Indirubin suppresses 4T1 murine breast cancer in vitro and in vivo by induction of ferroptosis and upregulation of Ptgs2

Indirubin, isolated from Indigo Naturals, is reported to have the inhibitory activity of MCF-7 human breast cancer cells in vitro. However, studies on its anti-breast cancer activity in vivo and underlying mechanism are insucient. We explored whether indirubin could trigger ferroptosis of breast cancer cells to exert antitumor activity. Bioinformatical analysis was performed to detected the expression of prostaglandin-endoperoxide synthase 2 (Ptgs2) in breast cancer tissues Ptgs2-related prognosis for patients with breast cancer. Growth of 4T1 cells was assessed using wound healing assay and MTT assay. The levels of 4-HNE, GPX4, PTGS2 and GSK-3β proteins were detected by Western blot, and the mRNA of Ptgs2 was tested by qPCR. The GSH and MDA were determined by commercial kits. Molecular docking was employed to study interaction between indirubin and GSK-3β. An 4T1 murine breast cancer was adopted to evaluate the in vivo antitumor activity of indirubin.


Results
Indirubin promoted ferroptosis of 4T1 breast cancer cells with deplete of GSH, increased MDA and 4-HNE level, as well as decreased GPX4 expression. Indirubin suppressed the growth of 4T1 breast tumor in vivo. Mechanism study showed indirubin up regulated Ptgs2 expression by promoting phosphorylation (Ser 9) of GSK-3β.

Conclusions
Indirubin suppresses 4T1 murine breast cancer in vitro and in vivo by induction of ferroptosis and upregulation of Ptgs2.

Background
Breast cancer is the most common cancer among women with increasing incidence year by year.
Furthermore, it is also one of the leading causes of death among women with cancer worldwide [1][2][3].
Chemotherapy is one of the important methods for the treatment of breast cancer, but the drug resistance and side effects make it imperative to nd new breast cancer treatment drugs [4][5][6].
Indigo Naturalis, the classical Chinese medicine, has been used in China for many centuries, and modern pharmacological studies have shown that it has anti-in ammatory, antiviral, antibacterial, anti-tumor and anti-psoriatic activities [7][8][9][10][11]. In the study of the pharmacological active components of Indigo Naturalis, indirubin has received the most attention [12]. Indirubin, the major active component of Indigo Naturalis, have been widely known for its clinical use for treatment of chronic myelocytic leukemia (CML) [13,14]. Recent studies have shown that indirubin can also be used to treat other types of cancer, like breast cancer [15][16][17]. However, the research on the mechanism of indirubin in the treatment of breast cancer is still insu cient.
Ferroptosis, a new type of cell death, is mediated by iron-dependent accumulation of lipid peroxidation. It was found that ferroptosis can be involved in the occurrence, progression and treatment of tumor. And emerging evidence indicates triggering ferroptosis of cancer cells is an important means for cancer therapy [18][19][20][21].
In this study, we aimed to explore whether indirubin could play an anti-breast cancer role by inducing the occurrence of ferroptosis. We analyzed the clinical data provided by NCBI database and found that breast cancer was related to the low level of prostaglandin-endoperoxide synthase 2 (Ptgs2), a molecular marker of ferroptosis. In vivo and in vitro studies suggested that indirubin could exert the anti-breast cancer effect by up-regulating PTGS2 and inducing ferroptosis.

Bioinformatical analysis of breast cancer-related genes
The gene expression pro le of GSE20713, GSE42568 and GSE 54002 in breast cancer and normal breast tissues were obtained from the free public database, NCBI-GEO database. Differentially expressed genes (DEGs) between breast cancer specimen and normal breast specimen were identi ed via GEO2R online tools with log FC < -2 and adjust P value < 0.05 [22]. Then, the raw data in TXT format were checked in Venn software online to detect the commonly DEGs among the three datasets. The prognostic relationships between selected genes and breast cancer were analyzed by Kaplan-Meier plotter and GEPIA online tools [23,24].
Cell viability analysis 4T1 cells were seeded in a 96-well plate and cultured in DMEM supplemented with 10% FBS overnight. After 24 h of drug treatment, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT) solution was added into each well and incubated at 37 °C for 4 h. The medium was removed and formazan crystals were later dissolved in dimethyl sulfoxide (DMSO). The absorbance was measured at 570 nm.
Wound healing assay 4T1 cells were seeded into 12-well plates and grown to a monolayer until 90% con uence. After removing the medium, the cell monolayer was scraped in a straight line with a sterile tip of 200 microliters. Next, the isolated cells were washed with phosphate buffer saline (PBS). The cells were treated with fresh medium containing 2% FBS with different drug concentrations for 48 h. Untreated cells served as the control. All the cells were photographed with a microscope before and after the drug treatment. Image J software was used to quantitatively calculate the wound area.
Animal experiment BALB/c mice (6-8 weeks old, female) were purchased from Guangdong Medical Laboratory Animal Center (Guangzhou, China) with permission No. SCXK 2017 − 0174. Animal experiments were approved by the Animal Care and Use Committee of Jinan University and conducted in accordance with National Institute of Health's Guide for the Care and Use of Laboratory Animals (7th edition, United States). All mice were kept in a non-pathogenic animal chamber with a temperature of 23 ± 1 °C and a dark period of 12 h, and fed with standard laboratory diet and water. The animals were allowed to acclimatize for a week before the experiment. Mice received a subcutaneous injection with 4T1 cells (2 × 10 6 cells) into the right ank. When the tumor diameter reached about 5 mm, the mice were randomly divided into 5 groups.
The mice were treated with saline, indigo (20 mg/kg), indirubin (low dose, 10 mg/kg; high dose, 20 mg/kg) and Adriamycin (1 mg/kg, clinically used in the treatment of breast cancer, as a positive control drug) every day (Supplementary Table 1). The tumor sizes were measured every 2 day using calipers, and the tumor volume was calculated by the following formula: volume (mm 3 The tumor size at 21th day post-injection was used as the endpoint reading. Mice were sacri ced with diethyl ether anesthesia. The data were presented as the mean volume X ± S. E.

Histological analysis
Breast tumors were xed with 4% paraformaldehyde for 48 h at room temperature. Then, they were dehydrated, transparent, para n-embedded and cut into 5 µm thick slices. Hematoxylin and eosin (H&E) were used for staining of tumor tissues. The histological changes were observed and imaged at 40x magni cation under an automatic scanning microscope (PreciPoint M8, Freising, Germany).

Western blotting
The cells and tumor tissues were lysed using RIPA (P0013C, Beyotime) lysis buffer containing 1 mM PMSF (P1005, Beyotime) on ice for 30 min. After centrifugation at 12,000 × g at 4 °C for 10 min, the supernatants were collected and total protein concentrations were determined using BCA protein assay kit (Pierce, Rockford, IL, USA). The protein samples were separated by 10% SDS-PAGE and then transferred to PVDF membranes (Millipore Corporation, Billerica, MA, USA). The membranes were blocked with 5% skim milk dissolved in TBST buffer at room temperature for 1 h and probed with the indicated primary antibodies at 4 °C overnight and then incubated with HRP-labeled secondary antibodies at room temperature for 2 h. Target proteins were detected using ECL Western Blotting Detection Reagent The cells and tumor tissues were lysed using RIPA lysis buffer containing 1 mM PMSF on ice for 30 min and then centrifuged at 12,000 × g at 4 °C for 10 min. Then, the resulting cell lysates were utilized to assess GSH content and MDA content, using commercially test kits (GSH, A061-1-1, Nanjing Jiancheng BioTechnology, Nanjing, China; MDA, S0131, Beyotime), respectively, according to the manufacturer's protocols.

Molecular docking
The molecular docking method was used for the analysis of indirubin and indigo binding with GSK-3β. The discovery Studio 3.0 docking program was adopted [25]. The structure of GSK-3β was downloaded from PDB database (PDB ID: 2O5K). The preparation of protein structure included adding hydrogen atoms, removing water molecules, and assigning Charmm force eld. All parameters were set as default.

Statistical analysis
Statistical analysis was performed using GraphPad Prism 6.0 software (San Diego, CA, USA). The statistical signi cance of differences between two groups was determined with unpaired Student's t-test and multiple comparisons were analyzed with one-way ANOVA. All data presented as mean ± standard deviation. Differences were considered statistically signi cant at p < 0.05.

Low expression of Ptgs2 in breast cancer
There were 587, 343 and 564 DEGs between breast tumors and normal breast tissues from GSE20713, GSE154002 and GSE42568, respectively, via GEO2R online tools. Then, we used Venn diagram software to identify the commonly down-regulated genes in the three datasets. Results showed that a total of 16 down-regulated genes (log FC < -2, P < 0.05) was identi ed in the breast cancer tissues (Fig. 1A, B). Noticeably, Ptgs2, a ferroptosis marker, was down-regulated in breast cancer patients, which indicated that ferroptosis may be related to breast cancer [26,27]. Furthermore, the prognostic information of Ptgs2 was analyzed employing the Kaplan-Meier plotter. Result showed that the low mRNA expression of Ptgs2 (HR, 0.82; CI, 0.7-0.96) was associated with the poorer overall survival for patients with breast cancer (Fig. 1C). In addition, we also performed GEPIA and found that Ptgs2 mRNA was signi cantly lower in breast cancer tissues than that in normal tissues (p < 0.05) (Fig. 1D). Results above showed the low expression of Ptgs2 in breast cancer.

Indirubin induces ferroptosis of breast cancer cell in vitro
In order to investigate whether indirubin has the inhibitory effect on breast cancer, we rstly performed wound healing assay to study the effect of indirubin on growth of 4T1 cells in vitro. As shown in Fig. 2A and B, indirubin (40 µΜ and 80 µM) signi cantly inhibited wound closure of 4T1 cells (p < 0.05, p < 0.01). Next, we detected the effect of indirubin on viability of 4T1 cells by MTT assay. Result showed that the viability of 4T1 cells was signi cantly inhibited in a dose-dependent manner after incubation with indirubin (Fig. 2C). These results demonstrated the anti-breast cancer effect of indirubin in vitro. Then, we explored whether ferroptosis was involved in the anti-breast cancer effect of indirubin. The pre-treatment of Fer-1 (ferrostatin-1, ferroptosis-speci c inhibitor, 25 µΜ) signi cantly reversed indirubin-induced inhibition of 4T1 cells viability (p < 0.05) (Fig. 2D). However, Nec-1 (necrosis inhibitor, 50 µΜ) and z-VAD (apoptosis inhibitor, 50 µΜ) didn't protect 4T1 cells from indirubin-induced inhibition of viability (Fig. 2D). These results indicated indirubin induced ferroptosis of breast cancer cell in vitro. The occurrence of ferroptosis is characterized by the depletion of GSH and accumulation of lipid peroxides. Meanwhile, the result showed that indirubin (40 µΜ and 80 µΜ) reduced the GSH content of 4T1 cells (p < 0.05, p < 0.05) (Fig. 2E). MDA and 4-hydroxynonenal (4-HNE) are major end products derived from lipid peroxides. As shown in Fig. 2F and G, indirubin (80 µΜ) signi cantly increased the level of MDA (p < 0.05) and 4-HNE (p < 0.05). In addition, glutathione peroxidase 4 (GPX4) plays a key role in reducing lipid peroxides. The expression level of GPX4 signi cantly decreased by indirubin (40 µΜ and 80 µM) (p < 0.05, p < 0.05) (Fig. 2G). These results indicated that indirubin induced ferroptosis of breast cancer cell in vitro.

Indirubin up-regulates the expression of PTGS2 in breast cancer cell
Given that low expression of Ptgs2 in breast cancer found by bioinformatics analysis in Fig. 1, We sought to explore whether indirubin-induced ferroptosis was related to PTGS2. We rstly detected the effects of indirubin on the expression of PTGS2 by qPCR and Western blotting. As shown in Fig. 3A, indirubin (40 µΜ and 80 µΜ) signi cantly elevated the mRNA level of Ptgs2 in 4T1 cells (p < 0.01, p < 0.001). Moreover, indirubin (80 µΜ) signi cantly increased the expression level of PTGS2 protein (p < 0.05) (Fig. 3B). Indirubin was reported as the potent inhibitor of glycogen synthase kinase-3 beta (GSK-3β) [28]. We found that indirubin could form hydrogen bonds with Arg141, Gln185 and Val135 of GSK-3β protein by molecular docking (Fig. 3C, D). Interestingly, indigo, a main component from Indigo Naturalis and isomer of indirubin, only forms the hydrogen bond with Arg141of GSK-3β (Fig. 3C, D). It was indicated that two carbonyl groups in opposite positions of indirubin bene ted its binding with GSK-3β. We further analyzed the effect of indirubin on phosphorylation (Ser 9) of GSK-3β in 4T1 cells by western blotting. Indirubin but indigo could promote phosphorylation (Ser 9) of GSK-3β (Fig. 3E). Correspondingly, indigo had no effect on the mRNA and protein expression of PTGS2 (Fig. 3F, G). In addition, indigo failed to triggering lipid peroxidation and ferroptosis ( Fig. 3H and Supplementary Fig. 1). Indeed, it has been reported that GSK-3β participates in regulating the expression of PTGS2 [29,30]. Hence, we speculated that indirubin may regulate the activity of GSK-3β to promote the expression of PTGS2 and then trigger the ferroptosis of tumor cells.

Indirubin suppresses breast cancer in vivo
Indirubin has been shown in vitro to induce ferroptosis of breast cancer cells, which may concern upregulation of Ptgs2. Therefore, we next sought to investigate whether indirubin could suppress breast cancer in vivo. Firstly, we evaluated the anti-tumor activity of indirubin using the mouse model of breast cancer by injection of 4T1 cells subcutaneously. Results showed that indirubin (20 mg/kg) signi cantly inhibited the growth of breast cancer with reduced tumor weight (p < 0.05) and volume (p < 0.01) compared with the control group (Fig. 4A-C). Histopathological observation of tumor tissue showed that indirubin resulted in breast tumor necrosis (Fig. 4D). Meanwhile, the positive drug ADR suppressed breast cancer (Fig. 4A-D). After indirubin (20 mg/kg) treatment, the GSH content of tumor tissues signi cantly decreased (p < 0.05) (Fig. 4F). Indirubin (10 mg/kg and 20 mg/kg) signi cantly increased the MDA level of tumor tissues (p < 0.01, p < 0.01) (Fig. 4G). In addition, indirubin also increased the 4-HNE level of tumor tissues, and decreased the expression of GPX4 of tumor tissues (Fig. 4E). As expected, Indirubin (10 mg/kg and 20 mg/kg) signi cantly increased the level of PTGS2 protein of tumor tissues (p < 0.01, p < 0.001) (Fig. 4H). However, treatment of indigo (20 mg/kg) didn't decrease the weight and volume of breast tumors and promote necrosis and lipid peroxidation of tumors consistent with in vitro results (Fig. 4). The results showed that indirubin promoted lipid peroxidation of tumors suppressed breast cancer in vivo.

Discussion
Breast cancer is one of the leading causes of cancer death in women. Chemotherapy is one of the effective but risky ways to treat breast cancer. Naturally derived compounds have the potential to develop new drugs because of their safety and e cacy. Indirubin and derivatives has the inhibitory effect on tumors [15][16][17]. In our study, effectiveness and pharmacological mechanism of indirubin was explored, which may contribute to development of anti-breast cancer drug.
Ferroptosis is a unique way of cell death, and it has been shown that ferroptosis plays an important role in the occurrence and development of tumors [31]. By inducing ferroptosis of tumor cells, it can be used for tumor therapy [32,33]. Ptgs2, also known as cyclooxygenase-2 (Cox-2), is the key enzyme in prostaglandin biosynthesis. As a matter of fact, cyclooxygenases can catalyze lipid oxidation [34,35]. Studies have shown that Ptgs2 is involved in the process of ferroptosis, and Ptgs2 is signi cantly upregulated in the ferroptosis induced by RSL3 and erastin [36]. Meanwhile, suppression of Ptgs2 can inhibit the ferroptosis of cells and play a protective role [26,27]. In our study, results of bioinformatical analysis showed that ferroptosis-related gene Ptgs2 was expressed lower in breast cancer, which indicated an e cient strategy for treating breast cancer. It is possible to induce ferroptosis of breast cancer cells by ug-regulation of Ptgs2. Then, in vivo and in vitro analysis results showed that indirubin could play an anti-breast cancer role by upregulation of Ptgs2 to induce ferroptosis.
GSK-3β is a protein related to the regulation of Ptgs2, which can increase the expression of Ptgs2 by inhibiting GSK-3β activity [30]. The activity of GSK-3β is regulated by phosphorylation at Ser 9 and Tyr 216 in opposing directions, with phosphorylation of Ser 9 decreasing GSK-3β activity and phosphorylation of Tyr 216 increasing GSK-3β activity [37]. However, previous studies have shown that indirubin is an inhibitor of GSK-3β [28]. Our results showed that indirubin promoted phosphorylation of GSK-3β at Ser 9 and then inhibited the activity of GSK-3β. Molecular docking further proved that indirubin with two carbonyl groups at different sides could form additional hydrogen bonds with Gln185 and Val135 of GSK-3β, compared to indigo with two carbonyl groups the same side indicating the special role of indirubin in regulating GSK-3β.

Conclusion
In conclusion, indirubin suppresses breast cancer by up-regulation of Ptgs2 and induction of ferroptosis in vitro and in vivo in a 4T1 murine breast cancer model, which may be related bind of indirubin with GSK-3β and then promote its phosphorylation at Ser 9 (Fig. 5).