Design, synthesis, and biological evaluation of evodiamine-indolequinone hybrids as novel NQO1 agonists against non-small cell lung cancer

2.1 Synthesis of NQO1-targeted amine evodiamine prodrugs

The indolequinone shelf was synthesized according to the route shown in Scheme 1 (Xu et al., 2017). Briefly, intermediate 2 was obtained from 5-methoxy-2-methylindole (1) as starting material by treatment with Vilsmeier's reagent made from POCl₃; subsequently, the indole nitrogen was agonisted by passing sodium hydride and the methyl group was introduced by adding iodomethane to obtain intermediate 3; intermediate 4 was obtained by introducing nitro on the benzene ring using concentrated nitric acid in glacial acetic acid as solvent; intermediate 4 was dissolved in anhydrous ethanol and the addition of Fremy'salt oxidized the aniline structure to intermediate 6 with quinone structure; The aldehyde group was reduced using a reducing reagent NaBH₄, and finally different diacid carboxylic acids were introduced on the alcohol hydroxyl group of intermediate 7 under DMAP conditions to obtain intermediate compounds 8a-d with different carbon chain lengths having a carboxyl group at the end. The target compounds 9a-d, 10a-d, 11a-d, and 12a-d were then produced by coupling 8a-d with these four amino-EVO derivatives via condensation reactions in the presence of EDCI and DMAP (Scheme 2).

Close

Scheme 1

Synthesis of indolequinone skeleton. Reagents and conditions: (i) DMF/POCl₃. (ii) DMF/NaH/CH₃I. (iii) AcOH/HNO₃. (iv) Sn/HCl/EtOH, reflux. (v) Fremy’s salt/Acetone. (vi) NaBH₄/MeOH/THF. (vii) Anhydride/DMAP/DCM.

Export to PPT

Close

Scheme 2

Synthesis of compounds 9a, 9b, 9c, 9d, 10a, 10b, 10c, 10d, 11a, 11b, 11c, 11d, 12a, 12b, 12c, and 12d. Reagents and conditions: (i) EDCI/DMAP/DCM, rt, 6–8 h.

Export to PPT

2.2 Relative efficacies of various compounds at inhibiting cancer cell proliferation

To analyze the cytotoxicity of these targeting compounds, the antiproliferative activity of compounds 9a-9d, 10a-10d, 11a-11d, and 12a-12d containing indoquinone nuclei against human lung cancer cells A549, H460, PC9, and gefitinib-resistant cells PC9/GR was initially screened by MTT assay. The compounds also acted on human normal liver cell line LO2 for 48 h. EVO was employed as the positive control. As shown in Table 1, most compounds had an antitumor efficacy on these cancer cells with IC₅₀ ranging from 2.54 to 34.48 µM. Compared with unmodified EVO, IC₅₀ decreased significantly. In particular, compounds 11b and 12d on the A549 cell line and PC9 cell line were more effective than other compounds. On A549 cells, the IC₅₀ of 2.72 µM and 3.66 µM, respectively, even more than EVO with IC₅₀ of 19.65 µM. On PC9 cells, the IC₅₀ of 3.80 µM and 2.54 µM, respectively, even more than EVO with IC₅₀ of 15.21 µM. Further observation revealed that compounds 11a-11d which are derivatives of 3-NH-EVO, compounds 12a-12d which are derivatives of 3-NH-10-OCH₃-EVO, exhibit better activity. These two series of EVO derivatives, when conjugated with indolequinone, demonstrate superior antitumor activity compared to that of the conjugation of indolequinone with derivatives of 2-NH-EVO and 2-NH-10-OCH₃-EVO. Previously, Dong et al. (2012) conducted structural modifications on EVO and found that derivatives substituted with methoxy or chlorine at the C-2 position exhibited weaker anticancer activity, indicating that the introduction of substituents at the C-2 position led to a loss of activity. Substitution at the C-3 position had the most significant impact on anticancer activity, with substitutions such as –F, –Cl, –OH, –NH, and –NHR enhancing the anticancer activity of the compounds. Subsequently, the research group reported a series of novel EVO derivatives substituted with different groups or modified scaffolds.

We then calculated the selectivity index (SI) of the compounds between normal and tumor cells (Table 1). The SI, defined as the ratio of the IC₅₀ value in normal liver cells LO2 to that in A549 cells, was 0.16 for EVO, and the SI value of compounds 11b and 12d were increased to 3.07 (19.2-fold) and 1.80 (11.3-fold), respectively. Compounds 11b and 12d exhibited weak cytotoxicity on LO2 cells, and was shown that these compounds kill cancer cells selectively over normal cells.

2.3 Bioreductive agonism of compounds 11b and 12d by NQO1

To verify that the compounds can target NQO1 to trigger anticancer activity, we selected the typical competitive inhibitor of the NQO1 enzyme, dicumarol (DIC), which inhibits its catalytic efficiency by interacting with the NAD(P)H binding site on the NQO1 enzyme (Lewis et al., 2017). In our literature research, we found that A549 cells exhibited a high expression of NQO1 protein (Xu et al., 2017), so we chose A549 cells to validate NQO1 targeting.

First, we verified the toxicity of DIC on A549 cells by exposing them to DIC for 48 h. The results are shown in Fig. 4A. The IC₅₀ value of DIC on tumor cells after 48 h of incubation was 140.5 ± 0.15 µM, indicating that the effect of pre-incubating A549 cells with 10 µM DIC for 1 h on the proliferation of tumor cells was small. Therefore, A549 cells were co-incubated with 10 µM of DIC followed by the addition of compound 11b and compound 12d. A549 cells incubated without DIC were used as a control. The inhibition was measured at 24 h, 48 h, and 72 h (Fig. 4B). IC₅₀ values are shown in Table 2. Without DIC, the IC₅₀ of compound 11b at 24 h, 48 h, and 72 h was 4.17 µM, 1.86 µM, and 1.25 µM, respectively. The IC₅₀ of compound 12d at 24 h, 48 h, and 72 h was 5.52 µM, 3.00 µM, and 1.72 µM, respectively. At the same time, the IC₅₀ of compound 11b at 24 h, 48 h, and 72 h was 7.65 µM, 3.95 µM, and 1.84 µM after pretreatment with the NQO1 enzyme competitive inhibitor DIC for 1 h and then addition of compound 11b. The antiproliferative activity of compound 11b against A549 decreased by 1.83, 2.12, and 1.47-fold, respectively (Table 2). The IC₅₀ of compound 12d was 8.77 µM, 4.80 µM, and 2.62 µM at 24 h, 48 h, and 72 h, respectively, and the antiproliferative activity of compound 12d against A549 decreased by 1.59, 1.60, and 1.52-fold, respectively. This indicates that the antiproliferative activity decreased after pretreating A549 cells with enzyme inhibitors followed by the action of compounds 11b and 12d, suggesting that the antitumor proliferative activity of compounds was closely related to NQO1.

Close

Fig. 4

(A) Antiproliferative activity of dicumarol acting on A549 cells. (B) Evaluation of NQO1-dependent cytotoxicity of compounds 11b and 12d.

Export to PPT

2.4 Effect of NQO1 on the stability of compound 11b

We used HPLC to assess the stability of compound 11b in the presence and absence of NQO1 in a pH 7.4 buffer (Fig. 5). The results showed that the decomposition of compound 11b is time-dependent. In the presence of NQO1, approximately 39 % of compound 11b decomposed within 24 h. In the absence of NQO1, about 83 % of compound 11b broke down within 24 h in the pH 7.4 buffer. In conclusion, compound 11b is a good substrate for NQO1. We detected the release process by LC-MS method (Fig. 6). After co-incubation with NQO1 enzyme and NADPH, the potential leaving group at position 3 of indole quinone nucleus underwent double electron reduction and split into two fragments ([M]⁺ m/z = 218.2) and ([M]⁺ m/z = 431.5). These findings suggest that NQO1 plays a role in promoting the release of compound 11b.

Close

Fig. 5

The stability of 11b in the presence and absence of NQO1 was determined by HPLC.

Export to PPT

Close

Fig. 6

LC-MS analysis of 11b after 30 min co-incubation with NQO1.

Export to PPT

2.5 Effect of compounds 11b and 12d on apoptosis

To better observe whether the inhibition of tumor cells by compounds 11b and 12d in MTT assay was related to the ability to induce apoptosis, we used Annexin V-FITC/PI kit to detect the effect of compounds on apoptosis. The assay results are shown in Fig. 7. Compound 11b and 12d were able to show concentration-dependent induction of apoptosis in A549 cells, and the total apoptosis rates were 2.18 %, 5.00 %, 14.30 %, and 27.86 % after compound 11b treated A549 cells using 0, 2, 4, and 8 µM, respectively. Compound 12d was assayed at the same concentration gradient, and compound 12d treated A549 cells using 2, 4, and 8 µM treatment of A549 cells resulted in total apoptosis rates of 5.98, 11.49 % and 34.54 %, respectively, indicating that compounds 11b and 12d could effectively induce apoptosis in A549 cells.

Close

Fig. 7

Compounds 11b and 12d induced apoptosis. A549 cells were incubated with varying concentrations of 11b and 12d (0, 2, 4, and 8 µM) for 48 h. At least three independent experiments were done for each condition.

Export to PPT

2.6 Effects of compounds 11b and 12d on the cell cycle

Most antitumor drugs inhibit cell proliferation by inducing cell cycle arrest (Yang et al., 2015). To investigate the effect of compounds on the cycle of A549 cells during proliferation, we used propidium iodide (PI) staining to detect the effect of 11b and 12d on the cell cycle. A549 cells were incubated at concentrations of 0, 2, 4, and 8 µM of 11b and 12d for 48 h. The cells were stained with PI alone and the cell cycle distribution was detected using flow cytometry. The results are shown in Fig. 8. Compound 11b caused an increase in the percentage of G2/M phase cells from 12.52 % to 31.52 %. Compound 12d caused an increase in the percentage of cells in the G2/M phase from 12.52 % to 24.65 %. EVO also increased the proportion of cells in the G2/M phase to 21.00 %, which is consistent with previously reported (Hong et al., 2014). And the above results suggest that these compounds may inhibit cancer cell proliferation in the G2/M phase through cell cycle arrest. At the same time, the percentage of G1 phase cells decreased, and the percentage of S phase cells remained basically unchanged.

Close

Fig. 8

Compounds 11b and 12d induced G2/M arrest. A549 cells were incubated with varying concentrations of 11b and 12d (0, 2, 4, and 8 µM) for 48 h. At least three independent experiments were done for each condition.

Export to PPT

2.7 Compounds 11b and 12d caused ROS burst in A549 cells

At lower concentrations, reactive oxygen species (ROS) function as signaling molecules, stimulating and hastening tumor proliferation and progression (Srinivas et al., 2019; Moloney and Cotter, 2018). Certain chemotherapeutic agents intensify oxidative stress and exhibit selective cytotoxicity against tumor cells (Maity et al., 2018). We used a 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) fluorescence probe to detect ROS levels in cancer cells subjected to compounds 11b and 12d. Fig. 9 shows that A549 cells were treated at 2 µM, 4 µM, and 8 µM for 48 h, respectively, and the ROS ratio varied with the concentration. The ROS response rates of compound 11b in A549 cells at 2 µM, 4 µM, and 8 µM were 65.1 %, 87.9 % and 93.4 %, respectively. The ROS response rates of compound 12d in A549 cells at 2 µM, 4 µM, and 8 µM were 54.2 %, 62.7 % and 95.1 %, respectively. The ROS response rate in A549 cells was 88.4 % under 8 µM EVO treatment. The above results indicate that compounds 11b and 12d can both stimulate ROS generation in A549 cells, with the proportion increasing in a concentration-dependent manner.

Close

Fig. 9

Compounds 11b and 12d caused ROS burst. A549 cells were incubated with varying concentrations of 11b and 12d (0, 2, 4, and 8 µM) for 48 h. At least three independent experiments were done for each condition.

Export to PPT

2.8 Western blot

To comprehend the mechanism of cell apoptosis induced by compounds 11b and 12d, western blot analysis was performed on related proteins. The experimental results (Fig. 10, S1) indicate that compounds 11b and 12d, after 24 h of treatment in A549 cells, downregulate the expression of Keap1 while upregulating the expression of Nrf2, NQO1, and HO-1. This suggests that compounds 11b and 12d increase cellular antioxidant capacity by agonizing the Keap1/Nrf2/NQO1 antioxidant pathway. Nrf2 is a transcription factor that regulates the expression of a series of antioxidant genes, including NQO1 and HO-1 (Younis, 2022). Keap1 is a negative regulator protein of Nrf2, which typically inhibits Nrf2 activity (Guo et al., 2015). Therefore, agonism of the Keap1/Nrf2/NQO1 pathway may be due to the compounds promoting oxidative stress within cells. Additionally, the upregulation of proteins P53, Bax, Caspase3, and Caspase9, along with the downregulation of Bcl2, suggests that the compounds induce apoptosis by agonizing g the apoptotic pathway in cells. Proteins Cyclin A2, CDK2, and CDK4 are all downregulated, and they participate in cell cycle regulation, particularly in the transition from the G1/S to G2/M phase (Oakes et al., 2014; Li et al., 2015). Thus, the downregulation of these proteins indicates that the compounds inhibit cell proliferation by suppressing the cell cycle regulation pathway.

Close

Fig. 10

A549 cells were separately incubated with compound 11b (0, 2, 4, 8 µM), compound 12d (0, 2, 4, 8 µM) and EVO (8 µM) for 24 h, respectively. The expressions of Keap1, Nrf2, NQO1, HO-1, P53, Bax, Bcl2, Caspase3, Caspase9, Cyclin A2, CDK2 and CDK4 were determined respectively by western blotting assay. GAPDH was used as an internal loading control. At least three independent experiments were done for each condition.

Export to PPT

ROS plays important signaling and regulatory roles within cells, excessive ROS can lead to oxidative stress and cell damage (Chio and Tuveson, 2017; Sies et al., 2017). The Western blot experiment results indicate upregulation of these antioxidant genes (Nrf2, NQO1, HO-1). It has been demonstrated that activation of NQO1 can cause DNA damage and cell death, which is associated with accumulation of reactive oxygen species (ROS) in the cells. The increase of NQO1 can further promote the production of ROS (Al-Gubory et al., 2010). This dual action may be one of the important mechanisms of drug pharmacology, but further research is needed to determine the specific cellular responses and biological effects.

2.9 Binding mode of the compounds to NQO1 protein

To gain insight into the possible interaction patterns of the synthesized compounds with indoquinone structures with NQO1 protein (PDB ID: 2F1O), molecular simulations of compounds 11b, 12d and EVO were performed. The docking results showed (Fig. 11) that the carbonyl oxygen in the amide bond of compound 11b forms two hydrogen bonds with GLY150 and TYR155, while another carbonyl group on the linker forms a hydrogen bond with PHE106. The oxygen atom near the indoloquinone moiety forms a hydrogen bond with TRP105, and one carbonyl oxygen on the indoloquinone ring forms a hydrogen bond with GLU117 (Fig. 11A), and the docking score was −11.256 kcal·mol⁻¹. In compound 12d, the carbonyl oxygen in the ester bond on the linker forms two hydrogen bonds with GLY150 and TYR155, and the oxygen atom in the single bond on the linker forms a hydrogen bond with HIS161. Additionally, one carbonyl oxygen on the indoloquinone ring forms a hydrogen bond with TYR128 (Fig. 11B), and the docking score was −11.930 kcal·mol⁻¹. For EVO, one hydrogen bond is formed between the side-chain NH on the indole ring and GLN104, and π-π interactions occur between the EVO skeleton and PRO68 (Fig. 11C), and the docking score was −8.070 kcal·mol⁻¹. The Docking Score can be used to evaluate the binding of small molecule ligands to receptors, the results of the calculations are shown in Table 3. The docking study was a beneficial complement and explanation to the mode of intermolecular action. The results of this study seem very promising, as the two new compounds showed important interactions with different residues and can also serve as a basis for further studies on the design of drug candidates against cancer.

Close

Fig. 11

Binding models of NQO1 with compound 11b (A), 12d (B) and EVO (C), respectively. The hydrogen bonds are indicated with red dashed lines. The carbons of compounds 11b, 12d and EVO are colored in orange. The oxygen atoms are colored in red, nitrogen atoms in dark blue, sulfur atoms in yellow and hydrogen atom is grayish-white. The figure was generated using Pymol. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Export to PPT

2.10 In vivo antitumor activity of compound 11b

In the study of the anti-proliferative activity of tumor cells, compound 11b was found to have the lowest IC₅₀ value against A549 cells (2.72 ± 0.12 µM), and is a potential antitumor agent, based on this, the antitumor activity of compound 11b in vivo was evaluated. A preliminary efficacy experiment on the LLC-xenograft mode was prepared. Using 5-fluorouracil (5-Fu) as a positive control, EVO and compound 11b were injected intraperitoneally for 13 days (20 mg/kg, once a day), as exhibited in Fig. 12A. The changes in tumor volume and tumor weight comparison among tumor-bearing mice in each group are shown in Fig. 12B and 12D, respectively. The ex vivo tumor tissue of the mouse model is depicted in Fig. 12E. From the data in the figures, it can be observed that with the increase in time, the tumor volume of mice in each group exhibited a growing trend. Although the tumor volume of mice treated with compound 11b and 5-Fu also showed a growing trend, the rate of tumor volume growth was significantly slower compared to the control group. At the end of the treatment, the tumor volume of mice in the compound 11b and 5-Fu groups was smaller than that of the control group and the EVO group. Compared to EVO, EVO modified with an indolequinone structure demonstrated the best inhibitory effect. It was remarkable that compounds 11b and 5-Fu had no significant difference in antitumor activity under the same administration route, dose, and time, and the tumor growth inhibition rate (TGI) was 39.13 and 27.54 %, respectively (Table 4).

Close

Fig. 12

Compound 11b inhibited tumor growth. C57BL/6 mice were divided into four groups (n = 6) including the control, 11b, 5-Fu, and EVO. Each group except the control group was given 20 mg/kg once a day for 13 days (i.p.). (A) Experimental scheme. (B) Plot of tumor volumes over time plotted as mean ± SEM. (C) Diagram of mice weight over time, shown as mean ± SD. (D, E) Weight and size of tumors on the last day. (F) Liver index in mice. (G) Kidney index in mice. (H) IHC staining of Bax, Bcl2, Caspase3. (I) Pathological sections of liver, kidney, and tumor tissue.

Export to PPT

From the changes in mouse body weight (Fig. 12C), it can be seen that the body weight of each group slightly increased but without significant differences. Calculating the liver-kidney index (Fig. 12F, G) showed that compared to the control, the liver-kidney index of the EVO group was slightly larger, indicating that administration of the drug EVO could cause liver and kidney damage in mice. Compared to the control, EVO group, and positive drug 5-Fu group, compound 11b caused a decrease in the liver-kidney index, indicating that compound 11b induced less liver and kidney damage in mice, with no significant toxicity.

By slicing, staining, and observing the pathological status of ex vivo tissues, the preliminary assessment of the toxicity of synthesized compounds on normal tissues of model animals and their effects on tumor tissues was made (Fig. 12I). Observation of liver tissue slices showed that compared to the control group, the liver tissues in the 5-FU and EVO groups exhibited relaxation, while the liver tissue of compound 11b did not show significant differences from the control group. Observation of kidney tissue slices revealed that the interstitial space of renal glomeruli in the 5-FU and EVO groups tended to increase compared to the control group, while compound 11b showed little change compared to the control group. Observation of tumor slices showed that the tumor tissues in the control and EVO groups had larger nuclei, disordered cell arrangement, and darker staining, while the tumor tissues affected by compound 11b appeared loose, with a disproportionate nucleus-cytoplasm ratio and vacuolation. It can be concluded that compound 11b can inhibit tumor growth and cause minimal liver and kidney damage.

Immunohistochemistry was performed on ex vivo tumor tissues from mice in each group, and the staining results are shown in Fig. 12H. The pro-apoptotic protein Bax is a cytoplasmic protein. When cells are exposed to apoptotic stimuli, Bax enters the mitochondria through a specific translocation pathway, promoting the release of cytochrome C, further agonizing Caspase3, and initiating cell apoptosis (Légiot et al., 2019; Jürgensmeier et al., 1998). Therefore, Caspase3 plays a crucial role in cell apoptosis (Wang et al., 2017). In Fig. 12H, it can be observed that the compound 11b group exhibits significantly brown staining for Bax and Caspase3 compared to the blank and EVO groups, and is stronger than the 5-Fu group. Regarding the expression of the anti-apoptotic protein Bcl2, compound 11b shows weaker brown staining compared to the blank and EVO groups. Therefore, it can be concluded that compound 11b inhibits tumor cell proliferation by promoting the expression of Bax and Caspase3 while suppressing the expression of Bcl2.

3.1 Chemistry

Most chemicals and solvents are purchased commercially and further purified and dried. The solvent was dried using a 4 Å molecular sieve. ¹H and ¹³C NMR raw data were collected by a Bruker AVANCE400 and AVANCE500 spectrometer and spectrums were exported through MestReNova software. Deuterated solvents included DMSO‑d₆, CDCl_3, and D₂O, and tetramethylsilane was used as an internal standard. The chemical shifts (δ) are reported in parts per million (ppm) and the coupling constant (J) unit is Hz. ESI mass spectra were performed on an Agilent 7250&JEOL-JMS-T100LP AccuTOF. TLC analysis was carried out on silica gel plates GF254 (Qingdao Haiyang Chemical, China). Silica gel column chromatography was performed with Silica gel 60 G (Qingdao Haiyang Chemical, China). The synthetic route of the indoloquinone structure started from 5-methyl-2-methoxyindole and was obtained by a series of substitution and oxygen reduction reactions (Zhang et al., 2005). The purity of all tested compounds was ≥95 % by HPLC analysis (S2).

3.2 General experimental procedure for the synthesis of target compounds

3.3 Cell viability assays

A549, H460, PC9, PC9/GR, and LO2 cell lines were acquired from Procell Life Science & Technology Co. Ltd. (Wuhan, China). The A549, H460, PC9, and LO2 cells were maintained in Roswell Park Memorial Institute (RPMI)-1640 medium. The PC9/GR cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing 10 % (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, and 100 mg/mL streptomycin. For the viability assays, 4000 cells were seeded in each well of a 96-well plate. The cells were allowed to attach for 12–24 h, subjected to the indicated inhibitors for 48 h, and incubated with 0.5 mg/mL MTT for 4 h. Cell viability was then determined from OD₄₉₂. All experiments were performed in triplicate.

3.4 Effect of NQO1 enzyme on the stability of compound 11b

A stock solution of 11b (10 mM) was prepared in DMSO and diluted using a PBS solution. Compound 11b (10 mM) was added to PBS (10 mM; pH = 7.4) buffer solution configured to 50 μM, and NADPH (100 μM) and NQO1 (10 mg/mL) were added. The group without the NQO1 enzyme was incubated at 37 °C as a control. Samples were taken at 0, 2, 6, 12, and 24 h; incubation was terminated by adding 100 µL of cold acetonitrile. HPLC was performed on Shimadzu liquid chromatograph LC-2030 Plus equipment. Mass spectrometry was performed on Shimadzu LCMS-2020. The mobile phase was acetonitrile: the linear gradient of water was 40:60 (v/v), the flow rate was 0.5 mL/min, and the detection wavelength was 265 nm.

3.5 Cell apoptosis analysis

A549 cells (2.5 × 10⁵/well) were seeded in six-well plates for 12–24 h. After adhesion, the cells were subjected to 2 µM, 4 µM, and 8 µM of compounds 11b and 12d for 48 h, harvested by trypsinization, and washed twice with cold phosphate-buffered saline (PBS). The cells were then centrifuged and their supernatants were removed. The washed cells were resuspended in 1 × binding buffer (500 µL) (Beyotime), subjected to Annexin V-FITC (5 µL), and incubated at room temperature for 5 min. The cells were then subjected to propidium iodide (PI; 10 µL) and incubated in the dark at room temperature for 5 min. The stained cells were analyzed by flow cytometry (NovoCyte; Agilent Technologies, Santa Clara, CA, USA).

3.6 Cell cycle analysis

A549 cells (2.5 × 10⁵/well) were seeded in six-well plates for 12–24 h. After adhesion, the cells were subjected to 2 µM, 4 µM, and 8 µM of compounds 11b and 12d for 48 h, harvested by trypsinization, and washed with cold PBS. The cells were then centrifuged and their supernatants were removed. Then 1 mL ice was used to pre-cool 70 % (v/v) ethanol and the latter was used to fix the cells at 4 °C for >2h. Staining buffer (0.5 mL) (Beyotime), propidium iodide (PI) staining solution (20×; 25 µL), and RNase A (50X; 10 µL) were added to the cells and they were incubated in the dark at room temperature for 30 min. The stained cells were analyzed by flow cytometry. (NovoCyte; Agilent Technologies, Santa Clara, CA, USA).

3.7 ROS burst assay

A549 cells (3 × 10⁵/well) were inoculated into six-well plates and attached overnight before treatment with diverse concentrations of specified compounds for 48 h. Then, the cell culture medium was discarded and cells were dyed with 10 µM DCFH-DA (Beyotime) dye for another 30 min at 37 °C. Next, labeled cells were cleaned using PBS three times. Eventually, the cells were resuspended with 500 µL of PBS before using flow cytometry. (BD Accuri C6).

3.8 Western blotting assay

A549 cells were seeded in six-well plates at a density of 2 × 10⁵ cells/well. Then the cells were treated with compounds in different concentrations and cultured at 37 °C for 24 h. Proteins were extracted by lysis buffer and stored at −20 °C. The protein concentrations were quantified by the BCA Protein Concentration Detection Kit. Then the proteins were separated by 10 % sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto polyvinylidene difluoride (PVDF). The membranes were blocked with 5 % non-fat milk in TBST (Tris buffered saline with 0.1 % Tween-20) for 2 h and incubated with primary antibodies at 4 °C overnight under gentle shaking. Then, the membranes were washed with TBST and further incubated with the secondary antibodies at 37 °C for 1.5 h. All membranes were washed with TBST three times for 30 min and protein blots were detected with chemiluminescence reagent (Thermo Fischer Scientifics Ltd.) and Tanon automated chemiluminescence imaging analysis system.

3.9 Molecular docking

The crystal structure of NQO1 (PDB: 2F1O) was obtained from a protein database. The protein preparation tool in Maestro ver. 11.5 was used for docking. Ligands and water were removed from the structure and hydrogens were added to it. Staged minimization was performed using the default setting. All docking studies were conducted in Maestro ver. 11.5. The image representing the best pose was prepared with PyMol (https://www.pymol.org/).

3.10 In vivo anti-tumor evaluation

Six-week-old male C57 mice were purchased from Chengdu Dashuo Experimental Animal Co. LTD. Adaptive feeding 1 week later, a total of 1 × 10⁶ LCC cells were subcutaneously inoculated into the right flank of C57 mice according to protocols of tumor transplant research, to initiate tumor growth. After the subcutaneous transplantation tumor formation, the transplantation tumor model C57 mice were randomly grouped into 6 mice per group. The groups treated with 11b were administered in a vehicle of 10 % DMSO/2% Tween 80/88 % saline every day by intraperitoneal injection, respectively. The positive control groups were treated with EVO every day and 5-Fu every day by intraperitoneal injection, respectively. The negative control group received a vehicle of 10 % DMSO/2% Tween 80/88 % saline through intraperitoneal injection. The mice were sacrificed after the treatments for a total of 13 consecutive days and the tumors were excised and weighed. The inhibition rate was calculated as follows: Tumor inhibitory ratio (%) = (1-average tumor weight of treated group/average tumor weight of control group) × 100 %. After execution, liver tissue, kidney tissue, and tumor tissue specimens were taken, fixed by 4 % paraformaldehyde, dehydrated in ethanol, embedded in paraffin, stained with HE, and observed microscopically. Paraffin sections of tumor tissues were eluted, incubated with antibodies, colored with DAB, restained with hematoxylin, observed on the immunohistochemical scanner, and saved with the corresponding results.

3.11 Statistical analysis

Data are means ± standard deviation (SD) for three independent experiments. T-test was used to identify statistically significant differences among cells exposed to the various target compounds and control drugs. GraphPad Prism ver. 8.0.2 (GraphPad Software, Inc., La Jolla, CA, USA) was utilized for all statistical analyses and graph plotting. Differences were considered statistically significant at p < 0.05.