellular organelles should enhance its therapeutic effectiveness by enabling direct damage.

Then, the PDT effect of MH@CAT–Ce6 on tumor cells and its regulatory roles in intracellular oxygen and GSH levels were further investigated. First, the toxicity of MH@CAT itself was evaluated. Following a 24-hour co-culture of MH@CAT with MDA–MB-231 cells, cell viabilities remained above 85% even at doses reaching up to 500 µg mL⁻¹, indicating the low intrinsic cytotoxicity of MH@CAT (Fig. 5(A)). Subsequently, MH@CAT–Ce6, containing various amounts of Ce6, was co-incubated with MDA–MB-231 cells to examine the dark toxicity of nanomedicines. Results indicated no cell-killing effect in the absence of light irradiation, even with 50 µM H₂O₂ (Fig. 5(B)). Furthermore, MH@CAT did not induce significant hemolysis (less than 5%) even at high concentrations (up to 500 µg mL⁻¹) after incubation at 37 °C for 2 h (Fig. S7). This result indicated that MH@CAT possesses good biocompatibility and is suitable for subsequent in vivo studies. Upon exposure to laser light (650 nm), MH@CAT–Ce6 containing 1 µM equivalent Ce6 reduced cell viability to approximately 32%. The addition of 50 µM H₂O₂ further intensified the phototoxicity, decreasing viability to around 13%. In comparison, at the same Ce6 dose and light irradiation, MH–Ce6 (MH loaded with Ce6 without CAT coating) exhibited approximately 55% or 44% cell viability in the absence or presence of 50 µM H₂O₂ (Fig. 5(C)). Evidently, the cancer cell-killing capability of MH–Ce6 was inferior to that of MH@CAT–Ce6, possibly due to its weaker regulation of oxygen and GSH levels. Furthermore, when contrasted with nanoparticle-loaded Ce6, free Ce6 demonstrated over 90% cell viability in the presence of 1 µM Ce6 post light irradiation (Fig. S8), suggesting poor photodynamic killing efficiency likely stemming from its restricted cellular uptake. Subsequent live/dead staining also confirmed the superior photodynamic killing efficacy of MH@CAT–Ce6 in the presence of 50 µM H₂O₂ (Fig. 5(D)). To validate the photodynamic killing efficacy of MH@CAT–Ce6 in other tumor cells, the same experimental tests were replicated using 4T1 cells. The outcomes of carrier toxicity (Fig. S9A), MH@CAT–Ce6's dark toxicity (Fig. S9B), MH@CAT–Ce6's PDT killing capacity (Fig. S9C), and live/dead staining results (Fig. S10) were all consistent with those observed in MDA–MB-231 cells.

Fig. 5 (A) Cytotoxicity evaluation of various concentrations of MH@CAT on MDA–MB-231 cells. (B) Cytotoxicity test of MH@CAT–Ce6 on MDA–MB-231 cells with or without H₂O₂ in the absence of light irradiation. (C) PDT effect of MH–Ce6's or MH@CAT–Ce6 on MDA–MB-231 cells with or without H₂O₂, under 650 nm light irradiation (20 mW cm⁻² for 15 min) in a nitrogen atmosphere. (D) Live/dead fluorescence staining images of MDA–MB-231 cells with (+) or without (−) irradiation. The green channel is live cells. The red channel is dead cells. (E) Intracellular ¹O₂ production of MDA–MB-231 cells examined using DPBF as the indicator under 650 nm light irradiation (20 mW cm⁻² for 20 min) in a nitrogen atmosphere. (F) Intracellular O₂ generation of MDA–MB-231 cells determined by RDPP. (G) Relative RDPP intensity after various treatments. (H) The relative contents of GSH in MDA–MB-231 cells and 4T1 cells after treatment with MH or MH@CAT. Mean values and error bars are defined as mean and s.d., respectively. ***P < 0.001, **P < 0.01, *P < 0.05, n.s., not significant.

Singlet oxygen, ¹O₂, generation is central to the photodynamic killing of cancer cells. Using the DPBF probe,⁴² it was confirmed that the MH@CAT–Ce6 system produced abundant ¹O₂ upon light irradiation, with ¹O₂ levels increasing with the concentration of MH@CAT–Ce6 (Fig. 5(E) and Fig. S11). In fact, the efficacy of photodynamic tumor cell killing hinges on the amount of ¹O₂ available to act on cells, and this parameter is regulated by the balance between ¹O₂ generation and consumption. However, the main factor restricting ¹O₂ generation in the tumor microenvironment is low in oxygen, while the high concentration of intracellular GSH exacerbates the consumption of ¹O₂. MH@CAT was specifically designed to overcome these dual constraints by supplying oxygen and depleting GSH, thereby enhancing intracellular ¹O₂ levels.

Based on the above concept, RDPP (an oxygen-quenched probe) was used to detect intracellular oxygen levels under different culture conditions.^40,41 Results showed that cells co-incubated with MH@CAT and 50 µM H₂O₂ exhibited the lowest RDPP fluorescence (Fig. 5(F) and (G) and Fig. S12), indicating the highest oxygen content in this group. These findings suggested that one reason for the optimal PDT effect of MH@CAT–Ce6 in combination with H₂O₂ on tumor cells is the elevated intracellular oxygen levels, which facilitate conditions for enhanced ¹O₂ generation.

In addition, MH@CAT was originally engineered to improve the consumption of intracellular GSH via its MnO₂ component. Therefore, the relative content of GSH in cells co-incubating with MH or MH@CAT was tested. Compared with the untreated group, MH and MH@CAT induced 15% and 25% reductions in intracellular GSH levels, respectively (Fig. 5(H)). While the difference is modest, MH@CAT was significantly more effective at GSH depletion. Overall, GSH depletion impairs cellular antioxidant capacity, thereby increasing ¹O₂ levels to enhance the efficacy of PDT.

In summary, it has been demonstrated that MH@CAT–Ce6 displays superior PDT efficacy at the cellular level through several mechanisms. Firstly, MH@CAT facilitates greater Ce6 delivery into cells, and assists Ce6 to co-localize on lysosomes and mitochondria. Secondly, the CAT in MH@CAT converts H₂O₂ to O₂, thereby boosting ¹O₂ production in tumor upon light activation. Thirdly, MH@CAT exhibits enhanced ability to deplete cellular GSH levels, preventing the scavenging effect of the generated ¹O₂ and thus potentiating its cytotoxic effects on tumor cells. All these features of the MH@CAT nanoplatform contribute to its improved photodynamic anticancer performance relative to free Ce6 or the MH–Ce6 nanoparticles.

In vivo tissue distribution and PDT effects of MH@CAT–Ce6

To further evaluate the biological and anti-tumor effects of MH@CAT–Ce6 in vivo, we established a subcutaneous 4T1 breast cancer model in mice. To enable direct delivery of nanomedicine to the tumor site, fully leverage the tumor microenvironment (e.g., hypoxia, high levels of H₂O₂ and GSH) for exerting the therapeutic advantages, and minimize systemic toxicity to the greatest extent, peri-tumoral injection was selected as the administration route. Using an in vivo imaging system, the retention of free Ce6 and nanoparticle-encapsulated Ce6 in tumor was first investigated at different time points. As shown in Fig. 6(A), the fluorescence signal from free Ce6 decreased rapidly after peri-tumoral injection, whereas the nanoparticle formulations (MH–Ce6 and MH@CAT–Ce6) exhibited sustained tumor retention. Quantitative analysis revealed half-lives of the Ce6 fluorescence signal to be ∼5 h, 9 h, and 14 h for free Ce6, MH–Ce6, and MH@CAT–Ce6, respectively (Fig. 6(B)). This indicates that the MH@CAT nanoplatform significantly enhanced the retention of Ce6 at the tumor site. The rapid clearance of free Ce6 is likely due to its rapid diffusion from the subcutaneous injection site, whereas nanoparticle encapsulation slows the release kinetics. Furthermore, the additional CAT coating on MH@CAT–Ce6 may further retard Ce6 release, accounting for its extended tumor residence compared to it of the uncoated MH–Ce6 construct.

Fig. 6 (A) In vivo fluorescence images and (B) relative fluorescence intensities of 4T1 tumors in mice taken at different time points post peri-tumoral injection of free Ce6, MH–Ce6, and MH@CAT–Ce6. (C) Ex vivo fluorescence images of major organs and tumors after medicine administration for 24 h. (D) The quantitative fluorescence intensities of major organs and tumors after medicine administration for 24 h. (E) Representative immunofluorescence images of HIF-1α staining (a characteristic factor of tumor hypoxia) of tumors resected from mice in different treatment groups. The blue channel is Hoechst, and the green channel is HIF-1α. (F) Relative fluorescence intensities of hypoxia areas of each treatment groups. Mean values and error bars are defined as mean and s.d., respectively. ***P < 0.001, **P < 0.01.

We hypothesized that the duration of peri-tumoral drug retention would serve as a key determinant of subsequent tumor accumulation. To test this, tumors and organs were harvested 24 h after administration of free Ce6, MH–Ce6, or MH@CAT–Ce6, and the Ce6 fluorescence signal in each tissue was quantified using a small animal fluorescence imaging system. The results supported our hypothesis – the tumor Ce6 fluorescence intensity followed the same rank order as the drug half-lives at the injection site: MH@CAT–Ce6 > MH–Ce6 > free Ce6 (Fig. 6(C)). Quantitative results showed that the fluorescence signal within MH@CAT–Ce6-treated mouse tumors was ∼4-fold higher than that of free Ce6 (Fig. 6(D)), indicating that more Ce6 delivered by MH@CAT was enriched in the tumor tissue. Additionally, free Ce6 showed a negligible signal in the heart, liver, spleen, lungs, and kidneys, while the nanoparticle formulations displayed weak fluorescence in the liver and kidneys. This suggests that the MH and MH@CAT carriers may facilitate clearance of Ce6 through the reticuloendothelial system.

Next, the ability of MH and MH@CAT to alleviate tumor hypoxia was investigated, which is a well-known impediment to effective photodynamic therapy. Since HIF-1α is a characteristic factor of tumor hypoxia,^6,47,48 immunofluorescence staining of HIF-1α was used to identify hypoxic regions. As expected, a significant reduction in signal intensity within tumor tissue sections from mice treated with either MH or MH@CAT, indicating a significant decrease in the expression of HIF-1α in the corresponding tumor cells (Fig. 6(E)). Quantitative analysis revealed that the HIF-1α fluorescence intensity in the MH@CAT group is significantly lower than that in the control group and MH group (Fig. 6(F)), indicating the superior efficacy of MH@CAT in alleviating tumor hypoxia. The decreased fluorescence intensity observed in the MH group could be attributed to the oxygen generated via the reaction of its MnO₂ component with endogenous H₂O₂, thereby demonstrating the capacity to alleviate hypoxia to an extent. The profound hypoxia-alleviating effect of the MH@CAT nanoplatform is likely attributable to its reactivity with tumor-associated H₂O₂, thereby increasing local oxygen availability. By mitigating the hypoxic tumor microenvironment, MH@CAT can better enable the PSs to generate cytotoxic ¹O₂ upon light activation, ultimately enhancing the photodynamic anticancer efficacy.

The ultimate intention of developing novel functional nanocarriers is to use them for disease treatment. After proving the effects of MH@CAT on promoting the accumulation of PSs in tumors and regulating the tumor microenvironment, the effect of MH@CAT–Ce6-based photodynamic therapy on tumor treatment was further explored. First, a subcutaneous 4T1 breast cancer model was established in mice. When the tumor size reached about 100 mm³, peri-tumoral administration and irradiation at 650 nm wavelength were performed, and the mice's body weight and tumor size were recorded every 2 days until 18 days (Fig. 7(A)). In order to study the PDT effect of MH@CAT–Ce6, corresponding control experimental groups were also established, such as the blank control, Ce6 with light, MH–Ce6 with light, MH@CAT without light, MH@CAT–Ce6 without light and MH@CAT–Ce6 with light.

Fig. 7 (A) Treatment procedure of 4T1 tumor in mice. (B) Tumor volume growth curves, (C) tumor photographs, and (D) tumor weights of 4T1 tumor-bearing mice in each treatment group. (E) Representative images of H&E staining of tumors resected from different treatment groups. (F) The body weight growth curves of 4T1 tumor-bearing mice in each treatment group. Serum biochemical analysis (G) ALT, (H) AST, (I) BUN, and (J) CREA of mice with different treatments. In all figures, the symbol “(+)” indicated the presence of laser irradiation. Mean values and error bars are defined as mean and s.d., respectively. ***P < 0.001, **P < 0.01, *P < 0.05, n.s., not significant.

Subsequently, the tumor growth inhibition elicited by different PDT regimens was analyzed. Tumor size measurements revealed that MH@CAT–Ce6 with light treatment elicited the most potent anti-tumor effect (Fig. 7(B)). To provide more accurate tumor quantification, we excised the tumors at day 18 and directly measured their size and weight (Fig. 7(C) and (D)). These ex vivo assessments largely corroborated the in vivo findings. Importantly, MH@CAT–Ce6 alone without light showed negligible anti-tumor activity, highlighting the light-dependent nature of the PDT effect. Evidently, both light irradiation and the carrier play crucial roles in modulating the tumor microenvironment, which is indispensable for effective PDT. Histological examination of the excised tumors by H&E staining corroborated the superior photodynamic efficacy of MH@CAT–Ce6 with light treatment, which induced the most extensive tumor cell destruction (Fig. 7(E)).

Evaluation of the biocompatibility of novel nanocarriers is a critical consideration for their potential clinical translation. To this end, the systemic tolerability of the MH@CAT platform was assessed during the course of the photodynamic therapy studies. Analysis of the mouse body weight changes revealed no significant differences between the treatment groups and untreated controls over the 18-day study period, suggesting that the nanomedicine formulations and treatment regimens did not elicit overt toxicity (Fig. 7(F)). This finding is consistent with the high biocompatibility profile observed in the in vitro cell studies. Subsequent assessment was conducted for serum biochemical markers of liver (ALT and AST) and kidney (BUN and CREA) functions, aiming to evaluate organ toxicity. Notably, the levels of these indicators remained within normal ranges in mice receiving the MH@CAT–Ce6 treatment, compared to untreated controls (Fig. 7(G–J)). Histopathological analysis of major organs including the heart, liver, spleen, lungs, and kidneys also revealed no evidence of treatment-induced tissue damage in MH@CAT–Ce6 with the light group (Fig. S13). Collectively, these comprehensive biocompatibility assessments demonstrate that the MH@CAT nanoplatform exhibits an excellent safety profile, both systemically and at the tissue level, in this murine tumor model. This favorable characteristic is highly encouraging and supportive of the further development and clinical translation of this nanomedicine system for photodynamic cancer therapy applications.

Conclusions

In summary, an intelligent nanocarrier platform, MH@CAT, has been developed by coating catalase onto the surface of protein-protected MnO₂ nanoparticles. On the one hand, MH@CAT enables the encapsulation of various small-molecule drugs, thereby addressing the poor drug loading capacity of original MnO₂ nanoparticles. On the other hand, MH@CAT sequentially facilitates oxygen generation and GSH consumption in the tumor microenvironment. Upon laser irradiation, the MH@CAT–Ce6 nanoparticles demonstrated robust ¹O₂ production, which is crucial for enhancing PDT efficacy. Both in vitro and in vivo studies confirmed the excellent biocompatibility and remarkable PDT effects of the MH@CAT–Ce6 formulation. Consequently, simultaneous hypoxia relief and GSH depletion using the intelligent MH@CAT–Ce6 nanoparticles exhibit immense potential for clinical cancer treatment.

Author contributions

Weijuan Jia: investigation, methodology, data curation, and writing – original draft. Aoxue Zhang: methodology. Haiwei Hou: methodology. Yazhong Bu: supervision. Di Liu: supervision and writing – review and editing. Ching-Husan Tung: supervision and writing – review and editing. Baoji Du: supervision, writing – review & editing, funding acquisition, and conceptualization.

Conflicts of interest

There are no conflicts to declare.

Data availability

Supplementary information (SI) is available. The SI includes materials, instruments, optimization of MH synthesis, XPS spectrum and stability of MH@CAT, O₂ generation curves of MH@CAT-Ce6, in vivo cellular uptake and PDT effects of MH@CAT-Ce6 in 4T1 cells, hemolysis experiments of MH@CAT-Ce6, and H&E staining images of major organs from 4T1 tumor-bearing mice. See DOI: https://doi.org/10.1039/d5tb01925g.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 22105154), the Funding for Basic Scientific Research and the ‘Young Talent Support Plan’ of Xi’an Jiaotong University (xzy012022047).

References

1. X. Li, J. F. Lovell, J. Yoon and X. Chen, Nat. Rev. Clin. Oncol., 2020, 17, 657–674.
2. P. Agostinis, K. Berg, K. A. Cengel, T. H. Foster, A. W. Girotti, S. O. Gollnick, S. M. Hahn, M. R. Hamblin, A. Juzeniene, D. Kessel, M. Korbelik, J. Moan, P. Mroz, D. Nowis, J. Piette, B. C. Wilson and J. Golab, Ca-Cancer J. Clin., 2011, 61, 250–281.
3. C. Donohoe, M. O. Senge, L. G. Arnaut and L. C. Gomes-da-Silva, Biochim. Biophys. Acta, 2019, 1872, 188308.
4. X. Zhao, J. Liu, J. Fan, H. Chao and X. Peng, Chem. Soc. Rev., 2021, 50, 4185–4219.
5. S. Kwiatkowski, B. Knap, D. Przystupski, J. Saczko, E. Kędzierska, K. Knap-Czop, J. Kotlińska, O. Michel, K. Kotowski and J. Kulbacka, Biomed. Pharmacother., 2018, 106, 1098–1107.
6. M. R. Younis, G. He, J. Qu, J. Lin, P. Huang and X.-H. Xia, Adv. Sci., 2021, 8, 2102587.
7. J. Chen, T. Fan, Z. Xie, Q. Zeng, P. Xue, T. Zheng, Y. Chen, X. Luo and H. Zhang, Biomaterials, 2020, 237, 119827.
8. J.-p Taquet, C. Frochot, V. Manneville and M. Barberi-Heyob, Curr. Med. Chem., 2007, 14, 1673–1687.
9. E. E. Wicks and G. L. Semenza, J. Clin. Invest., 2022, 132(11), e159839.
10. A. Bansal and M. C. Simon, J. Cell Biol., 2018, 217, 2291–2298.
11. P. Kuppusamy, H. Li, G. Ilangovan, A. J. Cardounel, J. L. Zweier, K. Yamada, M. C. Krishna and J. B. Mitchell, Cancer Res., 2002, 62, 307–312.
12. Z. Zhou, J. Song, L. Nie and X. Chen, Chem. Soc. Rev., 2016, 45, 6597–6626.
13. L.-H. Fu, Y. Wan, C. Qi, J. He, C. Li, C. Yang, H. Xu, J. Lin and P. Huang, Adv. Mater., 2021, 33, 2006892.
14. Z. Yang, Y.-M. Chan, D. S.-H. Chan, C. Wu, Z. Wang, Y. Jiang, D. Liu, Z. Xia, L. Zhang, Y. Cai and C.-Y. Wong, Small Methods, 2024, 8, 2400159.
15. C. Hao, Y. Shao, J. Tian, J. Song and F. Song, J. Colloid Interface Sci., 2024, 659, 582–593.
16. Q. Chen, L. Feng, J. Liu, W. Zhu, Z. Dong, Y. Wu and Z. Liu, Adv. Mater., 2018, 30, 1707414.
17. T. Lin, X. Zhao, S. Zhao, H. Yu, W. Cao, W. Chen, H. Wei and H. Guo, Theranostics, 2018, 8, 990–1004.
18. T. P. Szatrowski and C. F. Nathan, Cancer Res., 1991, 51, 794–798.
19. X. Cheng, H. D. Xu, H. H. Ran, G. Liang and F. G. Wu, ACS Nano, 2021, 15, 8039–8068.
20. K. Liang and H. Chen, WIREs Nanomed. Nanobiotechnol., 2020, 12, e1616.
21. S. I. Lim, C. I. Lukianov and J. A. Champion, J. Controlled Release, 2017, 249, 1–10.
22. T. Yamada and T. Komatsu, ChemMedChem, 2023, 18, e202300373.
23. S. Pardhiya, R. Gautam, J. P. Nirala, N. N. Murmu and P. Rajamani, Reprod. Toxicol., 2022, 113, 136–149.
24. L. Tian, Q. Chen, X. Yi, J. Chen, C. Liang, Y. Chao, K. Yang and Z. Liu, Small, 2017, 13, 1700640.
25. B. Niu, K. Liao, Y. Zhou, T. Wen, G. Quan, X. Pan and C. Wu, Biomaterials, 2021, 277, 121110.
26. K. Li, K. Xu, Y. He, Y. Yang, M. Tan, Y. Mao, Y. Zou, Q. Feng, Z. Luo and K. Cai, ACS Nano, 2023, 17, 4667–4687.
27. S. Qin, Y. Xu, H. Li, H. Chen and Z. Yuan, Biomater. Sci., 2022, 10, 51–84.
28. J. Yu, H. Xiao, Z. Yang, C. Qiao, B. Zhou, Q. Jia, Z. Wang, X. Wang, R. Zhang, Y. Yang, Z. Wang and J. Li, Small, 2022, 18, 2106100.
29. Z. Chu, J. Yang, W. Zheng, J. Sun, W. Wang and H. Qian, Coord. Chem. Rev., 2023, 481, 215049.
30. A. Raza, U. Hayat, T. Rasheed, M. Bilal and H. M. N. Iqbal, Eur. J. Med. Chem., 2018, 157, 705–715.
31. G. Yu, Z. Yang, X. Fu, B. C. Yung, J. Yang, Z. Mao, L. Shao, B. Hua, Y. Liu, F. Zhang, Q. Fan, S. Wang, O. Jacobson, A. Jin, C. Gao, X. Tang, F. Huang and X. Chen, Nat. Commun., 2018, 9, 766.
32. C. Lennicke, J. Rahn, R. Lichtenfels, L. A. Wessjohann and B. Seliger, Cell. Commun. Signaling, 2015, 13, 39.
33. A. Najafi, M. Keykhaee, H. Khorramdelazad, M. Y. Karimi, L. Nejatbakhsh Samimi, N. Aghamohamadi, M. Karimi, R. Falak and M. Khoobi, Biomed. Pharmacother., 2022, 153, 113483.
34. X. Cheng, L. He, J. Xu, Q. Fang, L. Yang, Y. Xue, X. Wang and R. Tang, Acta Biomater., 2020, 112, 234–249.
35. C. Glorieux and P. B. Calderon, Biol. Chem., 2017, 398, 1095–1108.
36. B. Du, W. Zhang and C.-H. Tung, Chem. Commun., 2019, 55, 5926–5929.
37. W.-F. Lai, W.-T. Wong and A. L. Rogach, ACS Appl. Mater., 2020, 12, 43341–43351.
38. N. Cini and F. Calisir, Nanomedicine, 2022, 17, 1961–1980.
39. J. Borges, J. Zeng, X. Q. Liu, H. Chang, C. Monge, C. Garot, K.-F. Ren, P. Machillot, N. E. Vrana, P. Lavalle, T. Akagi, M. Matsusaki, J. Ji, M. Akashi, J. F. Mano, V. Gribova and C. Picart, Adv. Healthcare Mater., 2024, 13, 2302713.
40. N. Lu, W. Fan, X. Yi, S. Wang, Z. Wang, R. Tian, O. Jacobson, Y. Liu, B. C. Yung, G. Zhang, Z. Teng, K. Yang, M. Zhang, G. Niu, G. Lu and X. Chen, ACS Nano, 2018, 12, 1580–1591.
41. X. L. Liu, X. Dong, S. C. Yang, X. Lai, H. J. Liu, Y. Gao, H. Y. Feng, M. H. Zhu, Y. Yuan, Q. Lu, J. F. Lovell, H. Z. Chen and C. Fang, Adv. Sci., 2021, 8, 2003679.
42. Y. Li, G. He, L.-H. Fu, M. R. Younis, T. He, Y. Chen, J. Lin, Z. Li and P. Huang, ACS Nano, 2022, 16, 17298–17312.
43. R. Yang, L. Chen, Y. Wang, L. Zhang, X. Zheng, Y. Yang and Y. Zhu, Front. Immunol., 2023, 14, 1237316.
44. R. Cheng and H. A. Santos, Adv. Healthcare Mater., 2023, 12, 2202063.
45. R. Mo and Z. Gu, Mater. Today, 2016, 19, 274–283.
46. M. Lan, S. Zhao, W. Liu, C.-S. Lee, W. Zhang and P. Wang, Adv. Healthcare Mater., 2019, 8, 1900132.
47. Z. Wang, X. Gong, J. Li, H. Wang, X. Xu, Y. Li, X. Sha and Z. Zhang, ACS Nano, 2021, 15, 5405–5419.
48. X. Jing, F. Yang, C. Shao, K. Wei, M. Xie, H. Shen and Y. Shu, Mol. Cancer, 2019, 18, 157.

---