Atom-precise fluorescent copper cluster for tumor microenvironment targeting and transient chemodynamic cancer therapy

Background Reactive oxygen species (ROS) have been widely studied for cancer therapy. Nevertheless, instability and aspecific damages to cellular biomolecules limit the application effect. Recently, significant research efforts have been witnessed in the flourishing area of metal nanoclusters (NCs) with atomically precise structures for targeted release of ROS but few achieved success towards targeting tumor microenvironment. Results In this work, we reported an atomically precise nanocluster Cu6(C4H3N2S)6 (Cu6NC), which could slowly break and generate ROS once encountered with acidic. The as-prepared Cu6NC demonstrated high biological safety and efficient chemodynamic anti-tumor properties. Moreover, Cu6NC enabled transient release of ROS and contained targeting behavior led by the tumor microenvironment. Both in vitro and in vivo experiments confirmed that Cu6NC demonstrated a low cytotoxicity for normal cells, while presented high cytotoxicity for tumor cells with a concentration-dependent manner. Conclusions This work not only reported a promising candidate for chemodynamic cancer therapy, but also paved the route to address clinical issues at the atomic level. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s12951-021-01207-6.


Introduction
Reactive oxygen species (ROS) are unstable molecules that contain oxygen, such as singlet oxygen, superoxide, hydroxyl radical and peroxide, etc. For the last decades, both experimental and clinical studies confirm that the alterations of ROS can affect the intracellular redox state [1][2][3]. Ever since, numerous efforts have been devoted to examine the role of ROS in cellular transformation and tumorigenesis [4,5]. It has been observed that raising ROS level could effectively promote the apoptosis of cancer cells and modulate the immune response (specifically, inhibition) towards the treatment of cancer as well as autoimmune diseases [6][7][8]. Moreover, the elevation of ROS levels also contributes to tissue regeneration [9]. Many treatments such as the exposure of cancer cells to chemotherapy and radiotherapy could upregulate ROS [10][11][12]. Nevertheless, caused by the intrinsic instability, it is difficult for ROS to realize sustainable release for persistent anti-cancer effect. Meanwhile, a well-coordinated and balanced redox system is typically present under normal physiological conditions. Thus, promoting ROS may produce cellular stress and damage, since ROS can damage cellular biomolecules (such as proteins, DNA, RNA) and result in mutations or even carcinogenesis. Hence, the controllable presentation of ROS is vital to take advantage of this double-edged sword for cancer therapy.
Over the past decades, remarkable progress has been achieved in the area of nanomaterials such as atomically precise assembled nanoclusters [13][14][15][16]. Amongst the nanoclusters, atomically precise metal nanoclusters (NCs) has explored promising potentials with fine-tuned properties [17][18][19]. The bottom-up programmed assembly of these NCs make it relatively simple to control over the size and to replace the surface organic ligand, providing good opportunities for particular applications [20][21][22][23][24]. By far, various metal based NCs (especially gold) have demonstrated effective anti-cancer properties [25,26]. For example, researchers have developed NCs that could make cancer cells more sensitive to radiation and promote the efficiency of radiotherapy [27][28][29][30]. Our previous work also explored photothermic behavior from coordinated-metal centers for cancer therapy [31]. By far, our group have designed and synthesized dozens of metal NCs with atomically precise assembly [32][33][34][35][36][37][38].
Benefiting from the precise formulas, both the structure and the physicochemical properties of atomically precise metal NCs are controllable [39,40]. Thus, one could reasonably hypothesize that with rational design, the additional introduction of the organic ligand could modulate targeting properties while the functional modification may generate sustained releasement of NCs in particular position.
Herein, we reported an atomically precise nanocluster Cu 6 (C 4 H 3 N 2 S) 6 , which were determined by the single crystal X-ray diffraction (SC-XRD). 2-mercaptopyrimidine was used to assemble the precise structures and led to a size of ~ 1 nm of each cluster as calculated from the SC-XRD analysis. Consistently, none of cardiotoxicity, kidney toxicity or liver toxicity were observed in the main organs and peripheral blood, suggesting the good compatibility of our Cu 6 NC. Specially, the structure Cu 6 (C 4 H 3 N 2 S) 6 was quite stable unless mixed in acidic environment. The slight addition of acid (pH ~ 6) would slowly break the structure and generate the burst of radicals, realizing the sustained releasement of ROS (Scheme 1). The in vitro and in vivo experiments confirmed that Cu 6 NC demonstrated a low cytotoxicity for normal cells, while presented high cytotoxicity for tumor cells with a concentration-dependent manner (Scheme 1). Such observations not only confirmed the biological safety and chemodynamic anti-tumor properties, but also indicated the targeting behavior led by the tumor microenvironment.

Results and discussions
Synthesis and characterization of Cu 6 (C 4 H 3 N 2 S) 6 nanoclusters Typically, the Cu 6 (C 4 H 3 N 2 S) 6 nanocluster (Cu 6 NC) was synthesized via one-pot route by blending 2-mercaptopyrimidine with Cu(NO 3 ) 2 ·3H 2 O (ratio of 1/1 by mole) in dimethyl formamide (detailed in "Experimental section"). According to the SC-XRD analysis, each of the ligand bridges three Cu ions. Specially, each sulfur atom of the ligand bridges two Cu ions, while each Cu ion is coordinated with two sulfur atoms and one nitrogen atom from three independent ligands (Fig. 1a). The exact structure information could be found in the Additional file 1: Table S1. As shown in Fig. 1b, the experimental Powder-XRD (PXRD) patterns were in good agreement with the simulated ones from the SC-XRD, indicating the phase purity of the as-synthesized crystalline products of Cu 6 NC. X-Ray photoelectron spectroscopy (XPS) (Additional file 1: Fig. S1a) confirmed that Cu 6 NC contained C, N, S and Cu elements. The peaks at 932.6 and 952.5 eV in the high-resolution Cu 2p spectrum (Additional file 1: Fig.  S1b) could be attributed to the 2p 2/3 and 2p 1/2 of Cu + , respectively [41,42], which was generated by the reductive mercapto groups. In the Fourier transform infrared (FTIR) spectra (Additional file 1: Fig. S2), the adsorption peaks of S-H stretch were absent in the spectra of Cu 6 NC and the C=C/C=N peaks were apparently redshifted, suggesting the formation of the coordination Scheme 1 Schematic of the Cu 6 NC for chemodynamic anti-tumor therapy. Once encountered with the acidic tumor microenvironment, the Cu 6 NC would slowly break and give birth to ROS via a sustained releasement, leading to high cytotoxicity for tumor cells between mercaptopyrimidine and Cu ions [43]. Such coordinated ligands led to considerable thermal stabilities. Under nitrogen atmosphere, the thermogravimetric analysis (TGA) revealed that large weight loss (~ 40%) was observed at 400 °C, suggesting that Cu 6 NC could be maintained from room temperature to nearly 400 °C (Additional file 1: Fig. S3).
Notably, Cu 6 NC demonstrated crucial luminescent properties, indicating potential applications for biomedical labeling and targeting. The normalized emission spectra of the Cu 6 NC showed long-wavelength emission bands of 700 nm (Additional file 1: Fig. S4), which was appealing for the biological test. Moreover, the fluorescence of Cu 6 NC was reserved in water as well as in many The experimental PXRD patterns and the simulated ones from the SC-XRD of Cu 6 NC, which were in good agreement. c The tracking monitoring of UV-vis absorption spectra for Cu 6 NC in HCl solution (pH ~ 6). d EPR of Cu 6 NC with and without the addition of HCl solution (pH ~ 6). The emerging peaks indicated the generation of ROS once HCl was added. e SEM images of the crystalline Cu 6 NC. Clearly shape could be observed. f SEM images of Cu 6 NC after the treatment of HCl solution (pH ~ 6). The crystals were totally pulverized biological media (Additional file 1: Fig. S5), indicating that Cu 6 NC held considerable stability for the potential biological applications. According to dynamic light scattering (DLS) analysis, the average size of Cu 6 NC components in aqueous was ~ 23 nm (Additional file 1: Fig.  S6), which was within the range enabling efficient uptake by tumors. Transmission electron microscopy (TEM) of the Cu 6 NC (Additional file 1: Fig. S7) demonstrated a homogenous distribution with the size around 20 nm, which was smaller than the kidney filtration threshold. Moreover, obvious light absorption of Cu 6 NC was observed at the wavelength of 370 nm during the solid state UV-vis absorption spectra test (Additional file 1: Fig. S8). Since the structure was quite stable in water, we further tracked the UV-vis absorption spectra curves of Cu 6 NC in the HCl solution (pH ~ 6), observing a slowly structure changes (Fig. 1c). Meanwhile, the PXRD revealed that the crystalline structures of Cu 6 NC were totally broken (Additional file 1: Fig. S9). More importantly, electron paramagnetic response (EPR) of Cu 6 NC displayed significantly changed signals with the addition of HCl solution (pH ~ 6) (Fig. 1d), proving that a large number of ROS was produced. On the contrary, none signals was detected when tested the pure ligands under the same condition (Additional file 1: Fig. S10). The morphology analysis by scanning electron microscopy (SEM) and the associated energy dispersive spectroscopy (EDS) also revealed that the crystalline Cu 6 NC with clear shape ( Fig. 1e and Additional file 1: Fig. S11) was totally pulverized ( Fig. 1f and Additional file 1: Fig. S12) after the treatment of acidic solution (pH ~ 6). This suggests that the slight addition of acid (pH ~ 6) would break the structure and generate the burst of radicals, realizing the releasement of ROS. We also performed the test of Fenton-like reactions with Cu 6 NC in different pH, and the result showed that Fenton-like reactions were stronger in lower pH conditions, supporting our results that Cu6NC can generate ROS under acid environment (Additional file 1: Fig. S13). The above results indicate that Cu 6 NC maintains a stable structure in the simulated neutral normal tissue environment (pH ~ 7.4), and at the same time, rapidly dissociates and releases the inner core and generates ROS after entering the simulated acidic tumor microenvironment (pH ~ 6).

Cytotoxicity studies of Cu 6 NC
The biological safety of Cu 6 NC was evaluated by cell viability. Both normal (cardiomyocytes, H9C2) and cancer cell lines were used as models to assess cellular responses by CCK-8 assay kit. Cu 6 NC demonstrated high cytotoxicity to tumor cells (Additional file 1: Fig. S14 and Fig. 2a). After treatment with 20 μM Cu 6 NC for 48 h, the cell viability of A375, MCF-7 and Kyse30 cancer cell lines were 26.3%, 19.6%, and 20.9%, respectively (Fig. 2a). And its ligand components showed almost no effect on tumor cell viability (Additional file 1: Fig. S15). This is consistent with the EPR results ( Fig. 1d and Additional file 1: Fig. S10), so we further confirmed the unique mechanism of Cu 6 NC leading to cytotoxicity through structural fragmentation. On the contrary, low cytotoxicity was observed in normal cardiomyocytes for Cu 6 NC (Fig. 2b). After 48 h of incubation, the inhibition rate of 20 μM Cu 6 NC was only 49.3% and the cell viability was 78.5% at the 15 μM (administration level), which was much higher than that tumor cells at the same level. As we all know, compared with normal cells, tumor cells survive through more glycolytic pathways, which leads to an acidic microenvironment [44,45]. The acid sensitivity of Cu 6 NC is the main reason for its weak cytotoxicity to normal cells compared with tumor cells. Notably, we also measured the cellular uptake capacity of Cu 6 NC into both A375 cells and normal cardiomyocytes (Fig. 2c, d) during flow cytometric test. Along with the increase of incubation durations, the proportion of positive cells was slightly raised in cardiomyocytes while obviously promoted in cancer cells. We speculated that tumor cells could attract more Cu 6 NC during the instantaneous consumption of Cu 6 NC, and the less reaction of Cu 6 NC with cardiomyocytes determines the less cellular uptake. Therefore, we believed that the uptake of Cu 6 NC was more effective in tumor cells. These phenomena confirmed that Cu 6 NC was tumor-specific.

Investigations of anti-tumor properties
Due to the selective cytotoxic property, Cu 6 NC was thoroughly evaluated for chemodynamic therapy (CDT) in vitro. The four groups of cells treated with different concentrations of Cu 6 NC (0, 2, 8 and 16 μM) were analyzed. Subsequently, the apoptosis and cell cycle were evaluated by flow cytometry after 24 h. As shown in Fig. 3a and Additional file 1: Fig. S16, Cu 6 NC promoted the apoptosis of A375 cells, thereby causing cell proliferation inhibition. Particularly, Cu 6 NC could mediate cell arrest in G2 phase ( Fig. 3b and Additional file 1: Fig. S17). Live/dead stain imaging showed more red fluorescence appeared with the increased concentration of Cu 6 NC (Fig. 2e), suggesting that the viability of cells in the Cu 6 NC decreased in a dose-dependent manner. Meanwhile, live/dead detection associated with flow cytometry (Additional file 1: Figs. S18 and S19) also demonstrated that after incubation with 0, 2, 8 and 16 μM Cu 6 NC, the cell death rates were 3.63%, 7.14%, 23.0% and 42.2%, respectively. Since the formidable metastasis ability is one of the reasons that tumor cells are difficult to completely clean, we also performed transwell migration assay in vitro and observed that Cu 6 NC could significantly inhibit tumor cells migration ( Fig. 3c and Additional file 1: Fig. S20). Furthermore, the cell colony formation assay was conducted to assess the chemodynamic efficacy of Cu 6 NC over a longer period of time.
The corresponding results revealed similar dose-dependent ( Fig. 3d and Additional file 1: Fig. S21). In the control group (0 μM), the colonies were densely packed and none effect on cell proliferation was observed. Specifically, the  cell survival fraction obviously decreased with the addition of Cu 6 NC. In the 16 μM group, even almost no colonies appeared, indicating that Cu 6 NC was excellent for CDT. Cu 6 NC induced considerable oxidative stress in vitro, leading to a specific tumor killing effect. Since ROS was detected during EPR test, we explored the underlying mechanisms of Cu 6 NC for CDT and detected the intracellular ROS by the probe DCFH-DA, which displayed green fluorescence upon oxidation by ROS. For A375 cells exposed to Cu 6 NC 12 h, the observed green fluorescence intensity was enhanced with the increasing of concentration ( Fig. 3e and Additional file 1: Fig. S22), suggesting production of ROS via Cu + -mediated Fenton reaction. In addition, by detecting the production of ROS in A375 cells at different time points (Additional file 1: Fig. S23), it was shown that Cu 6 NC could slowly and sustainably release ROS for at least 48 h. Moreover, we found that the ratio of GSH/GSSG in A375 cells treated with Cu 6 NC for 6 h was significantly reduced in a concentration-dependent manner (Additional file 1: Fig.  S24), which also indicated that the higher intracellular GSH levels in tumor cells could also be one of the possible mechanisms.
As a transition metal element and also an essential trace element for human, copper has been widely researched for its chemodynamic potential in cancer therapy [46]. Since tumor microenvironment has weakly acidic pH, low oxygen, and high glutathione, once chemodynamic properties enter such environment, chemodynamic cancer therapy is activated via its interaction with glutathione and reinforced by H 2 O 2 through release of ROS [47]. Cu 6 NC synthesized in this work is an ideal property for chemodynamic cancer therapy. On one hand, the materials characterization results showed that Cu 6 NC triggers Fenton-like reaction and generates ROS which the standard mechanism of CDT, indicating its chemodynamic performance to explain its cytotoxicity to cancer cells over normal cells. On the other hand, our biomedical assay reveals that cancer cells treated with Cu 6 NC were observed with a higher level of ROS generation and lower GSH/GSSG ratio, showing that chemodynamic cancer therapy is a mechanism of Cu 6 NC to treat cancer cells. Therefore, our results confirmed that Cu 6 NC holds a chemodynamic property with potential for the treatment of cancer.

In vivo cancer therapy
Inspired by the good in vitro CDT efficacy, tumor formation assay was performed in vivo by using human melanoma A375 tumor-bearing mice model. Mice with tumors of about 100 mm 3 were randomly divided into three groups: (1) Control group, (2) Cu 6 NC group (10 mg/kg), and (3) Cu 6 NC group (20 mg/kg). According to the corresponding dose of each group, Cu 6 NC was administered intraperitoneally every 2 days. The tumor volume was measured every other day (Fig. 4a, b). At the time of final harvest, the tumor volume in the control group increased rapidly, while in the dose of 20 mg/kg group the tumor volume was markedly reduced by about 4.5 times (compared with the control group). Besides, the measured tumor weight also decreased significantly with the increase of the treatment dose (Fig. 4d). To verify the inhibitory effect of Cu 6 NC, the proliferation and apoptosis in tumor tissues from each group were further detected. According to the immunohistochemical staining, the Ki67 protein level of the Cu 6 NC treatment group was significantly lower than that of the control group (Fig. 4c, e). Consistently, the TUNEL staining also demonstrated a promotion with the increase of the administration concentration (Fig. 4c, f ), further proving the ability of Cu 6 NC to induce apoptosis.
To evaluate the in vivo biocompatibility of Cu 6 NC, the main organs and peripheral blood of the mice were investigated after the mice were sacrificed. Specifically, H&E staining results of the mice in every group showed that no major pathological abnormalities was induced in main organs of mice (Fig. 4g). Concomitantly, none of cardiotoxicity, kidney toxicity or liver toxicity from Cu 6 NC was observed in the blood samples (Additional file 1: Fig.  S25), indicating the low in vivo toxicity of Cu 6 NC.
Therefore, the results confirmed that the mechanism of Cu 6 NC for cancer therapy includes CDT, where the break of Cu 6 NC generate ROS which plays a role in killing cancer cells [48]. Unlike many previous copper nanoparticles for CDT use where their structure is not atomically precise or with continuous release of ROS, Cu 6 NC demonstrates its strong strength in further clinical application due to its unusual chemodynamic mechanism [49,50]. Firstly, Cu 6 NC is with atomically precise structure that its detection, application and possible side effects once for clinical use can be precisely examined, where there is not uncertain components or allotropic substances existing which may cause unpredictable danger. Secondly, Cu 6 NC does not break to generate ROS under environment of normal tissues and is with minor harm to normal tissues which has been tested both in vitro and in vivo, which guaranteed its bio-safety where normal tissues and important organs are not damaged heavily. More importantly, the generation of ROS requires Cu 6 NC to interact with tumor microenvironment and break, which means it only works when getting close to cancer cells. Since Cu 6 NC is made of copper element which is an essential trace element and a ligand which has been tested to be with high biocompatibility, it breaks to unharmful components after its chemodynamic cancer therapy with cancer cells which further minimize the potential danger in clinical use.

Conclusions
ROS are unstable oxygen containing species, which make it difficult to be persistently subsistent and effective in clinical treatment. On the other hand, lacking of natural targeted properties, the surplus ROS are harmful to normal cells, generating damages or carcinogenesis. In this study, we designed and prepared Cu 6 NC with atomically precisely structures of Cu 6 (C 4 H 3 N 2 S) 6 , which could generate large amount of ROS in acidic environment. Such characters address the above conundrum: (1) Cu 6 NC could slowly degrade and sustainably release ROS; (2) the fragmentation of Cu 6 NC are induced under acidic environment, providing targeted properties. Thus, Cu 6 NC could be fragmented under the acidic tumor microenvironment with sustainable burst of ROS. Both in vitro and in vivo experiments confirm that Cu 6 NC not only present high cytotoxicity to tumor cells, but also ensure the biological safety. Our intelligent strategy not only develops a promising candidate for the targeted CDT of cancer cells, but also paves the way for addressing clinical issues at the atomic level.

Experimental section
Synthesis of Cu 6 (C 4 H 3 N 2 S) 6 2-Mercaptopyrimidine (≥ 99%) was purchased from Alfa Aesar. N,N-dimethylformamide (DMF) and Copper nitrate trihydrate(II) (Cu(NO) 3 , ≥ 99%) were bought from Sinopharm Chemical Reagent Co., Ltd., China. All reagents and solvents used were of commercially available reagent grade and used without further purification. 2-Mercaptopyrimidine (112 mg, 1 mmol) and Cu(NO 3 ) 2 ·3H 2 O (240 mg, 1 mmol) were dissolved in 20 ml DMF solution respectively, stir thoroughly to dissolve completely. Subsequently, add the DMF solution of 2-Mercaptopyrimidine dropwise to the solvent of Cu(NO 3 ) 2 ·3H 2 O. Cu 6 L 6 powder crystals (C 24 H 18 Cu 6 N 12 S 6 ) were obtained by filtration after vigorous stirring at room temperature for 2 h (a single crystal suitable for SCXRD measurements can be obtained by reducing the dosage in the same proportion).

Characterization
Powder X-ray diffraction (PXRD) patterns of the samples were acquired on a Riguku D/Max-2500PC X-ray diffractometer with Cu radiation (λ = 1.54178 Å). Fourier transform infrared (FT-IR) spectroscopy was conducted using a Bruker ALPHA II FT-IR spectrometer. X-ray photoelectron spectroscopy (XPS) analysis was obtained on an ESCALAB 250 instrument operated at 150 W and 200 eV with mono chromated Al Kα radiation. Thermogravimetric analyses (TGA) were performed on an SDT 2960 thermal analyzer from room temperature (RT) to 800 °C at a heating rate of 10 °C/min under a nitrogen atmosphere. Dynamic light scattering (DLS) data were obtained by Horiba nano Partica SZ-100V2. UV-vis absorption spectra were recorded using a Hitachi UH4150 UV-visible spectrophotometer in the range of 200-700 nm. Emission and excitation spectra at RT were recorded with an Edinburgh FLS 1000 fluorescence spectrometer, and luminescence microscopy images were recorded on an Olympus BX53 microscope. Electron paramagnetic resonance (EPR) spectra were recorded by a Bruker A 300 EPR spectrometer.

Single-crystal X-ray diffraction (SCXRD) analysis
SCXRD was performed on a Rigaku XtaLAB Pro diffractometer Cu-Kα radiation (λ = 1.54184 Å) at 200 K. Data collection and reduction were conducted with CrysAlis-Pro software. The structures were solved with intrinsic phasing methods (SHELXT-2015) and refined by fullmatrix least-squares on F2 using OLEX2, which utilizes the SHELXL-2015 module. The imposed restraints in least-squares refinement of each structure were commented in the corresponding CIF files. All non-hydrogen atoms were refined anisotropically, and the hydrogen atoms were included in idealized positions. The crystal structures are visualized by DIAMOND 3.2. The detailed information of the crystal data, data collection and refinement results are summarized in Additional file 1: Table S1.

Fenton-like reaction
The colorimetric method was used to analyze the absorbance change of methylene blue (MB), indicating that ⋅OH is produced.

Transwell migration assay
Migration assays of A375 in vitro were performed using a 24-well plate with a transwell polycarbonate permeable support (pore size, 8 μm; Corning Incorporated, Corning, USA). A375 cells were incubated and treated as described in the Apoptosis Assay. The harvested cells were resuspended in FBS-free medium and seeded in the upper chamber with 1 × 10 5 cells per well. The lower chamber was filled with 20% FBS medium (600 μL/well). The cells were then cultured for 48 h. Finally, the cells in the lower layer of the upper chambers were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Numbers of migrated cells were photographed with a fluorescence microscopy and processed by ImageJ software.

Colony formation ability identification
A375 cells were incubated and treated as described in the Apoptosis Assay. The cells were digested and seeded in 6-well plates (1000 cells per well). Then the cells are cultured in complete medium, and the medium was changed every 4 days. At 12 days, the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. The cell colonies were observed and photographed with a microscope, and the colonies containing more than 50 cells were counted using ImageJ software.

Assessment of glutathione (GSH)
Levels of GSH in A375 cells after treatment with Cu 6 NC were evaluated with GSH and GSSG assay kits (Beyotime Biotechnology, Jiangsu, China). In short, A375 cells were seeded in 6-well dishes (2 × 10 5 cells per well) and treated with different concentrations of Cu 6 NC (0, 2, 8 and 16 μM) for 6 h. Then, the cells were collected with centrifugation, and the supernatant was discarded. Samples were added protein remover M (10 mg/30 μL), subjected to 3 cycles of freezing-thawing, and then centrifugated at 1000g for 10 min at 4 °C. The supernatant was reserved for GSH measurement with GSH and GSSG assay kits according to the manufacturer's protocol.

In vivo chemodynamic therapy
All animal procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the Welfare and Ethics Review Committee of Zhengzhou University Laboratory Animal Center (Approval number: ZZU-LAC20200911 [14]). In order to generate a xenograft mouse model, on day 0, we planted A375 cells (5 × 10 6 tumor cells in 200 μL of PBS) in 6-week-old female BALB/c nude mice (Beijing Vital River Laboratory Animal Technology Co. Ltd, China) subcutaneously in the right flank. When the tumor volume reached about 100 mm 3 on the 4th day, we randomly divided the mice into three groups: normal control group (Control), Cu 6 NC group (10 mg/kg), and Cu 6 NC group (20 mg/kg), at the same time by intraperitoneal injection corresponding dose of Cu 6 NC to treat mice. Mice were exposed to these treatments every other day for a total of nine administrations. The length (L) and width (W) of the tumor were measured with a vernier caliper every 2 days, and the tumor volume (V) was calculated using the formula: V = L × W 2 /2. On the 20th day, the tumor tissues, peripheral blood, and major organs (heart, liver, spleen, lungs and kidneys) of mice were harvested. All harvested tissues were fixed with 4% formalin at least for 48 h. The fixed tissues were used for experiments such as hematoxylin and eosin (H&E) and immunohistochemistry staining. Apoptosis of tumor tissue sections was detected by staining with terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL). The proliferation of tumor tissue was detected by immunohistochemical Ki67 staining. H&E were used to stain sections of major organs in each group. All these stained sections could be observed with a fluorescence microscopy. After the peripheral blood was collected, a serum sample was obtained by centrifugation. Then, an automatic biochemical analyzer (Chemray 240 or 840, Rayto, China) was used to detect the indexes of liver function, kidney function and myocardial enzymes.

Statistical analysis
All results are presented as the mean ± standard deviation (SD). Two treatment groups were compared by Student's t test. Multiple group comparisons were performed by two-way analysis of variance with Tukey's post hoc test. All statistical analyses were carried out using GraphPad Prism 5. *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001 were deemed as significant differences. "NS" indicates no significant differences.
Additional file 1. Additional table and figures.