Dual antibody-aided mesoporous nanoreactor for H2O2 self-supplying chemodynamic therapy and checkpoint blockade immunotherapy in triple-negative breast cancer

Triple-negative breast cancer (TNBC) represents a formidable challenge due to the absence of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) expression, rendering it unresponsive to conventional hormonal and targeted therapies. This study introduces the development of mesoporous nanoreactors (NRs), specifically mPDA@CuO2 NRs, as acid-triggered agents capable of self-supplying H2O2 for chemodynamic therapy (CDT). To enhance therapeutic efficacy, these NRs were further modified with immune checkpoint antagonists, specifically anti-PD-L1 and anti-CD24 antibodies, resulting in the formation of dual antibody-aided mesoporous nanoreactors (dAbPD−L1/CD24-mPDA@CuO2 NRs). These NRs were designed to combine CDT and checkpoint blockade immunotherapy (CBIT) for precise targeting of 4T1 TNBC cells. Remarkably, dAbPD−L1/CD24-mPDA@CuO2 NRs exhibited tumor-targeted CDT triggered by H2O2 and successfully activated immune cells including T cells and macrophages. This integrated approach led to a remarkable inhibition of tumor growth by leveraging the collaborative effects of the therapies. The findings of this study introduce a novel and promising strategy for the integrative and collaborative treatment of refractory cancers, providing valuable insights into addressing the challenges posed by aggressive breast cancer, particularly TNBC. Supplementary Information The online version contains supplementary material available at 10.1186/s12951-023-02154-0.


Introduction
Triple-negative breast cancer (TNBC) is a breast cancer subtype characterized by the absence of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER-2) expression.TNBC is known for its high invasiveness, metastatic potential, propensity for relapse, and poor prognosis [1][2][3].Due to the lack of ER, PR, and HER-2 receptors, hormone therapy and targeted therapies commonly used in clinical practice are ineffective, leaving patients with limited treatment options.Chemotherapy is the primary treatment modality; however, resistance to conventional therapies arises due to the overexpression of epidermal growth factor receptor (EGFR) proteins on the cell surface, leading to short-lived responses, severe side effects, and systemic toxicity [4,5].Additionally, monotherapy targeting TNBC-specific receptors has shown limited efficacy.For instance, although EGFR is highly expressed in 70-78% of basal-like TNBC cells [6], EGFR-targeted therapy alone has been unsatisfactory.Enhanced inhibitory effects necessitate the combined use of downstream signaling inhibitors [7].Therefore, selecting appropriate specific receptors for TNBC is crucial, and standardized treatment approaches for TNBC remain elusive.
Considering these challenges, researchers turned to alternative approaches, such as immunotherapy, which showed promise in the treatment of TNBC.Immunotherapy involves harnessing the immune system's functionality and specificity to treat malignant tumors.Tumor cells interact with T cells through antigen-presenting cells, and these interactions were facilitated by specific surface protein receptors that could either stimulate or inhibit T cell activity.Proteins that inhibited T cell activity were called tumor immune escape proteins or immune checkpoints [8,9].In recent years, many studies had pointed out that TNBC was more suitable for immunotherapy using immune checkpoint inhibitors (ICIs) compared to other subtypes of breast cancer.It was because TNBC had higher levels of tumor-infiltrating lymphocytes (TILs), higher PD-L1 expression on the tumor, and a greater number of nonsynonymous mutations.These characteristics provided direct targets for ICIs, correlated with better responses to ICIs in other tumors, and gave rise to tumor-specific neoantigens, which activated neoantigen-specific T cells to mount an antitumor immune response [10].Not only was PD-L1 found to be highly expressed in TNBC, but CD24 and CD47 were also discovered to have a similar situation.However, unlike PD-L1, CD24 and CD47 protected cancer cells from attack by directly interacting with the Siglec-10 signaling pathway in macrophages.Therefore, it was necessary to block the connection between them using ICIs -CD24 and CD47 antibodies, enabling macrophages to begin phagocytosing cancer cells more effectively.In contrast to the anti-CD47 antibody, the anti-CD24 antibody demonstrated no detectable binding to human red blood cells, thereby significantly reducing toxicity.This is because CD47 is recognized as a transmembrane protein of human red blood cells [11,12].Therefore, CD24 is a potent and more appropriate anti-phagocytic "don't eat me" signaling molecule that directly protects cancer cells from attack by Siglec-10-expressing macrophages.Previous research also confirmed that by downregulating the CD24 and CD47 proteins on breast cancer cells using the tumor suppressor gene ZBTB28, the phagocytic activity of macrophages increased.Effectively blocking both proteins inhibited the proliferation of late-stage breast cancer cells [13].Therefore, in addition to PD-L1 inhibitors, CD24 inhibitors could have potentially emerged as a novel immunotherapeutic approach for treating TNBC.In the past, studies on melanoma treatment demonstrated that combination therapy had a higher objective response rate compared to monotherapy [14][15][16].Consequently, in the previous TNBC treatment strategies, there was hope to enhance the anti-tumor immunotherapeutic effect through dual or triple blockade of immune checkpoints.
Although TNBC had a relatively higher response rate to ICIs, for many patients, the efficacy of monotherapy was still insufficient.As cancer treatment evolved, the current trend gradually shifted from a single treatment modality to combination therapies, aiming to multiply the therapeutic effects.In both early TNBC and metastatic TNBC, the combination of ICIs and chemotherapy demonstrated exceptional therapeutic effects, but it also led to an increase in side effects.These side effects included anemia, nausea, hair loss, fatigue, peripheral neuropathy, neutropenia, and hypothyroidism.These side effects might have had a certain impact on the patient's quality of life and the smooth progress of treatment [17,18].Chemotherapy lacked tumor specificity, harming both tumor and normal cells, leading to side effects.Even with imaging-guided positioning, physical treatments like photothermal, ultrasound or radiation therapy might have caused damage to surrounding tissues or induced cancer metastasis [19].Chemodynamic therapy (CDT) utilized the tumor microenvironment to destroy in situ tumors by delivering biocompatible catalysts that converted H 2 O 2 into therapeutically effective reactive oxygen species (ROS) [20].In comparison to photodynamic therapy, CDT did not rely on light, photosensitizers, or oxygen, reducing limiting factors [21].Furthermore, its process of generating free radicals did not require oxygen, reducing dependency on other conditions.This treatment approach could reduce side effects experienced by patients during therapy and improve the challenges faced by other treatments in terms of tissue depth and hypoxic tumor microenvironments [22].As a result, CDT was considered a promising novel cancer treatment strategy.
Nevertheless, owing to the intricate intracellular environment of tumor cells, the therapeutic efficacy of CDT is significantly limited.Integrating CDT with other treatment modalities has emerged as a burgeoning trend in cancer therapy.For example, the integration of CDT with photothermal therapy (CDT/PTT) and CDT with chemotherapy (CDT/chemotherapy) has demonstrated a significant increase in antitumor activity compared to individual treatments [23][24][25].CDT relies on the generation of reactive oxygen species (ROS) to induce toxicity in cancer cells.A Fenton-like reaction is employed, wherein metal ions react with higher concentrations of H 2 O 2 in the acidic tumor microenvironment, leading to the production of toxic •OH [26].Previously, numerous nanomaterials have utilized transition metals with catalytic activity to trigger chemical reactions of endogenous H 2 O 2 within tumors, thereby generating free radicals that inhibit tumor formation.These transition metals, including Fe, Cu, Mn, and Co, have all been proven effective in inducing CDT [18].Fe 2+ -mediated Fenton reactions require relatively high acidity, resulting in lower catalytic efficiency.In contrast, Cu 2+ -catalyzed Fenton-like reactions can increase the reaction rate by approximately 60 times compared to Fe 2+ [27].Despite the high efficiency of Cu 2+ -catalyzed Fenton-like reactions in weakly acidic and neutral media, Cu 2+ easily dissolves in water, which may cause premature decomposition before reaching the tumor region.This could reduce the nanoparticle concentration in the tumor area and increase the risk to normal tissues.Consequently, an appropriate nanoparticle delivery platform is needed to encapsulate copper oxide nanoparticles, ensuring that the CDT reaction occurs exclusively within the tumor region.Thus, in addition to CDT, this study proposes the integration of immunotherapy as an alternative treatment approach to enhance the effectiveness of TNBC treatment while minimizing the impact on normal tissues and skin.
In this study, the development of mesoporous nanoreactors (NRs) is reported, specifically mPDA@CuO 2 NRs composed primarily of CuO 2 and mPDA.The adhesive properties of the mPDA surface, inspired by mussels [28], were utilized to successfully conjugate antibodies onto the surface of mPDA@CuO 2 NRs, resulting in the formation of dAb PD−L1/CD24 -mPDA@CuO 2 NRs (Scheme 1 A).These dual antibody-conjugated nanoreactors demonstrate potential for checkpoint blockade immunotherapy (CBIT) by effectively targeting and blocking PD-L1 and CD24 proteins present on breast cancer cells, specifically Scheme 1 (A) Schematic illumination of the formation of dAb PD−L1/CD24 -mPDA@CuO 2 NRs as a nanotherapeutic agent for H 2 O 2 self-supplying CDT and CBIT simultaneous in TNBC TNBC cells.Moreover, the dAb PD−L1/CD24 -mPDA@CuO 2 NRs exhibit targeted chemodynamic therapy (CDT) with the ability to self-supply H 2 O 2 within the tumor microenvironment, leading to efficient suppression of 4T1 breast tumors (Scheme 1B).The findings highlight that the designed dAb PD−L1/CD24 -mPDA@CuO 2 NRs significantly enhance antitumor efficacy through the synergistic effects of H 2 O 2 self-supplying CDT and CBIT, offering a promising therapeutic approach for breast cancer treatment, particularly for TNBC.

Preparation of dual antibody-aided mesoporous nanoreactors Preparation of CuO 2
Initially, 0.2 g of CuCl 2 powder was dissolved in 5 mL of deionized water (DI-H 2 O).Subsequently, 0.16 mL of a 2.573 M H 2 O 2 solution and 5 mL of a 0.05 M Tris solution at pH 8.5 were added to the CuCl 2 solution.The mixture was thoroughly mixed at room temperature for 10 min.
To remove excess reagents, the resulting CuO 2 NRs were washed three times with ethanol through centrifugation at 12,000 rpm.Finally, the obtained CuO 2 NRs were resuspended in DI-H 2 O for subsequent experiments.

Preparation of mPAD@CuO 2
In brief, 180 mg of F127 powder was dissolved in 4.5 mL of ethanol and mixed thoroughly at room temperature until the solution transitioned from turbid to transparent.Subsequently, 4.5 mL of DI-H 2 O, 45 µL of mesitylene solution, and 18 mg of DA were added to the F127 solution.The resulting mixture was continuously stirred at room temperature for 10 min.To form PDA@CuO2, 180 µL of NH 4 OH was added to the solution mentioned above, and the mixture was stirred in the dark for 45 min.Unreacted substances in the supernatant were eliminated through centrifugation at 12,000 rpm.For the production of mesoporous PDA@CuO 2 NRs (mPDA@CuO 2 NRs), the PDA@CuO 2 NRs were dispersed in a 1:1 (v:v) mixture of ethanol and acetone.The suspension was shaken for 30 min to remove the F127 template, followed by two washes with ethanol through centrifugation at 12,000 rpm (this etching process was repeated twice).Finally, the resulting precipitate was stored at 4 °C for subsequent experiments.
For preparation of dAb PD−L1/CD24 -mPDA NPs or dAb PD−L1/CD24 -mPDA@CuO 2 NRs, mPDA NPs or mPDA@CuO 2 NRs were dispersed in 200 µL of pH 8.5 0.05 M Tris buffer.Subsequently, 10 µL of anti-PD-L1 and 10 µL of anti-CD24 antibodies were added to the solution containing mPDA NPs or mPDA@CuO 2 NRs.The mixture was gently shaken overnight at 4 °C.To remove unbound antibodies, the samples were washed three times with DI-H 2 O through centrifugation at 12,000 rpm.This process yielded dAb PD−L1/CD24 -mPDA NPs or dAb PD−L1/CD24 -mPDA@CuO 2 NRs.The conjugation efficiency was determined by measuring the antibody concentration in the supernatant using ELISA.

Apparatus
The morphology of CuO 2 , mPDA NPs, and mPDA@CuO 2 NRs was characterized by transmission electron microscopy (TEM, H-7800, Hitachi), and the elemental composition of materials were characterized by energy dispersive spectroscopy on the scanning electron microscope (EDS-SEM, SU8220, Hitachi).The zeta potential and dispersion stability of materials were determined by dynamic light scattering (DLS, SZ-100, HORIBA, Japan).The catalyzed activity of mPDA@CuO 2 NRs was analyzed by UV/VIS/NIR spectroscopy (MODEL V-700, JASCO, Japan).

Studies of catalytic performance
The pH-triggered release of CuO 2 from mPDA@CuO 2 NRs for H 2 O 2 catalysis was first investigated.For this purpose, 100 µL of H 2 O 2 (1 mM) was mixed with 850 µL of mPDA@CuO 2 NRs solution at pH values adjusted to 5.5 and 7.4.The mixture was incubated at 25 °C for 15-60 min.Afterward, 50 µL of TMB was added to the solution, and the absorbance intensity at 650 nm was measured using UV/VIS/NIR spectroscopy to determine the pH-dependent release efficiency of mPDA@CuO 2 NRs.
Furthermore, to evaluate the catalytic performance of mPDA@CuO 2 NRs, the mPDA@CuO 2 NRs were pretreated in an acidic solution (pH 5.5) for 1-48 h.Then 850 µL of the pretreated mPDA@CuO 2 NRs was mixed with 100 µL of H 2 O 2 at different concentrations (0.1, 1.0, and 10 mM) at 25 °C for 1 min.Subsequently, 50 µL of TMB was added to the solution, and the absorbance intensity at 650 nm was measured using UV/VIS/NIR spectroscopy (MODEL V-700, JASCO, Japan) to determine the catalytic activity of mPDA@CuO 2 NRs.

In vitro cell studies
In this study, two TNBC cell lines, 4T1 (mouse) and MDA-MB-468 (human), were utilized.The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 2.2 mg/mL sodium carbonate, 10% fetal bovine serum (FBS), and 50 µg/mL each of gentamicin, penicillin, and streptomycin.Before seeding into experimental wells, the cells were harvested using a 0.05% trypsin-ethylenediaminetetraacetic acid (EDTA) solution and washed three times with PBS buffer (pH = 7.4).TNBC cells were seeded at a density of 1.5 × 10 4 cells per well in 96-well plates and cultured for 24 h.Subsequently, different concentrations (12.5, 25, 50, 100, and 200 µg/mL) of materials, namely mPDA NPs, dAb PD−L1/CD24 -mPDA NPs, mPDA@CuO 2 NRs, and dAb PD−L1/CD24 -mPDA@CuO 2 NRs, were added to the cells, followed by incubation for an additional 24 h.After 24 h, the culture medium was removed, and the cells were incubated with 120 µL of XTT solution for 2 h.Following this, 100 µL of XTT solution from each well was transferred to a separate 96-well counting plate.The cell viability of 4T1 or MDA-MB-468 cells was determined by measuring the optical density (OD) at 490 nm using a SpectraMax M2 microtiter plate reader.
To simulate a realistic tumor environment and investigate the generation of excessive •OH as the H 2 O 2 concentration increased, 4T1 cells were cultured at a density of 1 × 10 4 cells per well in U-end 96-well plates for 72 h to form spheroid 3D cultures.This step aimed to confirm the CuO 2 in mPDA@CuO 2 NRs as a source of excessive •OH production.The formed 4T1 tumor spheres were subsequently transferred to Transwell inserts.The wells were then supplemented with medium containing mPDA NPs, mPDA@CuO 2 NRs, Ab CD24 -mPDA@CuO 2 NRs, and Ab PD−L1 -mPDA@CuO 2 NRs at a concentration of 100 µg/mL, respectively.Following a 48-h incubation period, the appearance and morphology of the 4T1 tumor cell spheroids were examined using inverted fluorescence microscopy (Nikon eclipse Ti2, Japan).This evaluation aimed to assess the efficiency of chemodynamic therapy (CDT) for 4T1 tumor cell spheroids.
To investigate the activation of CD8 + T cells, freshly isolated CD8 + T cells from WT C57BL/6 mice were co-cultured with 4T1 cancer cells in 12-well plates for 48 h.Prior to co-culturing, the 4T1 cells were pretreated with dAb PD−L1/CD24 -mPDA NPs for 6 h at 37 °C.A blank control consisting of the co-culture system with only medium was included.After 48 h of treatment, the culture supernatants from each group were individually collected to analyze the levels of interferon-gamma (IFN-γ).

ROS generation assay
The DCFDA/H2DCFDA -Cellular ROS Assay Kit was employed to assess the generation of free radicals upon the addition of mPDA@CuO 2 NRs to cells.4T1 cells were cultured in 12-well plates at a density of 5 × 10 4 cells per well and incubated for 24 h.Subsequently, the original culture medium was removed, and mPDA@CuO 2 NRs were resuspended in DMEM culture medium containing 1 × 10 − 4 M of H 2 O 2 for an additional 24-h incubation period.Following this, the cells were washed once with PBS, and 500 µL of DCFDA solution (2.5 µM) was added for 45 min.After two additional washes with PBS, the cells were examined using inverted fluorescence microscopy (Nikon eclipse Ti2, Japan) for monitoring purposes.

Target efficiency of dual antibody-aided mesoporous nanoreactors
4T1 cells were seeded at a density of 5 × 10 5 cells in 12-well plates.Following 24 h of incubation, the original culture medium was replaced with 1 mL of fresh DMEM medium containing mPDA NPs, Ab CD24 -mPDA NPs, or Ab PD−L1 -mPDA NPs at a final concentration of 100 µg/ mL.After a 2-h incubation period, the residual materials were removed, and the cells were subjected to 10 washes with PBS.The cells were subsequently observed and recorded using inverted fluorescence microscopy.

Anti-tumour effect in vivo
All animal experiments conducted in this study were approved by the Institutional Animal Care and Use Committee of China Medical University, Taiwan, and adhered to the guidelines for experimental animal care (IACUC NO.CMUIACUC-2021-109-1).The mice were housed at a room temperature of 26 °C.Female C57BL/6J mice weighing approximately 25-30 g (5-6 weeks old) were obtained from BioLASCO (Taipei, Taiwan) and were used to validate the effectiveness of the proposed approach.Prior to the commencement of the experiment, the mice were acclimatized for a minimum of two weeks.To establish the tumor models, 4T1 cells (1 × 10 5 cells in 50 µL DMEM) were subcutaneously implanted into the mice.Tumor size was measured using a caliper every 2-3 days, and mice with tumor sizes reaching 100 mm 3 were selected for the experimental studies.
Tumor size and mouse body weight were recorded every other day, and tumor volume was calculated using the following formula:

Histology
Mice were sacrificed after 4 days of treatment with mPDA@CuO 2 NRs or dAb PD−L1/CD24 -mPDA@CuO 2 NRs, and tumor tissues were collected and stained with IFN-γ or CD68 to evaluate the ability of Ab PD−L1/CD24 -mPDA@CuO 2 NRs for T cells and CD68 + infiltrating macrophages reactivation observed under a digital microscope (TissueFAXS PLUS + HistoQuest).

Statistical analysis
The data are expressed as the mean ± standard deviation (S.D.) on the basis of at least three independent experiments.Statistical analysis was performed using Student's t-test.Differences were considered to be statistically significant for a *p value < 0.05.

Results and discussions
Characterization of mPDA@CuO 2 NRs CuO 2 clusters were initially synthesized through the reaction of copper(II) chloride (CuCl 2 ), H 2 O 2 , and alkaline Tris buffer (pH 8.5) at room temperature for 30 min.The resulting CuO 2 clusters served as a template for the preparation of mPDA@CuO 2 NRs.Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) images revealed that the CuO 2 clusters exhibited a distinctive urchin-like shape, with an average particle size of 260.4 ± 15 nm, indicating their composition of multiple CuO 2 clusters (Fig. 1A, left).On the other hand, mPDA displayed a spherical, hollow, and smooth morphology, with an average particle size of 184.0 ± 10 nm (Fig. 1A, middle).Upon the formation of mPDA@CuO 2 NRs, the central portion of the original CuO 2 clusters was coated with mPDA, while the uncoated outer part retained a meteor hammer shape, with an average particle size of 283 ± 16 nm (Fig. 1A, right).Zeta potential analysis demonstrated that the surface potential of CuO 2 clusters was 61.3 ± 2.5 mV, which changed to -23.5 ± 2.8 mV upon coating with mPDA (-42.5 ± 3.6 mV).This change can be attributed to the deprotonation of phenol groups in mPDA, resulting in a negatively charged surface at neutral pH (Fig. 1B).These results confirm the successful coating of mPDA onto the surface of CuO 2 clusters, leading to the formation of mPDA@CuO 2 NRs.Energydispersive X-ray spectroscopy (EDS) further supported the successful preparation of mPDA@CuO 2 NRs.The elemental composition analysis indicated that mPDA predominantly consisted of carbon (C) (71%), oxygen (O) (22%), and nitrogen (N) (7%), while CuO 2 clusters was primarily composed of copper (Cu) (78%) and oxygen (O) (22%).After coating mPDA onto CuO 2 clusters, the observed elements were carbon (C) (48%), nitrogen (N) (9%), copper (Cu) (20%), and oxygen (O) (23%), confirming the successful preparation of mPDA@CuO 2 NRs with approximately 20% of CuO 2 clusters remaining uncoated with mPDA (Fig. 1C).

Acid-induced •OH generation from mPDA@CuO 2 NRs
In an acidic environment, the dissociation of CuO 2 clusters into Cu 2+ and H 2 O 2 , along with the subsequent Fenton-type reaction between these dissociation products, generates •OH for cancer chemodynamic therapy (CDT).However, the rapid decomposition of CuO 2 clusters to Cu 2+ and H 2 O 2 in an acidic environment limits the duration of CDT.From the results in Fig. S1, a discernible color change in TMB to a blue-green hue, exhibiting a distinct absorbance peak at approximately 650 nm, was observed when the CuO 2 NRs were incubated in an acidic solution for 30 min, compared to mPDA@CuO 2 NRs.This suggests that the coating of mPDA can impede the rapid decomposition of CuO 2 NRs.To address this, we developed mPDA-coated CuO 2 NPs (mPDA@CuO 2 NRs) with acid-triggered degradation capability to prevent their rapid decomposition in the acidic tumor microenvironment.To demonstrate the acid-triggered H 2 O 2 self-supplying ability of mPDA@CuO 2 NRs, we suspended the NRs in an acidic solution (pH 5.5) without the addition of H 2 O 2 and incubated them for different time intervals.As expected, a noticeable color change from TMB to a blue-green color with a distinct absorbance peak at approximately 650 nm was observed after 15 min of incubation under mild acidic (pH 5.5) conditions, but not under neutral (pH 7.4) conditions (Fig. 2A).These findings confirm that the mPDA@CuO 2 NRs possess acid-responsive characteristics, allowing them to produce Fenton catalytic Cu 2+ and H 2 O 2 for selfsupplying CDT in the acidic tumor microenvironment [29].Subsequently, we performed an initial pretreatment of the mPDA@CuO 2 NRs by resuspending them in an acidic solution (pH 5.5) for 1-48 h to etch away the outer CaO 2 clusters layer and mPDA coating layer of the mPDA@CuO 2 NRs.As a result, the spikes on the surface of mPDA@CuO 2 NRs disappeared after 1 h incubation, causing a morphological transition from an urchin-like shape to a sphere-like shape (Fig. 2E, middle).The resulting mPDA@CuO 2 NRs were then mixed with different concentrations (0-10 mM) of H 2 O 2 .The results depicted in Fig. 2B demonstrated that the mPDA@CuO 2 NRs were capable of generating sufficient •OH to oxidize TMB only in the presence of high concentrations of H 2 O 2 (above 1 mM).No significant peak at 650 nm was observed when the H 2 O 2 concentration was below 1 mM during the initial 1 h.Notably, after 24 h of incubation in the acidic solution (pH 5.5), the mPDA structure gradually loosened, leading to the formation of a hollow sphere-like shape (Fig. 2E, bottom).This structural change allowed for more CuO 2 clusters to decompose into Cu 2+ and H 2 O 2 in the acidic environment, resulting in sufficient •OH generation to oxidize TMB even at lower H 2 O 2 concentrations (0.1 mM, mimicking the tumor microenvironment) (Fig. 2C&D).The inner CuO 2 clusters layer underwent Fenton reaction, generating •OH.The absorption peaks at 370/650 nm increased with the H 2 O 2 concentration and the duration of mPDA immersion in the solution, indicating that •OH generation was directly proportional to the H 2 O 2 concentration.This observation is likely due to the slow loosening of the mPDA structure in a neutral environment and its rapid loosening in an acidic environment.These results demonstrate that the mPDA coating effectively prevents the rapid decomposition of CuO 2 clusters into Cu 2+ and H 2 O 2 in a neutral environment, thus reducing potential side effects in normal tissues.

In vitro ROS generation and CDT efficacy
pH-sensitive mPDA@CuO 2 NRs can undergo decomposition within cancer cells following endocytosis, leading to a Fenton-like reaction between the released Cu 2+ and  3A demonstrate that 4T1 cancer cells incubated with mPDA@CuO 2 NRs for 6 h exhibited significantly higher green fluorescence (Fig. 3A, middle) compared to the untreated control group (Fig. 3A, left).Moreover, the ROS-associated green fluorescence signal was still observable even after incubating the cells with mPDA@CuO 2 NRs for over 12 h (Fig. 3A, right).This observation suggests that mPDA@CuO 2 NRs efficiently generate •OH within cancer cells, allowing for a sustained two-stage Fenton-like reaction and longer-lasting chemodynamic therapy (CDT).
Subsequently, we quantitatively assessed the in vitro CDT efficiency of mPDA@CuO 2 NRs against cancer cells (MDA-MB-468 cells and 4T1 cells) using the XTT assay.As depicted in Fig. 3B, MDA-MB-468 cells treated with 25 µg/mL of mPDA@CuO 2 NRs for 24 h exhibited a cell viability of approximately 85.6%.However, when an additional H 2 O 2 concentration of 0.1 mM was introduced into the culture medium, the cell viability significantly decreased to 56.2%.This decrease in cell viability can be attributed to the fact that the low concentration of mPDA@CuO 2 NRs (25 µg/mL) was insufficient to generate an adequate amount of H 2 O 2 , resulting in an inadequate production of •OH.Interestingly, when the concentration of mPDA@CuO 2 NRs was increased to 50 µg/mL, more MDA-MB-468 cells were killed.Intriguingly, mPDA@CuO 2 NRs exhibited lower toxicity towards 4T1 cells, which can be attributed to their rapid proliferation rate (Fig. 3C).Nonetheless, even at a concentration of 200 µg/mL, the cell viability of 4T1 cells decreased to 44.5%.Furthermore, when an additional H 2 O 2 concentration of 0.1 mM was introduced, the cell viability further decreased to 20.9%.These results indicate that MDA-MB-468 cells were more sensitive to mPDA@CuO 2 NRs, with an IC 50 value of 51.9 µg/ mL, while 4T1 cells had an IC 50 value of 190.7 µg/mL.Moreover, when H 2 O 2 was introduced at a concentration similar to the tumor microenvironment (0.1 mM), the IC 50 values significantly decreased to 25.6 µg/mL for MDA-MB-468 cells and 106.9 µg/mL for 4T1 cells.These findings further confirm that mPDA@CuO 2 NRs, which

Investigation of PD-L1 and CD24 expression in TNBC cells
In this study, our objective is to combine chemodynamic therapy (CDT) with checkpoint blockade immunotherapy (CBIT) for the treatment of triple-negative breast cancer (TNBC).To assess the expression levels of immune checkpoint proteins in TNBC cells, specifically PD-L1 and CD24, we conducted flow cytometry analysis on two types of human TNBC cells (MDA-MB-231, and MDA-MB-468) as well as one mouse TNBC cell line (4T1 cells).The data presented in Fig. 4 demonstrate that, among the examined cell lines, only 4T1 cells exhibit concurrent expression of both PD-L1 and CD24, with the potential for CD24 expression reaching up to 99.7%.Conversely, MDA-MB-231 and MDA-MB-468 cells exclusively express either PD-L1 or CD24.More specifically, MDA-MB-231 cells display a high expression level of PD-L1 (90.9%) without notable expression of CD24 (0.6%).Although MDA-MB-468 cells do manifest expression of both CD24 (42.1%) and PD-L1 (2.7%), the level of PD-L1 expression is extremely low, verging on negligible.From the results, it is evident that not all cells will express both PD-L1 and CD24 simultaneously.Based on these findings, we endeavored to immobilize Ab PD−L1 and Ab CD24 concurrently on the surface of mPDA@CuO 2 NRs.This led to the creation of dAb PD−L1/CD24 -mPDA@CuO 2 NRs, allowing for a broader and more comprehensive application of CBIT for TNBC.This strategy facilitates the integration of CDT and dual CBIT in treating TNBC.

Quantification of Ab PD−L1 and Ab CD24 immobilization
The quantities of Ab PD−L1 and Ab CD24 bound to mPDA@CuO 2 NRs were determined by analyzing the unbound antibodies in the supernatant using an ELISA method.The amount of Ab CD24 immobilized on 100 µg of mPDA@CuO 2 NRs increased with higher concentrations of added Ab CD24 .At 5 µg of added Ab CD24 , approximately 3.0 ± 0.2 µg of Ab CD24 was bound to 100 µg of mPDA@CuO 2 NRs, resulting in a conjugation rate of 59.9 ± 3.4% (Fig. 5A).On the other hand, the quantity of immobilized Ab PD−L1 reached saturation (1.6 ± 0.1 µg Ab PD−L1 /100 µg mPDA@CuO 2 NRs) at 3.5 µg of added Ab PD−L1 , with a conjugation rate of 46.6 ± 3.6%.Although the amount of immobilized Ab PD−L1 could be increased to 1.8 ± 0.2 µg with 5 µg of added Ab PD−L1 , the conjugation rate significantly decreased to 35.9 ± 4.2% (Fig. 5B).The reduced conjugation rate might be due to the majority of the surface area being occupied by Ab CD24 .This could possibly be attributed to the higher affinity of Ab CD24 with mPDA, resulting in a lower conjugation rate for Ab PD−L1 compared to Ab CD24 .Taken together, these results demonstrate that approximately 3.0 ± 0.2 µg of Ab CD24 and 1.6 ± 0.1 µg of Ab PD−L1 can be immobilized on 100 µg of mPDA@CuO 2 NRs.

Study of in vitro CBIT efficiency
The expression level of PD-1/PD-L1 in tumor tissue is known to be associated with clinical outcomes, tumor metastasis, and overall survival in various cancers, including melanoma, breast cancer, and pancreatic cancer.In this study, we aimed to develop dAb PD−L1/CD24 -mPDA@CuO 2 NRs to enhance the immune cells' ability to detect and eliminate cancer cells.To verify the efficacy of dAb PD−L1/CD24 -mPDA@CuO 2 NRs in checkpoint blockade immunotherapy (CBIT), we executed successful blockage of PD-L1 and CD24 on 4T1 cells.This was done by pretreating the cells with dAb PD−L1/CD24 -mPDA@CuO 2 NRs for 6 h and subsequently staining for residual, unblocked PD-L1 and CD24.The results revealed that approximately 20.2 ± 4.8% of CD24 and 38.5 ± 6.2% of PD-L1 on the 4T1 cell surface were effectively blocked by dAb PD−L1/CD24 -mPDA@CuO 2 NRs (Fig. S3).Following this, the cells were co-cultured with mouse T cells for an additional 48 h.The results shown in Fig. 6B revealed that the morphology of 4T1 cells exhibited slight differences after co-culturing with T cells compared to the control group and the group treated with dAb PD−L1/CD24 -mPDA NPs alone.This can be attributed to the PD-L1-PD1 signaling between 4T1 cells and T cells, which transmits a "don't eat me" signal to T cells [30].However, when 4T1 cells were pretreated with dAb PD−L1/CD24 -mPDA NPs to block PD-L1-PD1 signaling and reactivate T cells, the T cells successfully attacked and destroyed the 4T1 cells.Furthermore, we investigated the concentrations of IFN-γ secreted by T cells, as activated CD8 + cytotoxic T lymphocytes (CTLs) exert their antitumor effects by releasing IFN-γ, tumor necrosis factor-alpha (TNF-α), and other cytotoxins [31].As depicted in Fig. 6C, only a minimal amount of IFN-γ (302.3 ± 8.1 pg/mL at 48 h of co-culture) was secreted when T cells were directly co-cultured with 4T1 cells.However, when T cells were co-cultured for 48 h with 4T1 cells that had been pretreated with dAb PD−L1/CD24 -mPDA NPs, the concentration of secreted IFN-γ significantly increased to 1125.3 ± 28.9 pg/mL.These results confirm that dAb PD−L1/CD24 -mPDA NPs effectively block PD-L1-PD1 signaling between 4T1 cells and T cells, reactivate T cells, and promote the secretion of sufficient IFN-γ to induce cancer cell death [32].
As depicted in Fig. 7B, treatment with mPDA NPs did not exhibit significant anti-tumor efficiency due to the lack of H 2 O 2 self-supplying CDT and T-cell activation.However, at day 17, mice treated with dAb PD−L1/CD24 -mPDA NPs showed a reduction in tumor volume (1362.9± 284.8 mm 3 ) compared to the control group (3218.5 ± 498.6 mm 3 ) and mPDA NPs treated group (2809.5 ± 477.5 mm 3 ).This reduction can be attributed to the binding of dAb PD−L1/CD24 -mPDA NPs to CD24 and PD-L1 molecules on tumor cells, which can target cytotoxic T lymphocytes against tumor cells and induce cytokine production (e.g., IFN-γ).However, after 17 days of CBIT treatment, the tumor began to grow rapidly, possibly due to the incomplete blockade of PD-L1 on the tumor cell membrane by dAb PD−L1/CD24 -mPDA NPs.Therefore, combining CBIT with other treatment strategies (e.g., chemotherapy, photothermal therapy, and CDT) is necessary to achieve complete ablation of TNBC.Furthermore, Fig. 7B demonstrates that tumor growth could be effectively inhibited (463.1 ± 134.9 mm 3 at day 17) when mice received mPDA@CuO 2 NRs for H 2 O 2 self-supplying CDT.However, tumor recurrence was observed after 21 days of H 2 O 2 self-supplying CDT treatment, likely due to the revival of non-affected tumor cells caused by insufficient generation of •OH.
In this study, we integrated CBIT with H 2 O 2 self-supplying CDT to achieve better tumor inhibition.Our findings revealed that dAb PD−L1/CD24 -mPDA@CuO 2 NRs inhibited tumor growth in most mice, and no significant tumor recurrence was observed until day 32 (545.2 ± 129.6 mm 3 ).This indicates that the binding of dAb PD−L1/CD24 -mPDA@CuO 2 NRs to CD24/PD-L1 on the surface of 4T1 tumor cells resulted in the generation of •OH, T-cell activation, and secretion of IFN-γ, exhibiting a synergistic effect of CBIT and H 2 O 2 self-supplying CDT.Finally, the potential in vivo toxicity of dAb PD−L1/CD24 -mPDA@CuO 2 NRs treatment was evaluated, and no significant differences in body weight among the treatment groups were observed (Fig. 7C), indicating negligible off-target side effects for all treatments.
To further confirm the ability of the dAb PD−L1/CD24 -mPDA@CuO 2 NRs to reactivate the T-cells via blocking PD-L1-PD1 signaling, we analyzed the secretion level of IFN-γ in the tumor environment after treatment with immunohistochemistry.As shown in Fig. 7D, higher IFN-γ secretion was observed after the dAb PD−L1/CD24 -mPDA@CuO 2 NRs treatment.Semiquantification showed that the tumors treated with mPDA@CuO 2 NRs found low-density of IFN-γ + CD8 T cells (398.5 ± 89.3 IFN-γ + CD8 T cells/mm 2 ) in the tumor area.In contrast, the tumors treated with the dAb PD−L1/CD24 -mPDA@CuO 2 NRs showed significantly increased density of IFN-γ + CD8 T cells (1512.6 ± 178.4 IFN-γ + CD8 T cells/mm 2 ) in the tumor area.These results could be attributed to the high tumor accumulation of the dAb PD−L1/CD24 -mPDA@CuO 2 NRs via binding to CD24 and PD-L1 on the 4T1 tumor cells, which could block PD-L1-PD1 signaling to reactivate T-cells and trigger IFN-γ secretion.We have also verified that dAb PD−L1/CD24 -mPDA@CuO 2 NRs can reactivate macrophages by blocking CD24-Siglec-10 signaling.This was demonstrated by staining for CD68 in tumor tissues and analyzing them using immunohistochemistry post-treatment.The results, illustrated in Fig. 7E, reveal a substantial presence of CD68 + infiltrating macrophages in tumor tissues treated with dAbPD-L1/CD24-mPDA@CuO 2 NRs compared to those treated with mPDA@CuO 2 NRs.Taking together, dAb PD−L1/CD24 -mPDA@CuO 2 NRs have

Conclusions
In conclusion, this study successfully developed dAb PD−L1/CD24 -mPDA@CuO 2 NRs, tumor microenvironment-activated nanoreactors that effectively generated toxic •OH through H 2 O 2 self-supply within the tumor environment.These nanoreactors exhibited the ability to reactivate T cells and infiltrating macrophages by blocking the CD24-Siglec-10 and PD-L1-PD1 signaling pathways, leading to the secretion of IFN-γ and enhanced cancer cell killing.Upon internalization by cancer cells, the dAb PD−L1/CD24 -mPDA@CuO 2 NRs underwent decomposition within the acidic endo/lysosomal compartments, releasing Fenton catalytic Cu 2+ ions and H 2 O 2 .This decomposition led to the production of toxic •OH, which in turn induced lipid peroxidation and caused cancer cell death.The synergistic effect of H 2 O 2 self-supplying CDT and CBIT demonstrated by the dAb PD−L1/CD24 -mPDA@CuO 2 NRs in the acidic tumor microenvironment presents a promising drug-free synergistic therapy approach for breast cancer, particularly for the treatment of triple-negative breast cancer (TNBC).

Fig. 2 (
Fig. 2 (A) Colorimetric detection of •OH generated by mPDA@CuO 2 NRs at different pH values based on the TMB assay.UV-vis spectra and photographs (inset) of TMB aqueous solution incubated with different concentrations of H 2 O 2 in the presence of mPDA@CuO 2 NRs (100 µg/mL) pretreated in an acidic solution for (B) 1 h, (C) 24 h, and (D) 48 h.(E) TEM images of mPDA@CuO 2 NRs (top), mPDA@CuO 2 NRs treated in an acidic solution for 1 h (middle), and mPDA@CuO 2 NRs treated in an acidic solution for 24 h (bottom).Scale bar = 500 nm

Fig. 3 (
Fig. 3 (A) Fluorescence and bright-field images of DCFH-DA-stained 4T1 cancer cells after exposure to mPDA@CuO 2 NRs for 6 and 12 h.The scale bar represents 100 μm.(B) In vitro CDT potency of mPDA@CuO 2 NRs after 24 h of incubation with MDA-MB-468 cells in the presence of different concentrations of H 2 O 2 .The values are expressed as means ± SD (n = 8).(C) In vitro CDT potency of mPDA@CuO 2 NRs after 24 h of incubation with 4T1 cells in the presence of different concentrations of H 2 O 2 .The values are expressed as means ± SD (n = 8)

Fig. 4
Fig. 4 Flow cytometry analysis of PD-L1 and CD24 protein expression on the cell membrane of 4T1, MDA-MB-231, and MDA-MB-468 cells

Fig. 6
Fig. 6 4T1 cells were cultured in an ultra-low attachment plate to form tumor spheroids, followed by treatment with different materials in the culture medium over time.(A) The volume of 4T1 tumor spheroids after treatment with PBS (Ctrl), mPDA NPs, mPDA@CuO 2 NRs, Ab PD−L1 -mPDA@CuO 2 NRs, Ab CD24 -mPDA@CuO 2 NRs, and dAb PD−L1/CD24 -mPDA@CuO 2 NRs.Scale bar = 500 μm.(B) In vitro CBIT efficiency tested by blocking PD-L1-PD1 signaling between cancer cells and T cells using dAb PD−L1/CD24 -mPDA NPs.(C) Generation of IFN-γ from T cells after co-culturing with 4T1 cells treated with dAb PD−L1/CD24 -mPDA NPs.The values are expressed as means ± SD (n = 3).Asterisks indicate a significant difference between the T cells treated and dAb PD−L1/CD24 -mPDA NPs + T cells treated groups (Student's t-test, *p ≤ 0.05).Note: The absence of black and red bars is due to their values being extremely low, almost equivalent to 0