Cancer Cell Membrane Biomimetic Mesoporous Nanozyme System with Efficient ROS Generation for Antitumor Chemoresistance

Single-atom nanozymes (SAZs) with reaction specificity and optimized catalytic properties have great application prospects in tumor therapy. But the complex tumor microenvironment (low content of H2O2) limits its therapeutic effect. In this study, we developed a bionic mesoporous Fe SAZs/DDP nanosystem (CSD) for enhanced nanocatalytic therapy (NCT)/chemotherapy by simultaneously encapsulating the chemotherapeutic drugs cisplatin (DDP) and Fe SAZs with high peroxidase (POD) activity into the cancer cell membrane. CSD could evade immune recognition and actively targets tumor sites, and DDP upregulates endogenous H2O2 levels by activating nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, thereby enhancing SAZs-mediated hydroxyl radical (·OH) production, which subsequently leads to mitochondrial damage and intolerance to chemotherapy drug. We used the HGC27/DDP cell line for in vitro and in vivo experiments. The results showed that CSD achieved good therapeutic benefits, without any side effects such as inflammatory reaction. This system can induce multiple antitumor effect with H2O2 self-supply, mitochondrial damage, and ATP downregulation and eventually lead to chemosensitization.


Introduction
With the development of anticancer drugs, the quality of life of cancer patients has been gradually improved, but the emergence of multidrug resistance (MDR) has brought difficulties to chemotherapy, which is also the main reason for the failure of chemotherapy [1][2][3]. MDR is characterized by the cross-resistance of cancer cells to a variety of anticancer drugs with different structures and mechanisms of action. The mechanisms of MDR include overexpression of drug efflux proteins, changes in apoptosis signaling, and activation of DNA repair mechanisms [4]. Methods to deal with MDR include the combination of chemotherapeutic drugs with efflux protein inhibitors, proapoptotic drugs, and tyrosine kinase inhibitors. However, simple combination chemotherapy is limited by the different pharmacokinetic properties of the drugs. Application of nanocarriers is a strategy to solve the above problems, such as liposomes, polymer micelles, mesoporous silica nanoparticles, gold nanoparticles, and cell membrane coating technology [5][6][7][8][9]. Loading drugs into traditional nanocarriers can achieve the same drug distribution in vivo, and nanoscale particles can also target tumor tissues through high permeability and retention effects [9,10].
Nanozymes are a class of artificially simulated enzymes that have both the unique properties of nanomaterials and the catalytic efficiency and enzymatic reaction kinetics similar to natural enzymes [11]. Emerging single-atom nanozymes (SAZs), integrating single-atom technology and inherent enzyme-like active sites, elevate nanozyme technology to the atomic level and provide new opportunities to break through its inherent limitations. In SAZs, atomically dispersed metal centers maximize atom utilization efficiency and active site density [12]. SAZs have an excellent catalytic activity 10-100 times higher than that of traditional nanozymes [13,14]. Metal-based SAZs have great potential to regulate tumor microenvironment to generate ROS such as ·OH, which could disrupt mitochondrial activity within tumor cells, thereby downregulating ATP synthesis; thus, redox microenvironment homeostasis was disrupted, and the end result is that tumor cells were killed through apoptosis or necrosis. In this process, MDR effect was synchronously tackled [15]. However, the ·OH content catalyzed by SAZs with peroxidase (POD) activity is highly dependent on the H 2 O 2 content [16]. Intracellular reducing agents such as glutathione (GSH) balance the H 2 O 2 level, which conspicuously suppressed SAZ-mediated ·OH production [17]. Therefore, improving the intracellular H 2 O 2 concentration could effectively improve the therapeutic effect of SAZs.
In this study, a cancer cell membrane bionic mesoporous Fe SAZs/cisplatin cascade catalysis nanoplatform (CSD) was developed to weaken tumor resistance to DDP and remodel the TME for enhancing nanozyme-based NCT (Scheme 1). Tumor cell membrane biomimetic nanoparticles have good tumor homology targeting and biosafety and have been widely studied in recent years [18,19]. CSD not only have homologous targeting at the cellular level, but at the animal level. As a typical platinum-based drug in clinical chemotherapy, cisplatin (DDP) can interact with the nuclear DNA of cancer cells to form Pt-DNA adducts that could cut off intracellular gene replication and transcription, thereby triggering many cellular responses, including cell cycle arrest, DNA inhibition replication, and transcriptional processes as well as apoptosis and necrosis [20][21][22]. It also activates nicotinamide adenine dinucleotide phosphate (NADPH) oxidase to generate superoxide anion (O 2 • − ) [23,24], which is subsequently dismutated by superoxide dismutase (SOD) to promote the production of H 2 O 2 . This motivates us to combine DDP with SAZs to satisfy both tumor catalytic therapy treatment and chemotherapy, as SAZs could destroy cell homeostasis to further weaken MDR and improve cisplatin resistance. In addition, CSD could successfully escape the interception of the liver and kidney, prolonging its circulation time in the body. Eventually, DDP upregulates H 2 O 2 levels in the TME, and Fe SAZs can generate a large amount of ·OH, causing mitochondrial damage and downregulating ATP content, thereby improving the efficacy of DDP-mediated chemotherapy. This cascade reaction could achieve coadjutant NCT/chemotherapy and significantly inhibit tumor growth without physiological toxicity. This work achieves precise cancer treatment and enhanced efficacy of DDP through novel biomimetic nanotechnology, which has good clinical translation potential.

Results and Discussion
SAZs have uniformly dispersed active sites and well-defined coordination structures, which can mimic the activities of various natural enzymes and can be used for biological detection and cancer treatment [25,26]. The atomically dispersed Fe SAZs were obtained by firstly anchoring of iron ions on mesoporous carbon sphere precursors, followed by a pyrolysis process in an argon atmosphere, as shown in Figure 1(a). Firstly, mesoporous carbon sphere precursors were prepared by an organic-organic self-assembly approach that uses Pluronic F127 as soft template, ethanol/water  Oxidative Medicine and Cellular Longevity mixture as solvent, and dopamine as carbon and nitrogen source, respectively. The precursor was then mixed with Fe 3+ in water to anchor iron atom. Finally, the Fe SAZs were obtained by pyrolysis process at 600°C for 4 h. The isolated and high-density bright spots observed in aberrationcorrected high-angle annular dark field scanning transmission electron microscopy (AC-HAADF-STEM) image ( Figure 1(b)) implied the existence of single metal atoms. The uniform distribution of elements of C, N, and Fe in samples was observed in the EDS mappings (Figure 1(c)). Energy dispersive X-ray spectroscopy (EDX) ( Figure S1) result shows that Fe SAZs are composed of two elements, namely, Fe and C. And XRD energy spectrum also showed no obvious diffraction peaks of Fe and its oxides, suggesting that Fe is distributed in atomic form ( Figure 1(d)). The N 2 adsorption-desorption measurements demonstrated a mesoporous size distribution of Fe SAZs (Figures 1(e) and 1(f)). According to the Barrett-Joyner-Halenda method, the specific surface area of prepared sample was calculated as 437 m 2 /g and has a fine pore structure. According to the results of ICP-AES, the iron content is about 1.1%.

Oxidative Medicine and Cellular Longevity
Cell membrane-coated biomimetic nanoparticles are mainly composed of a layer of cell membrane wrapping functional nanoparticles, thus forming a nanoparticle corecell membrane-shell structure [27,28]. In this study, we first prepared cancer cell membrane vesicles (CV) and then encapsulated Fe SAZs and DDP into CV to obtain a biomimetic hybrid material CSD. Transmission electron microscopy (TEM) image of as-prepared Fe SAZs is shown in Figure 2(a). And CSD showed a clear gray membrane coating on the outside of the Fe SAZs (Figures 2(b) and 2(c)). The drug loading efficiency of CSD was about 17.2% calculated by UV-Vis absorption spectroscopy (Figure 2(d)). The size of Fe SAZs and CSD was quantitatively calculated for three consecutive days (Figure 2(e)), and the results showed that the single-atom enzyme we prepared had good stability. The zeta potential of Fe SAZs was about -12.5 mV, and the zeta potential of CSD modified by cancer cell membrane was close to the level of CV (Figure 2(f)). Cancer cell membrane marker, Epcam, was detected using western blotting ( Figure S2). This also indicates successful encapsulation of the cell membrane. The tumor microenvironment (TME) is different from the physiological environment of normal cells [14]. Due to the insufficient supply of nutrients and oxygen at the tumor site, glycolysis produces lactic acid, so the pH of the tumor site is usually lower than that of the normal adjacent tissues, showing a weak acid trend [29]. We carried out TMB (3,3 ′ ,5,5 ′ -tetramethylbenzidine) chromogenic reaction experiments under different pH   Figure S3), and stimulate more H 2 O 2 production. And the coincubation of CSD and H 2 O 2 will gradually release DDP, which is because the ROS generated by CSD could damage the cell membrane (Figure 2(i)). Usually, the immune system recognizes and eliminates tumor cells in the TME [30]. However, in order to survive and reproduce, tumor cells can employ different strategies to suppress the body's immune system so that it could not properly recognize and kill tumor cells, thereby surviving all stages of the antitumor immune response [31][32][33]. The above-mentioned characteristic of tumor cells is called immune escape. Tumor cell membranes retain cell adhesion molecules on the surface of the source cell, including cadherins, selectins, integrins, and immunoglobulin superfamily.
These receptors enable tumor cell membrane-coated nanoparticles to escape immune clearance and exhibit cognate targeting behaviors that greatly enhance their cancerspecific accumulation and retention [18,34,35]. We prepared erythrocyte membrane-coated Fe SAZs and DDP by a similar method, named RSD. Fe SAZs and DDP were coated with cancer cell membrane extracted from HGC27 cell line to form CSD (HGC27), which were used to function as control groups to verify the targeting ability of CSD. CSD, CSD (HGC 27), and RSD were labeled with Dil and coincubated with tumor cells (Figure 3(a)). The results showed that CSD is more easily phagocytosed by tumor cells than RSD as there are no receptor proteins on the erythrocyte membrane that recognize cancer cells. In addition, the number of CSD (HGC27) phagocytosed by HGC27/DDP was less than CSD, which may be due to the membrane protein change. In vitro cell uptake experiments were carried out with CSD, CSD (HGC 27), and RSD containing different concentrations of DDP, and the results also showed that CSD had 5 Oxidative Medicine and Cellular Longevity the best HGC27/DDP cell-targeting ability ( Figure S4). Next, we detected ·OH formation of different groups with the hydroxyphenyl fluorescein (HPF) fluorescent-staining experiment; the result showed that CSD-induced green fluorescence was the brightest (Figure 3(b)). Mitochondria, as the energy-producing structures of cells, are the main sites of cellular aerobic respiration. High concentrations of ·OH trigged by CSD group could damage mitochondria (Figure 3(c)), then the energy production channels of mitochondria are inhibited, and the content of ATP in cells is significantly reduced (Figure 3(d)). While RSD has a moderate tumor therapeutic effect, CSD with good homologous targeting ability induced the admirable killing efficiency. Initially, CSD could be recognized and preferentially phagocytosed by cancer cells, causing the release of DDP and Fe SAZs. Fe SAZs could catalyze the production of toxic ·OH from endogenous H 2 O 2 , and the subsequently generated ·OH in situ could disrupt mitochondrial activity, reducing ATP content. Simultaneously, DDP upregulate NADPH activity, which would induce tumor cells to produce more H 2 O 2 ; this repeated cycle of treatment could continuously replenish H 2 O 2 and maximize the effectiveness of DDP, thereby profoundly killing tumor cells. Cell viability in the CSD group was less than 15%, indicating that our complementary treatment was found to be effective for achieving enhanced NCT/chemotherapy. Similarly, we adjusted the loading of DDP in CSD, the concentrations of DDP were 0, 2, 4, and 8 μg/mL (Figure 3(e)), respectively, and cell viability decreased with increasing DDP concentration without exhibiting drug resistance. Similarly, we detected the fluorescence images of HGC27/DDP cells stained with FDA and PI under different formulations; the results showed that CSD could maximumly eliminate cancer cells (Figure 4(a)).
The good in vitro experimental results of CSD motivated us to continue to explore the tumor accumulation ability and therapeutic efficiency of CSD in vivo. First, a subcutaneous tumor model was established in Balb/c nude mice. We explored the in vivo pharmacokinetics and biodistribution of CSD and RSD. Due to the coating of the cell membrane, both CSD and RSD could significantly improve the circulation time in the body and reduce the accelerated blood clearance (ABC), thereby showing good long-term blood circulation. In vivo biodistribution experiment also showed that CSD and RSD achieved better immune evasion. Compared to RSD, only a small amount of CSD and CSD (HGC-27) accumulated in liver. However, RSD will not actively target tumor sites, and the constructed biomimetic nanoparticle CSD could better retain key proteins on the HGC27 cell membrane and escape immune clearance, and more importantly, CSD could specifically target drug-resistant tumor cells, resulting in better tumor accumulation (Figures 5(a) and 5(b)). In addition, ordinary cell membrane-encapsulated CSD (HGC27) have weak tumor targeting properties, which also reflects the homologous targeting ability of CSD. These results encouraged us to continue our in vivo antitumor experiments. The animals were randomly divided into four groups, each of which was subjected to one of the following treatments: (1) PBS, (2) CS, (3) RSD, and (4) CSD. When the tumor volume in these groups grew up to about 200 mm 3 , treatments were administered intravenously to each group. We measured tumor growth curves and mouse body weight every two days during antitumor period. As shown in Figure 5  Oxidative Medicine and Cellular Longevity were not affected by the various treatments. This is very important. Although many nanomaterials have good killing effect on tumors, they also have corresponding physiological toxicity, which limits their long-term biological and clinical value [36]. The treatment schedule is shown in Figure 5(c). As shown in Figure 5(d), the tumor volume of the control group increased within two weeks, while the CS group containing only Fe SAZs showed almost no vis-ible tumor suppressive effect, which is due to insufficient H 2 O 2 content in the tumor physiological environment [37][38][39], Fe SAZ-mediated ·OH production was little.
Notably, the RSD group achieved a moderate tumor suppression, and tumor volume in the CSD group was significantly suppressed during the treatment cycle. Cancer cell membrane-encapsulated CSD achieved immune escape and homologous targeting capabilities and accumulated  7 Oxidative Medicine and Cellular Longevity specifically at tumor sites, greatly improving their cancer therapeutic effects. And Fe SAZ-mediated nanozymecatalyzed therapy and DDP-mediated chemotherapy complement each other and improve the TME during the treatment, further increasing the corresponding effect of Fe SAZs/DDP. After the treatment, the tumors were weighed, and it was found that the average tumor mass in the RSD group was about 0.47 g, while that in the CSD group was only about 0.1 g, which was consistent with the tumor volume in mice ( Figure 5(e)). Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL) and Ki-67 staining of tumor tissues from all groups are shown in Figure 5(g); the CSD group showed a large amount of tumor cell apoptosis and lower cell proliferation effect. ROS can destroy the active components of tumor cells, causing damage to their cell membranes, nucleic acids, proteins, etc., eventually leading to cell death [40,41]. TME stimulated by DDP upregulates H 2 O 2 levels, thereby promoting the ROS content catalyzed by CSD. CSD exhibited the highest ROS fluorescence production, indicating that our prepared material is an ideal redox destroyer.
After the treatment, slices of the basic organs (heart, liver, spleen, lung, and kidney) of the control group and CSD group were taken for pathological toxicity analysis, as shown in Figure 6(a). There were no obvious physiological and morphological changes and inflammatory responses in both groups. And we conducted further blood biochemical analysis (Figures 6(b)-6(e)). All the indicators were normal. This result shows that the health of the mice was not affected after the treatment. The CSD we prepared not only has a good tumor inhibition rate but also exhibits far-reaching biosafety. The nanocarrier can maintain good biocompatibility of cell membranes, has relatively low cytotoxicity and immunogenicity, and would not cause significant side effects and rejection, reducing drug accumulation and toxic side effects in internal organs.

Conclusion
In conclusion, we have designed a cancer cell membrane biomimetic mesoporous single-atom nanozyme drug delivery system for self-enhancing catalytic therapy of drug-resistant tumors. CSD with cancer cell membrane modifications are

Data Availability
All data reported in this study are available upon request by contact with the corresponding author. The detailed experimental procedures are included in the Supporting Information file.

Conflicts of Interest
The authors declare no conflict of interest.

Authors' Contributions
Wenxue Tang and Xiang Li contributed equally to this work.

Acknowledgments
This work was supported by the Henan Province Medical Science and Technology Research Program Joint Construction Project (LHGJ20190326). Figure S1: EDX of Fe SAZs. Figure S2: cancer cell membrane marker, Epcam, was detected using western blotting. Figure  S3: the activity analysis of NADPH oxidase (NOX) in HGC27/DDP cells after treated with 10 μg/mL DDP for different time points. Figure S4: nanoparticle uptake by HGC27/ DDP cells at different concentrations. (Supplementary  Materials)