The breast cancer cell lines MCF-7, MDA-MB-231 and 4T1 were purchased from the American Type Culture Collection (Rockville, MD, United States). Cells were maintained in DMEM or RPMI 1640 medium supplemented with 10% foetal bovine serum (Si Ji Qin Bioengineering, China) and 1% antibiotics (penicillin and streptomycin) at 37°C in a 5% CO₂ incubator.

The cells (1-5 × 10³ cells per well) were incubated in 96-well plates overnight and then treated with different concentrations of Statmp-151 for 24, 48, and 72 h. Afterward, 5 mg/ml MTT (20 μl) was added to each well for 3 h. Finally, the supernatant was removed, and 150 µl of DMSO was added to each well. The optical density values were determined at 490 or 570 nm by a SpectraMAX M5 Microplate Spectrophotometer. All experiments were performed three times with three replicates.

Colony formation was measured by seeding cells in 6-well plates at 500–800 cells per well and treating the cells with various concentrations of Statmp-151 after approximately 24 h of incubation. The culture medium containing Statmp-151 was replaced every three days. All cells were fixed and stained with 0.5% crystal violet after 12 days.

For determination of the effect of Statmp-151 on tumour apoptosis, apoptosis detection kits were used. The cells (1 × 10⁵ cells per well) were incubated in 6-well plates overnight and then treated with different concentrations of Statmp-151. After 24 h, the cells were harvested and washed twice with cold PBS. Following the manufacturer’s instructions, the cells were stained with FITC-conjugated Annexin V and PI (Propidium Iodide) and detected by FCM.

The cells (1 × 10⁵ cells per well) were incubated in 6-well plates overnight, and treated with Statmp-151 for another 24 h. Then, the cells were harvested and washed twice with cold PBS, stained with JC-1 according to the instructions, and finally detected by FCM.

The cells (1 × 10⁵ cells per well) were incubated in 6-well plates overnight and then treated with different concentrations of Statmp-151. After 24 h, the cells were harvested and washed twice with cold PBS. The ROS levels were monitored using 10 µM DCFH-DA for 30 min and then detected by FCM.

For analysis of the expression of the corresponding proteins in MDA-MB-231 and 4T1 cells after incubation with Statmp-151 for 24 h, harvested cells were lysed with RIPA buffer for 1 h. Then, the protein concentrations were measured and equalized. SDS-PAGE with the optimal concentration of proteins selected according to their molecular weight was performed, and then, the proteins were transferred onto polyvinylidene difluoride nitrocellulose membranes. After incubation with 5% skim milk for 1 h, the target protein was incubated with the corresponding primary antibodies overnight at 4°C. The next day, the cells were washed several times and then incubated with the corresponding secondary antibodies, followed by washing several times. Finally, the protein bands were visualized using an enhanced chemiluminescence kit. Monoclonal β-actin proteins were used as a reference.

A wound-healing migration assay was performed to evaluate cell migration. When cells grew to 80% confluence in 6-well plates, they were renewed with 2% FBS containing different concentrations of Statmp-151. Images were taken at 0 and 24 h by an inverted microscope.

All animal experiments were conducted in accordance with the principles and procedures approved by the Committee on the Ethics of Animal Experiments of State Key Laboratory of Biotherapy, Sichuan University. BALB/c mice aged 6–8 weeks were purchased from the experimental Animal Centre of Sichuan University. Approximately 1 × 10⁶ 4T1 cells were inoculated into the right lower limb of the mice. We randomly divided the mice into 4 groups (n = 5) when the tumours reached an average volume of 100 mm³. Statmp-151 (10, 20, and 30 mg/kg) or vehicle was intraperitoneally injected once a day for 15 days. After administration, the body weights and tumour sizes of the mice were measured every 3 days. At the end of the animal experiment, the weight of the tissue and tumour was recorded, and then, the tumour volume was measured and photographed. The tumour tissue was stored in paraformaldehyde for further analysis.

For evaluation of the safety of Statmp-151 during the treatment, all the mice were observed continuously for general conditions, such as body weight, appetite, mental state and other clinical features. Blood was obtained from the eyeball for biochemistry analysis. Haematoxylin and eosin staining was performed of paraffin sections of lung, liver, spleen, heart and kidney tissues of the mice treated with Statmp-151.

The results are presented as the mean and standard deviation (M±SD). Statistically significant p values were labelled as follows: *p < 0.05; **p < 0.01; ***p < 0.001. All statistical workflows were performed in GraphPad Prism (version 7.0).

TMP was reported to possess multiple activities, as described above. For chemical structure modification, there are three common fragments of TMP to utilize (Figure 1A). Carboxyl, bromo and hydroxyl groups are operable reactive groups for various modification strategies. Stat3-targeted inhibitors, such as Stattic, BBI608, LLL12, HJC0146, LY-5 and STA-21, are listed in Figure 1B. Among them, Stattic was the first small molecule inhibitor of Stat3 activation and dimerization (Schust et al., 2006). This compound is often used for further development to discover Stat3 inhibitors (Ji et al., 2015). In this study, we used (5-bromo-1,1-dioxidobenzo[b]thiophen-2-yl) (piperidin-1-yl) methanone (E28) (Ji et al., 2015) as the lead compound to discover Stat3 inhibitors containing TMP. We replaced the piperazine ring of E28 with a piperazine ring for further modification (in blue circle) and explored the group linking TMP with the piperazine ring. In terms of the scaffold of Stattic, we also explored the substituents on the benzene ring (in red circle), and a total of 38 compounds were designed (Figure 1C). We employed the Glide docking program to conduct a virtual screening against these 38 rationally designed Statmp series and finally found that Statmp-151 could exhibit potent activity against Stat3, had the best docking score and showed a more favourable conformation (Figure 2). Binding mode analysis indicated that the pyrazine ring and sulfone form hydrogen bonds with Lys591, while the bromo substituent forms halogen bonds with Arg595. Then, wet laboratory experiments were conducted to explore whether Statmp-151 was consistent with the primary drug design. The synthesis and characterization of Statmp-151 are shown in the supporting information.

The effects of Statmp-151 on the proliferation of MCF-7, MDA-MB-231 and 4T1 cells were detected by MTT assays (Liu et al., 2019). In 4T1 and MDA-MB-231 cells where Statmp-151 was used at concentrations of 0–20 µM, and in MCF-7 cells where Statmp-151 was used at concentrations of 0–40 µM for 24, 48 and 72 h. The results showed that the IC₅₀ at 72 h was 2.20 ± 0.01 µM in 4T1 cells, 4.75 ± 0.39 µM in MDA-MB-231 cells, and 8.06 ± 0.08 µM in MCF-7 cells (Figure 3A). Pan J reported the anti-tumour activity of TMP which is not ideal. The ligustrazine IC₅₀ results showed it is about 10 mmol/l in breast cancer. The above results significantly improved the anti-tumour activity of maternal TMP (Pan et al., 2015).

A clone formation experiment was utilized to further verify the effects of Statmp-151 on the proliferation of MDA-MB-231, MCF-7 and 4T1 cells. As shown in Figure 3B,C, cancer cell proliferation was reduced in a concentration-dependent manner, which was consistent with the results of the MTT assay.

For further analysis of the effect of Statmp-151, we performed Annexin V and PI double dye detection of MDA-MB-231 and 4T1 cells (Zhu et al., 2016). As shown in Figure 4A,B, Statmp-151 was able to induce breast cancer cell apoptosis in a dose-dependent manner after treatment for 24 h compared to that in the control group.

To confirm the apoptotic effect of Statmp-151 on breast cancer cells, we determined the expression of Bcl-2, Bax and Cleaved caspase-3 (CC3) in MDA-MB-231 and 4T1 cells after treatment with Statmp-151 for 24 h. As shown in Figure 4C,D, the expression level of Bcl-2 was strongly decreased in 4T1 cells, Bax expression was not changed, and CC3 expression was significantly increased. This treatment showed no effect on Bcl-2/Bax ratio expression and increased CC3 expression in MDA-MB-231 cells. Collectively, these results showed that Statmp-151 could induce breast cancer cell apoptosis.

Mitochondrial membrane potential is a marker of the early apoptotic pathway. During mitochondrial pathway-mediated apoptosis, mitochondrial membrane permeability increases, stromal calcium outflow and mitochondrial membrane rupture occur, and apoptotic induction factor (AIF) and cytochrome C are released, thus initiating a cascade reaction to activate the caspase family and ultimately induce apoptosis (Reed, 1997). Mitochondrial membrane potential and intracellular ROS level jointly have a marker effect on the growth and development of cancer cells. When the mitochondrial membrane potential changes or the ROS level changes, it often indicates the loss of homeostasis of cancer cells. Once a compound has an inhibitory effect on cancer cells, these two indicators usually change (Zhao et al., 2016; Burke, 2017).

According to this knowledge, we detected the changes in mitochondrial membrane potential (ΔΨm) by FCM after treatment with different concentrations of Statmp-151 in breast cancer cells. As shown in Figure 5A,B, treatment with 10 µM Statpm-151 led to a 17.28% loss of ΔΨm for 4T1 cells. The loss was 52.26% in MDA-MB-231 cells under the same conditions.

Tumor cells are often in an environment with a high level of reactive oxygen species (ROS), and intracellular ROS levels are closely related to the stability of the mitochondrial membrane potential (Li et al., 2011). The effect of Statmp-151 on membrane potential was identified in previous experiments, and the ligustrazine structure of Statmp-151 has been reported to reduce intracellular reactive oxygen species (Zhang et al., 2010). In view of this, we also determined the effect of Statmp-151 on ROS by FCM using DCFH-DA. The results indicated that Statmp-151 led to increased ROS levels in a dose-dependent manner. As shown in Figure 5C,D, the results showed that after 24 h of treatment with 10 µM Statmp-151 in 4T1 and MDA-MB-231 cells, ROS levels were 18.89 and 48.72%, respectively. These results confirmed that Statmp-151 induced cell apoptosis via the mitochondrial-mediated apoptotic pathway.

Phosphorylation by Jak kinase is a key step in the activation of Stat3 (Li et al., 2019); therefore, we detected whether Statmp-151 could influence the expression of p-Stat3. The results confirmed that Statmp-151 could decrease the expression of p-Stat3 but had no effect on total Stat3 in MDA-MB-231 and 4T1 cells Figure 6. The results implied that Statmp-151 inhibited the phosphorylation of Stat3 proteins, consistent with the original drug design.

Tumour cell migration is one of the committed steps in cancer metastasis (Li et al., 2016). We conducted wound healing assays on 4T1 and MDA-MB-231 cells to explore the therapeutic potential of Statmp-151. As shown in Figure 7A,B, the results indicated that Statmp-151 significantly inhibited the migration of both 4T1 and MDA-MB-231 cells in a concentration-dependent manner. The migration of these cells to the wound area was significantly inhibited after incubation with Statmp-151 for 24 h at 10 µM.

To further verify the mechanism of the antimigratory effects of Statmp-151, we determined the MMP-9 and MMP-2 levels by western blotting. The results suggested that Statmp-151 could significantly inhibit the expression levels of MMP-2 and MMP-9 in 4T1 cells Figure 7C. These results suggested that Statmp-151 possessed a potent ability to inhibit breast cancer cell migration.

To evaluate the anti-tumour efficacy of Statmp-151 in vivo (Yang et al., 2015), we established a 4T1 tumour-bearing mouse model. The mice were administered Statmp-151 daily at doses of 10, 20 and 30 mg/kg for 15 days (Heppner et al., 2000; Zeng et al., 2020). There was a significant reduction in tumour growth, while the body weight of mice had no significant changes (Figure 8A,B). On the final day, there were significant reductions in tumor size and tumour weight compared with the control group (Figure 8C,D). These results suggested that Statmp-151 had potent anti-tumour activity in vivo.

As mentioned above, the body weights of the mice were not significantly changed. To further evaluate the safety of Statmp-151, we determined the serum biochemical indexes and weights of the heart, liver, spleen, lung and kidney. Paraffin sections of the lung, liver, spleen, heart and kidney were stained with haematoxylin and eosin. As shown in Figure 9A,B, there were no obvious differences in organ weight or serum biochemical indexes. Furthermore, no pathologic changes were observed in the lung, liver, spleen, heart or kidney compared with those in the control group Figure 9C. Therefore, Statmp-151 was considered to be a relatively safe small molecule compound.