Cu(II)-Containing Metal-Organic Framework with 1D Hexagonal Channels for Cyanosilylation Reaction and Anticancer Activity on Osteosarcoma Cells.

A novel metal-organic framework (MOF) has been produced via Cu(NO3)2·6H2O reaction with 3-(1H-tetrazol-5-yl)pyridine (HL) in water, and its chemical formula is {[(Cu(L)2(H2O)2](H2O)8) n } (1). Due to its high density of coordinately unsaturated sites along with large one-dimensional (1D) hexagonal channels, the activated complex 1 (1a) was explored as the solvent-free heterogeneous catalyst for cyanosilylation under mind conditions. The inhibitory function of compound 1a against the survival rate of OS-732 osteosarcoma cells was evaluated via Cell Counting Kit-8 (CCK-8) detection kit. Furthermore, the Annexin V-FITC/PI detection kit and the active oxygen (ROS) detection kit was carried out to determine the cell apoptosis levels and the ROS accumulation in OS-732 osteosarcoma cells after treatment by compound 1a.

Introduction

The cyanosilication of carbonyl compounds with TMSCN is a kind of transfer organic reaction to generate the cyanohydrins. Cyanohydrins are the useful intermediates for synthesizing β-hydroxyamines, α-hydroxyketones, α-hydroxyacids, β-aminoalcohols, α-aminonitriles, and other different compounds.[1−3] A lot of diverse catalysts are active, including the Lewis acid. However, it is necessary to note that the formations of the C–C bond of the compounds are often realized via the presence of a homogeneous catalyst, which is difficult to recycle and separate.[4−6] Thus, it is extremely important to seek an efficient heterogeneous catalyst in the process of organic synthesis.

Metal–organic frameworks (MOFs) are porous crystal materials composed of metal centers and organic connectors. They are favored by chemists and material scientists because of their extensive application, strong designability, and ordered structure.[7−11] There are a few outstanding review papers summing up various kinds of organic conversion catalytic systems on the basis of the MOF in the past several years. One of the most valuable catalytic systems on the basis of MOF may be the cyanosilylation of aldehydes with trimethylsilylcyanide because the cyanohydrins are very useful as fine pharmaceutical and chemical intermediates.[12−17] On the other hand, recent investigations have exhibited that many coordination complexes based on copper have excellent anticancer activities. A variety of coordination complexes of Cu(I) and Cu (II) have been generated.[18−20] For example, [(Cu(II))(4,4′-dimethyl-2,2′-bipyridine)(acetylacetonate)(NO3)(H2O)], a copper compound (belonging to the Casiopeinas family, generated by L. Ruiz and his colleagues) is entering the first phase of clinical trials; [Cu(trishydroxymethylphosphine)4]·[PF6], HydroCuP, a phosphine copper(I) complex, it possesses antiproliferative effects with highly selective, and it has revealed good performances in the preclinical researches; Wang et al, have developed a sphere-like Cu(II)-based coordination polymer architecture, which possesses strong anticancer activities in vitro against three selected cancer lines (namely, NCI-H446, HeLa, and MCF-7). In the research, based on the solvothermal reaction approach, a new metal–organic framework (MOF) on the basis of Cu(II) {[(Cu(L)2(H2O)2](H2O)8)} (denoted as complex 1 hereafter) with high porosity can be produced via Cu(NO3)2·6H2O reacting with 3-(1H-tetrazol-5-yl)pyridine (HL) in the mixed solvent of dimethylformamide (DMF) and water. The product gained was completely investigated using thermogravimetric analyses, PXRD, diffraction of single-crystal X-ray, and Fourier transform infrared (FT-IR) spectra as well as EA. Because of its high density of the coordinately unsaturated positions along with large one-dimensional (1D) hexagonal channels in the framework, the activated complex 1 (denoted as 1a hereafter) was researched as the heterogeneous catalyst without solvent for cyanosilylation under the mind conditions. Furthermore, its catalysis mechanism was also investigated. In the biological studies, 1a’s inhibitory effect on the OS-732 osteosarcoma survival rate was evaluated for the first time. The results of CCK-8 suggested that 1a had outstanding anticancer activity in vitro and the half-maximal inhibitory concentration (IC50) value was 4.01 ± 0.12 μg/mL. The results of Annexin V-FITC/PI assay reflected that compound 1a could evidently facilitate the apoptosis levels of OS-732 osteosarcoma cells dose dependently. In addition to this, the data of H2DCF-DA determination also demonstrated that 1a could significantly cause the accumulation of ROS in OS-732 osteosarcoma cells. Above all, compound 1a exhibited excellent anticancer activity on the OS-732 osteosarcoma cells by triggering ROS accumulation and inducing OS-732 osteosarcoma cell apoptosis.

Results and Discussion

Crystal Structure of Complex 1

The targeted 1 was generated through the reaction between HL ligand and Cu(NO3)2·3H2O in the mixed solvent of DMF and water at 100 °C for 72 h, which was gained as the blue crystals in moderate yield. The chemical formula of complex 1 was determined via combining the TGA measurement, the EA along with the diffraction of the single-crystal X-ray, which was found to be {[(Cu(L)2(H2O)2](H2O)8)}. The +2 oxidation valence of the Cu ion in 1 has been confirmed by its blue crystal as well as the bond valence sum (BVS) software, which shows that the value of center Cu(II) ion is 1.97.[21] In accordance with the data of single crystal acquired under room temperature, the structural optimization outcomes display that 1 situates in the space group R–3c of the triangular crystal system, and it revealed the three-dimensional (3D) supramolecular net involving the 1D hexagonal channels generated through the two-dimensional layer stacking. The structural analysis suggested that the fundamental molecular repeating unit consists of 0.5 Cu(II) ion, a completely deprotonated organic ligand of L–, and four disordered molecules of DMF along with a terminal coordinated molecule of H2O as suggested via TGA. According to Figure a, the Cu1 center shows an octahedral coordination geometry, which is completed through two tetrazol nitrogen atoms and two pyridinyl nitrogen atoms originate from four ligands of L, and the other two binding positions are completed through two coordinated molecules of H2O. The lengths of the Cu–N bond span from 2.018(4) and 2.036(4) Å, and the Cu–O bond length is 2.426 Å; on the basis of several nitrogen-donor ligands, these values are all in the normal range of the Cu–N and Cu–O bond distances in other Cu(II)-based MOFs.[22−24] For the ligand of L–, there is only one coordination pattern, namely, via prying the nitrogen atom, two copper(II) cations separated of the crystal are linked with a tetrazolyl atom (Figure b). The tetrazol ring is not coplanar with a pyridinyl ring, which creates a 62.137° dihedral angle. Each copper ion is linked to four neighboring copper ions by the ligand of L, and each ligand of L acts as a two-linked node. In this connection pattern, a two-dimensional hierarchical net containing large triangular windows was found. According to the radius of van der Waals, the window size is 9.62 Å (Figure c). In addition, the superposition of the two-dimensional layers on the c-axis provides a three-dimensional supramolecular net having one-dimensional channels, and the channel size is 7.12 Å (Figure d). It can be found that a large number of coordination molecules of water pointed to the center of the channel, which can generate the open metal sites with high density after the removal of the coordination molecules of water. The solvent 1’s free volume is 49.3% without considering the coordinated molecules of H2O. The outcomes suggested that 1 possesses a large number of free N atoms and open copper(II) positions as the Lewis base positions, which reveals great application potential, especially in catalysis. From the topological point of view, each Cu ion could be considered as a four-connected node bonded with four adjacent L– ligands, and each ligand of L– could be regarded as a two-linked connector, thus 1’s entire skeleton could be reduced to a four-linked network with kgm type with {3∧2.6∧2.7∧2} point symbol (Figure S2).

Figure 1

(a) 1’s coordination surrounding. (b) View of the coordination patterns for the ligand. (c) View of 1’s two-dimensional layered network. (d) 1’s three-dimensional packing view displaying one-dimensional channels.

PXRD, TGA Analysis, and Gas Sorption Property

For the determination of the product phase purity, the powder X-ray diffraction patterns on the complex created were harvested under ambient temperature. Figure a reflects the 1’s patterns of PXRD, and for the simulated PXRD patterns, its single-crystal diagram is conformed to the results of the experiment based on the data of diffraction, and this reflects the structural integrity and phase purity of the generated samples. TGA was implemented to study the thermal stability of the as-prepared complex 1 in an atmosphere. In accordance with Figure b, the curve of TGA reveals 20.1% of weightlessness from 30 to 240 °C, which is the same as the release of eight lattice water molecules and two coordinated water from each formula unit (calcd: 19.8%). After further heating, in the temperature range of 245–470 °C, there is no obvious weight loss phenomenon, indicating that the solvents are completely removed in this range of temperature. Between 470 and 600 °C, the second weightlessness corresponds to the collapse of the structural skeleton and the decomposition and organic ligands. To obtain completely activated 1 (hereinafter referred to as 1a), the newly as-generated complex 1 was immersed into MeOH for 72 h, and the sample was then heated for 72 h at 80 °C to acquire the solvent-free 1a product. The integrity of the skeleton was confirmed via the PXRD measurement, and the existence of lattice solvent was proved through the TGA analysis. At 77 K, the N2 adsorption was conducted on the activated samples to explore the 1a’s permanent porosity and its specific surface area. From the results illustrated in Figure c, for the complex 1a, the N2 gas adsorption has type I reversible isotherm. The surface areas calculated by BET and Langmuir, respectively, are 783 and 1074 m2/g. The experimental micropore volume is 0.36 cm3/g, which is close to the theoretical value. Based on the measured structure of the crystal, the analysis of DFT indicates that the pore size of 1 is approximately 7.8 Å.

Figure 2

(a) 1’s PXRD models. (b) Complex’s TGA curves. (c) Sorption data of N2 for complex 1a at 77 K. (d) 1a’s pore size distribution calculated from the DFT method.

Catalytic Cyanosilylation Reaction

Considering a larger metal pore density along with window dimensions of the synthesized skeleton 1a, we utilized 4-nitrobenzaldehyde as the model substrate to test its activity as the solid heterogeneous catalyst in different aldehydes cyanation and carried out cyanosilylation via utilizing the mixture of trimethylsilylcyanide, aldehyde, and 1a among CH2Cl2 under the ambient temperature. Ninety-four percent of 4-nitrobenzaldehyde can be transformed into 2-(4-nitrophenyl)-2-[(trimethylsilyl)oxy]acetonitrile when skeleton 1a was used as 2 mol % catalyst and after being placed in dichloromethane under the ambient temperature for 10 h (item 1, Table ). With the reaction time extended to 24 h, the yield of the product increased slightly, only 96%. Furthermore, without other products being determined, the product yield was regarded to be 4-nitrobenzaldehyde conversion. Under the same conditions, the yield of complex 1 is only 32% when it is utilized as a catalyst. Although the relationship between the structure and catalytic activity is not distinct in the research, compound 1a’s higher dialog rate than that of 1 may finally be associated with its one-dimensional nanosize channel, which is used for accessible metal centers and higher Lewis acidity at the Cu(II) site. The influences of the amount of solvent and catalyst on the cyanation of 1a catalyst were researched. When the loading of the catalyst improved between 1.0 and 2.0 mol %, the product yield increased from 75 to 94% (items 6 and 7, Table ), while the loading of the catalyst enhanced deep (reach 5%), the yield only enhanced 1% (Table , item 8). We also investigated the catalytic reactions among distinct solvents, including CHCl3, THF, MeCN, CH2Cl2, and MeOH, among which 94% yield of CH2Cl2 was the best for this transformation, while 63% yield of THF was the worst (item 7, Table ). We carried out a blank test via 4-nitrobenzaldehyde with the condition of without catalyst under ambient temperature; after a 10 h reaction, the conversion was only 12%. At the same time, the reactivity of free ligand (HL) and Cu(NO3)2·3H2O in CH2Cl2 was investigated, and the acquired yields were 16 and 21%, respectively (Table , items 10 and 11).

Table 1

Refinement for the Parameters of Cyanosilylation for 4-Nitrobenzaldehyde with 1a as well Aas TMSCNa

entry	cat	quant.a (mol %)	time	solvent	conv. (%)b
1	1a	2	10	CH2Cl2	94
2	1a	2	24	CH2Cl2	96
3	1	2	10	CH2Cl2	32
4	1a	1	10	CH2Cl2	75
5	1a	5	10	CH2Cl2	94
6	1a	2	10	MeCN	82
7	1a	2	10	THF	63
8	1a	2	10	CH3OH	76
9			10	CH2Cl2	12
10	Cu(NO3)3·3H2O	2	12	CH2Cl2	21
11	HL	2	12	CH2Cl2	16

The conditions of the reaction are solvent (2.0 mL), 1.0 mmol of TMSCN, 4-nitrobenzaldehyde (0.50 mmol), and catalyst at ambient temperature.

Calculated through the approach of GC–MS: mol(product)/mol(aldehyde) × 100.

Table 2

Outcomes for the Cyanation of Aldehydes under the Presence of 1a

entry	R1	R2	time (h)	conversion (%)a
1	Ph	H	10	89
2	2-CH3C6H4	H	10	82
3	2-CH3OC6H4	H	10	77
4	3-NO2C6H4	H	10	93
5	3-ClC6H4	H	10	91
6	1-naphth	H	10	73
7	9-nathra	H	10	32

Conversion determined by GC.

We also compared the catalyst 1a’s activities in the reaction of other substituted aromatic aldehydes with the trimethylsilylcyanide, and acquired the corresponding cyanohydrin derivatives in 32–91% yields (Table ). Aromatic aldehydes (for instance, chloro and nitro) that have the substituents of strong electron-withdrawing groups reflected the highest reactivities (items 4 and 5 in Table ), which may be associated with substrate electrophilicity increase. As expected, the aldehydes (for instance, methoxy and methyl) having electron-donating groups indicated the lower reaction yields (items 2 and 3, Table ). The yield of aromatic aldehydes containing a larger naphthalene ring is the lowest, which may be due to their larger molecular weight and inability to enter the channel of 1a.

According to the former reports along with the results of the experiment, the rational reaction mechanism was proposed and the catalyst 1a’s cyanosilylation process was described.[25−27] The unstable molecules of water in the 1a channel are removed via heating to expose the metal unsaturated center formerly. The aldehydes were activated through the Cu(II) unsaturated coordination center with the aim of reacting with TMSCN (Scheme ). The products were replaced by aldehydes, and the aldehydes were continuously activated in the next catalytic cycle.

Scheme 1

1a’s Mechanism Facilitating the Reaction of Catalysis

In accordance with the literature, a catalyst based on MOF is reusable and recyclable, and the recycling of catalysts is very significant in catalysis, especially for heterogeneous catalysis.[28−30] On this basis, the experiments of recovery catalytic were implemented, and the recoverability of 1a catalyst in the benzaldehyde cyanosilicate reaction was discussed. After the reaction, 1a catalyst can be easily separated from the solution of the reaction washed with methanol. The catalyst 1a was regenerated via heating at 100 °C for 5 h in a vacuum and employed directly in the next experiment cycle. The results of the experiment are reflected in Figure , which suggested that 1a catalyst could be reused more than four times under the optimal conditions, and the cyanoformylation catalytic activity was not significantly decreased.

Figure 3

Recovery test of benzaldehyde cyanoformylation catalyst 1a.

Compound Reduced the Viability of the OS-732 Osteosarcoma Cells

After the synthesis of compound 1a with a new structure, its anticancer activity on the OS-732 cells was evaluated by measuring the OS-732 osteosarcoma cell viability after compound 1a treatment using the CCK-8 detection kit. From the results shown in Figure A, we can see that the OS-732 cell viability was evidently inhibited after compound 1a treatment. The inhibition effect of compound 1a on the OS-732 cell survival rate showed a dose-dependent relationship. The IC50 value of compound 1a on OS-732 cell was 2.81 ± 0.17 μg/mL, this is lower than the IC50 value of the control drug (10.04 ± 0.68 μg/mL), suggesting that compound 1a had a stronger anticancer activity compared with the control drug. Cu(II) has toxicity for the organisms; in the experiment, we also detected 1a’s toxicity to HEK-293 human normal cells. The results shown in Figure B suggest that compound 1a has almost no influence on the growth of HEK-293 cells, indicating that there was no toxicity of the modified compound 1a on HEK-293 cells, which is significantly different from the high toxicity of Cu(II). The cell viability results exhibited that compound 1a had outstanding anticancer activity on OS-732 cells and low cytotoxicity on the normal cells.

Figure 4

Compound 1a reduced the osteosarcoma cell OS-732 survival rate with no cytotoxicity. CCK-8 was employed to detect the OS-732 cell viability after treatment with compound 1a at serial concentrations (1, 2, 4, 8, 10, 20, 40, 80 μg/mL) for one day. The IC50 value of 1a was calculated through the SPSS software (A). The HEK-293 cell viability after treatment with compound 1a with serial concentrations for 24 h (B).

Compound 1a Induces OS-732 Osteosarcoma Cell Apoptosis

In our previous experiment, we proved that 1a exhibited an outstanding inhibitory effect against the OS-732 cell survival rate. However, the detailed mechanism of 1a is still unclear. According to the reports, the majority of anticancer drugs exert inhibitory activity by inducing the apoptosis of OS-732 cells. Apoptosis is a vital process to keep the stability of the tissue environment and then clear harmful or unnecessary cells. Thus, determining the apoptosis of cancer cells may reflect the precession of cancer disease. After compound 1a treatment, the apoptotic level of OS-732 cells was determined via the Annexin V-FITC/PI detection kit. The data illustrated in Figure exhibits that 1a enhanced the levels of apoptotic OS-732 cell to 88.53% and 65.21% under the concentration of 3 × IC50 and 1 × IC50, respectively, which are significantly different from that of the control group. This research demonstrated that 1a could induce the apoptosis of OS-732 cells, which also explained the inhibitory effect of compound 1a on the OS-732 cells.

Figure 5

Induced percentage of OS-732 osteosarcoma apoptotic cells after compound 1a treatment. The OS-732 cells were seeded into six-well cell culture plates, and compound 1a was added for the treatment for 24 h. The Annexin V-FITC/PI detection kit was applied for the determination of the OS-732 cell apoptosis at 625 and 488/525 mm after compound 1a treatment.

Compound 1a Stimulates ROS Accumulation in OS-732 Cells

As we have proved, compound 1a could significantly induce the OS-732 cell apoptosis in a dose-dependent manner. The ROS is an important inducer of cancer cell apoptosis; thus, the ROS accumulation in the cells could be used as an indicator for the cancer cell apoptosis. In this experiment, the H2DCF-DA detection kit was utilized to determine the ROS accumulation in OS-732 osteosarcoma cells after compound 1a treatment for one day. In accordance with the results in Figure , the ROS level was evidently enhanced after the treatment of compound 1a, and this inhibition exhibited a dose-dependent relationship. Compound 1a can respectively induce the ROS positive cell levels of 91.02 and 87.22% at the concentrations of 1 × IC50 and 3 × IC50. These results indicate that compound 1a could inhibit the viability of cancer cells by inducing the accumulation of ROS in OS-732 cells.

Figure 6

Increased accumulation of ROS in OS-732 cells after treatment of compound 1a. The OS-732 cells were seeded into six-well plates, and compound 1a was added for the treatment for 24 h. ROS accumulation in OS-732 cells was measured with flow cytometry at 530 and 488 nm after compound 1a incubation for 24 h.

Conclusions

To recap briefly, we provided a novel metal–organic framework on the basis of Cu(II) containing open metal positions and one-dimensional hexagonal channels under the solvothermal conditions. The as-generated 1 was completely investigated with the thermogravimetric analyses, PXRD, the diffraction of single-crystal X-ray, the FT-IR spectra, and EA. Due to its high density of coordinately unsaturated sites along with large 1D hexagonal channels in the framework, the activated complex 1 (denoted as 1a hereafter) was researched as the without a solvent heterogeneous catalyst for cyanidation under the conditions of mind. The mechanism of catalysis in these also has been explored at length. Additionally, in the biological functional study, the CCK-8 results indicated that 1a could evidently reduce the OS-732 osteosarcoma cell survival rate in a dose-dependent fashion. Furthermore, the results of the Annexin V-FITC/PI assay revealed that 1a could obviously facilitate the apoptosis of the OS-732 osteosarcoma cells. Moreover, H2DCF-DA also demonstrated that compound 1a caused the ROS accumulation in the OS-732 osteosarcoma cells. In this present research, it was summarized that 1a showed anticancer activity against the osteosarcoma cells by inducing cancer cell apoptosis and triggering ROS production.

Experimental Section

Chemicals and Measurements

All of the reagents, materials, and raw solvents employed for the generation of the complex could be acquired from the market, and these could be employed without processing. Also, the infrared spectra could be detected via the FT-IR spectrophotometer of Nicolet Avatar 360. For the massive sample PXRD, they were determined with MiniFlex using Cu Kα (with λ of 1.5418 Å) at ambient temperature. TGA can be performed with a Q50 TGA (TA) heat analysis equipment at 5 °C/min heating rate in the flow of nitrogen. By employing a Micromeritics ASAP 2020 system, the sorption data of CO2 and N2 could be detected. The gas chromatography was implemented by applying the flame ionization detector with Agilent GC-7890A containing a capillary column (Agilent 19091J-413).

Preparation and Characterization for {[(Cu(L)2(H2O)2](H2O)8)} (1)

The mixture synthesized from 0.036 g and 0.15 mmol Cu(NO3)2·3H2O and 15 mg and 0.1 mmol 3-(1H-tetrazol-5-yl)pyridine HL was added into a mixed solution of 4 mL of DMF, 1 mL of EtOH, and 1 mL of H2O in a 25 mL glass vial; this mixture was heated for 72 h at 100 °C and then cooled to ambient temperature. After cleaning with DMF, the red square plate crystals were obtained. Yield: about 61% (based on Cu). Anal. calcd for complex 1 (C24H38Cu2N20O11): N, 30.79%; H, 4.21%; C, 31.68%; Found for the complex 1: N, 30.24%; H, 4.57%; C, 31.41%. IR (KBr, cm–1, Figure S1): 3383 m, 3105 m, 1608 s, 1562 m, 1474 m, 1433 s, 1414 s, 1351 s, 1272 w, 1215 s, 1169 m, 1154 m, 1092 m, 1051 w, 1026 s, 953 m, 872 m, 835 m, 811 s, 769 s, 691 m, 652 m, 635 m.

The X-ray data can be gained via employing an Oxford Xcalibur E diffractometer. To analyze the strength data, software CrysAlisPro was applied, and this data was subsequently converted to the HKL files. The spherical harmonic function is employed for the correction of empirical absorption, which is performed in the scaling algorithm of SCALE3 ABSPACK. Also, the original structural modes can be constructed by employing the direct manner based SHELXS program; afterward, the least-squares manner based SHELXL-2014 program was applied for a modification. By utilizing the entire nonhydrogen atoms, the anisotropic parameters could be mixed with the AFIX commands. The lattice water molecules are highly disordered, which could not be well modulated through the optimization of crystal structure; hence, their densities of electrons are removed from HKL files with the operation of SQUEEZE embedded into the PLATON software to get a new set of HKL and ins files, which were used in the following structural refinements. The compound’s refinement details along with the parameters of crystallography are detailed in Table .

Table 3

Compound’s Refinement Details and the Parameters of Crystallography

empirical formula	C24H26Cu2N20O5
formula weight	801.73
temperature/K	293(2)
crystal system	trigonal
space group	R–3c
a/Å	16.2136(9)
b/Å	16.2136(9)
c/Å	47.021(3)
α/(deg)	90
β/(deg)	90
γ/(deg)	120
volume/Å3	10704.9(13)
Z	9
ρcalcg/cm3	1.119
μ/mm–1	1.521
data/restraints/parameters	2449/75/146
goodness-of-fit on F2	1.073
final R indexes [I > = 2σ (I)]	R1 = 0.0820, ωR2 = 0.2513
final R indexes [all data]	R1 = 0.0854, ωR2 = 0.2550
largest diff. peak/hole/e Å–3	1.05/–0.87
CCDC	1 971 058

CCK-8 Assay

In our investigation, the Cell Counting Kit-8 (CCK-8) detection was performed to determine the inhibitory function of 1a on the OS-732 osteosarcoma cell survival rate. This conduction was completed under the guidance of the instructions. 1a was first dissolved in the dimethyl sulfoxide (DMSO) solution at the 1000 μM concentration for stock solution preparation. Then, it was filtered and sterilized with a 0.42 μM Millipore filter. Finally, 1a solution was diluted to a variety of working concentrations with the cell culture medium. Subsequently, the OS-732 osteosarcoma cells in the logical growth phage were collected and then seeded into 96-well plates at 1 × 104 cells/well ultimate destiny. All cells were inoculated overnight in an incubator at the condition of 5% CO2 and 37 °C; then, the compound (1, 2, 4, 8, 10, 20, 40, 80 μg/mL) was added into the wells for 24 h treatment. After the treatment of the compound, the culture medium was discarded and then a fresh medium containing 10% CCK-8 reagent (Dojindo Laboratories, Kumamoto, Japan) was added into the wells. Afterward, the microplate reader (ELX808; Bio Tek, Winooski, VT) was applied for the absorbance detection of each well, and the viability of the OS-732 osteosarcoma cells was plotted based on the absorbance.

Annexin V-FITC/PI Assay

After the treatment of compound 1a, the apoptotic percentage of the OS-732 cells was determined through the flow cytometer using Annexin V-FITC/PI staining detection kit (BD Biosciences, New Jersey). This preformation was finished totally in accordance with the protocols’ guidance with minor modification. In short, the OS-732 cells in the logarithmic growth phase were collected and then planted into six-well plates at 1 × 105 cells/well ultimate destiny. After incubation at 5% CO2 and 37 °C conditions overnight, the cells were then treated with 1a for 24 h. Afterward, the OS-732 cells were harvested, cleaned, and resuspended in a 1× binding buffer. Subsequently, 5 μL of Annexin V-FITC and 5 μL of propidium iodide (PI) solution were added into the cell suspension and then inoculated in darkness for 20 min. The apoptosis of the cell was determined in the flow cytometry at 525/625 and 488 nm. All of the studies were implemented in triplicate.

H2DCF-DA (Dichlorodihydrofluorescein Diacetate) Assay

The H2DCF-DA assay was performed in the experiment to detect the ROS accumulation level in the OS-732 cells after compound 1a treatment. This investigation was implemented on the basis of the manufactures’ instructions. Briefly, the OS-732 cells in the logarithmic growth phase were collected and then planted into six-well plates with 1 × 104 cells/well ultimate destiny. Next, the DCFH-DA probe was preloaded into the cells before conducting the compound treatment. Subsequently, the cells were incubated with 1a for 24 h. Afterward, the cells were cleaned with PBS and then fluorescence levels of each group were measured at 530 and 488 nm utilizing the flow cytometry (BD VIA, New Jersey), and then they were analyzed with FlowJo7.6 software. All the investigations required three repetitions.