*3.3. Cell Experiment*

After Cu-MOF/GOD@HA endocytosis into tumor cells, Cu(-) in Cu-MOF were reduced to Cu(I) by GSH in tumor. Cu(I) catalyzes H2O2 to generate a large number of ·OH. The excessive ·OH oxidize signal molecules, cytokines, proteins, nucleic acids, carbohydrates, lipids and etc to promote apoptosis of tumor cells [1,2,4,5,7].

MTT experiments were used to evaluate the lethality of Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA [37–40]. The viabilities of MCF-7 cells treated with Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA at several concentrations (0, 10, 20, 30, 50, 70, 100 μg mL–1) levels were calculated, as illustrated in Figure 4A. The quantitative results showed that 10 μg mL–<sup>1</sup> the Cu-MOF/GOD@HA nanocomposites gave rise to an MCF-7 cell viability of 49.8%, while the same concentration of Cu-MOF and Cu-MOF/GOD produced an MCF-7 cell viability of 87.9% and 83.1%, respectively. This could be attributed to the targeting ability of HA. 20 μg mL–<sup>1</sup> the Cu-MOF nanocomposites gave rise to an MCF-7 cell viability of 84.5%, while the same concentration of Cu-MOF/GOD and Cu-MOF/GOD@HA produced an MCF-7 cell viability of 36.5% and 19.9%, respectively. This could be due to the loading of GOD. GOD catalyzes Glu to supply more H2O2 for CDT. The 50% inhibitory concentration (IC50) values of Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA for MCF-7 cells were found to be 70.8, 17.5, 9.7 μg mL–1. In general,

Cu-MOF/GOD@HA improves its own tumor lethality in two aspects, one is the catalytic ability of GOD, and the other is the targeting ability of HA. When MCF-7 cells were incubated with lower concentration Cu-MOF/GOD@HA, the targeting performance of HA plays a dominant role. When MCF-7 cells were incubated with higher concentration Cu-MOF/GOD@HA, the catalysis performance of GOD plays a dominant role. Moreover, trypan blue was used to observe the lethality of Cu-MOF/GOD@HA more intuitively as illustrated in Figure 5B. It is clearly shown that the incubation of MCF-7 cells with Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA for 4 h leads to the significant decrease on the number of cells and the morphology of MCF-7 cells became irregular. Catalytic ability of GOD and targeting ability of HA improve the lethality of Cu-MOF/GOD@HA.

**Figure 5.** Cytotoxicity of different nanocomposites under different conditions. (**A**) MCF-7 cells were incubated with Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA at concentrations of 0, 10, 20, 30, 50, 70, 100 μg mL–1. (**B**) Microscopy images of MCF-7 cells with trypan blue staining. MCF-7 cells were incubated with Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA for 4 h at the levels of 20 μg mL–1. Scale bar: 100 μm. Values of *p* < 0.05 were considered statistically significant, with \*, \*\*, \*\*\* represent *p* < 0.05, *p* < 0.01 and *p* < 0.001, respectively.

#### *3.4. In Vivo Antitumor Efficacy*

The above discussions about the Cu-MOF/GOD@HA mediated CDT mechanisms indicate that the CDT process is significantly sensitive to GSH, which to a large extent determines the efficiency of CDT. In the present case, the variation of cell viability is evaluated by regulating the level of GSH in the cancer cell interior. For this purpose, MCF-7 cells were pre-incubated with GSH at the concentration of 2.5 and 5.0 mmol L–1, respectively. Then, the cytotoxicity of Cu-MOF was assessed. As shown in Figure 6A, the increase of GSH level in tumor cell environment leads to a significant increase on the

lethality of Cu-MOF to MCF-7 cells. The corresponding IC50 values of Cu-MOF were 70.8, 24.0 and 19.1 μg mL–1, at GSH levels of 0, 2.5 and 5.0 mmol L–1, respectively. Moreover, the variation of cell viability was evaluated by regulating the level of H2O2 in the cancer cell interior. For this purpose, MCF-7 cells were pre-incubated with H2O2 at the concentration of 100 μmol L–1. Then, the cytotoxicity of Cu-MOF was assessed. As shown in Figure 6B, the increase of H2O2 level in the tumor cell environment leads to a significant increase in the lethality of Cu-MOF to MCF-7 cells. The corresponding IC50 values were 70.8 and 20.3 μg mL–1, at H2O2 levels of 0 and 100 μmol L–1. Furthermore, in order to investigate the influence of Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA on the production of intracellular ROS, ROS detection kit was used for cell staining after MCF-7 cells were incubated with 50 μg mL–<sup>1</sup> of Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA for 4 h. The results in Figure S10 indicate that the loading of GOD and targeting ability of HA are beneficial for the generation of ·OH.

**Figure 6.** (**A**) The dependence of MCF-7 cell viability on the concentration of GSH (0, 2.5 and 5.0 mmol L–1) by incubation with Cu-MOF at the levels of 10, 20, 30, 50, 70, 100 μg mL–1. (**B**) The MCF-7 cell viability on H2O2 (100 μmol L–1) by incubation with Cu-MOF at the levels of 10, 20, 30, 50, 70, 100 μg mL–1. Values of *p* < 0.05 were considered statistically significant, with \*, \*\*, \*\*\* represent *p* < 0.05, *p* < 0.01, and *p* < 0.001, respectively.

The in vivo antitumor efficacy of the Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA was investigated in nude mice bearing MCF-7. Once the tumors had grown to approximately 90 mm3, the mice were intertumorally injected with either PBS (control, 10 mmol L–1), Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA (The concentrations were all 2.5 mg kg–1) in 10 mmol L–<sup>1</sup> PBS every 2 days (*n* = 3). A total of seven injections were performed over 2 weeks. During the treatment process, the mice in control (PBS) and experimental groups exhibit virtually no difference in their body weight, indicating negligible systemic toxicity of Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA itself (Figure S11A). As shown in Figure S11B, tumor sizes in the PBS group were evidently increased from ~100 mm<sup>3</sup> to ~774 mm3. the tumor sizes in the groups by injecting 2.5 mg kg–<sup>1</sup> Cu-MOF, Cu-MOF/GOD were increased from ~90 mm<sup>3</sup> to ~238, 95 mm3, respectively. On the contrary, the tumor sizes

in the groups by injecting 2.5 mg kg–<sup>1</sup> Cu-MOF/GOD@HA were decreased from ~90 mm<sup>3</sup> to 45 mm3. At the end of the treatment (15 days), the tumor tissues were excised from the mice and weighed. As shown in Figure S11C, the tumor sizes in the experimental and groups (PBS, Cu-MOF, Cu-MOF/GOD and Cu-MOF/GOD@HA) decreased by 7.72, 2.98, 1.08 and 0.57 with respect to the initial-tumor volumes. The tumor sizes were obviously increased in the PBS group. In contract, the tumor sizes for the experimental group were remarkably decreased (Figure S11D).

In the concentration range of 10–100 μg mL–1, Cu-MOF/GOD@HA exhibited no obvious hemolytic effect, and the hemolytic rate at each concentration level was less than 4%. This result well indicated the excellent blood compatibility of the Cu-MOF/GOD@HA. Figure S11E illustrated that after H&E staining of the major organs of mice including heart, liver, spleen, lung and kidney, no obvious pathological changes in these tissues were observed in the presence of Cu-MOF/GOD@HA. In order to avoid the possible hemolysis or blood cell aggregation after intravenous injection, the hemolysis experiment of Cu-MOF/GOD@HA was carried out (Figure S12).
