MCL-1 plays an oncogenic role in breast cancer by modulating chemoresistance and stemness properties by activating Wnt/β-catenin

Background Breast cancer is the most common malignant tumor and the leading cause of death in women. Chemotherapy is one of the most important treatments for breast cancer. However, the development of chemotherapy resistance is the main cause of 20–30% of breast cancer patients developing metastasis, leading to death. MCL-1, an anti-apoptotic protein, has not been found to contribute to chemotherapy resistance in breast cancer. We used large gene panels to detect pathological sections of tumors in drug-resistant and sensitive patients. We validated protein proling by IHC in a larger cohort of samples. We performed the function of MCL-1 by knockdown and overexpression in vitro and in vivo. Luciferase assay and CHIP assay were used to prove the regulatory network between MCL-1 and LRP6. We found that MCL-1 is more highly expressed in drug-resistant breast cancer tissues than it is in sensitive breast cancer tissues. Functional studies have revealed that MCL-1 plays an important role in drug resistance by regulating apoptosis in breast cancer cells. We found that overexpression of MCL-1 enhances the chemoresistance and stemness of breast cancer cells in vitro and in vivo, while silencing has the opposite effect. Mechanistically, by downregulating and upregulating MCL-1, we show that MCL-1 regulates LRP6 and activates the WNT/β-catenin signaling pathway in breast cancer cells. Finally, we found that a high level of MCL-1 expression predicts a poor prognosis in breast cancer. Our work highlights the role of MCL-1 in chemoresistance and stemness. The MCL-1-WNT/β-catenin axis might be used as a new clinical target for breast cancer therapy. enhance the characteristics of were with HA with or The that with suggest with and increases binding between and in signaling pathway mediates stem cell renewal and drug resistance by promoting translocation to the nucleus of β-catenin [34–36]. Our data provide evidence that MCL-1 promotes cancer stem cell characteristics and drug resistance though the WNT/β-catenin signaling pathway, which offers a novel strategy for targeted therapy in breast cancer. cell in of signaling predictive value of MCL-1.


MCL-1 overexpression and inhibitor
MB231 and MCF-7 cell lines were stably transfected with a lentiviral vector (pReceiver-Lv105-pure vector) by lentiviral transduction (MOI = 3), and they were selected by treatment with 5 µg/ml puromycin (219453925, MPbio, USA). Human MCL-1 cDNA (GeneCopoeia, EX-Z9207-Lv105-B, cDNA clone) or an empty vector was seeded into the pReceiver-Lv105-pure vector and were selected by treatment 2 µg/ml puromycin. MCL-1 was analyzed by WB and qRT-PCR. The MCL-1 inhibitor (M-i) S63845 was purchased from MCE. We used it at 10 nm, which is a concentration that results in no cytotoxicity.

Cell viability assay
Cells (5000 cells/well) were seeded in 96-well plates and placed in the cell culture incubator at 37 °C overnight to allow the cells to attach, after which they were treated with taxol at concentrations of 0, 1.625, 3.125, 6.250, 12.500, 25.000, 50.000, 100.000, 200.000 nM in the presence or absence of M-i (5 nm in DMSO) for 48 h. Cell viability was determined by MTT ((3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay (Sigma-Aldrich, Saint Louis, MO, USA), and then the percentages of viable cells was measured by absorbance value (OD) (VICTOR X5, PerkinElmer, USA) at 490 nm. The experiments were repeated three times.

Mammosphere formation assay
Cells were grown in serum-free DMEM supplemented with 1X B27 and were plated in ultralow attachment plates at a density of 10,000 viable cells/mL, and then they were grown for 10 days. Mammospheres were observed under a microscope and were photographed. The experiments were repeated three times.

Colony formation assay
NC and MCL-1-overexpressing cells were seeded into a 24-well plate (5 × 10 2 cells/well) and cultured in DMEM for 48 h. Then, the xed colonies formed were stained with crystal violet. The colony formation assay was repeated three times.

Flow cytometry
CD44 and CD24 were analyzed using ow cytometry. NC and MCL-1-overexpressing cells were then suspended in ow cytometry staining buffer at a nal cell concentration of 1 × 10 6 cells/mL. Cells were incubated with the following uorescent monoclonal antibodies (eBioscienceTM, Thermo Fisher Scienti c): CD44 and CD24. Cells were washed two times with 2 mL of ow cytometry staining ow cytometry buffer and resuspended in 500 µL of ow cytometry staining buffer. Flow cytometry analyses were performed on a BD ow cytometer (Calibur, BD, USA). The experiment was performed three times.
2.9 RNA extraction and real-time PCR RNA was extracted from cells using TRIzol reagent (Takara, Beijing, China), and total RNA was reverse transcribed into cDNA. Quantitative real-time PCR was carried out using TB Green™ Premix Ex Taq™ II (Takara, Beijing, China) with a CFX96™ Real-Time System (ABI Quant Studio 7 Flex, Applied Biosystems, USA).
GAPDH was used as the control, and the relative mRNA levels were analyzed by the 2 (−△△Ct) method. The primer sequences are listed in Table 1.   cells were implanted in the inguinal mammary gland of mice. Ten days later, when the tumors reached a volume of 100 mm 3 , we randomly allocated the mice to groups in which they received taxol (5 mg/kg), MCL-1 with taxol (5 mg/kg), M-i (2.5 mg/kg) with taxol (5 mg/kg) or DMSO. Mice were euthanized 35 days after the administration.

Immunohistochemistry
Immunohistochemistry (IHC) was conducted on para n-embedded sections. Before staining, para n sections were incubated at 60 °C for 1 h for dewaxing and treatment with dimethylbenzene two times for 15 min. The tissue sections were soaked in a series of ethanol solutions with concentrations of 100%, 100%, 95%, 80% and 75% for 5 min and then soaked in distilled water for 5 min. Afterwards, the tissue sections were incubated with sodium citrate antigen retrieval solution for 10 min and in 3% H2O2 for 15 min, which was followed by 5% BSA for 40 min at 37 °C. Then, the tissue sections were incubated overnight at 4 °C with a primary anti-MCL-1 antibody (1:200 dilution in PBS, CST, Boston, USA) or an anti-CD44 antibody (1:200 dilution in PBS, CST, Boston, USA) and goat anti-mouse or anti-rabbit IgG for 30 min at room temperature. MCL-1 expression was visualized by DAB staining, and tissue sections were counterstained with hematoxylin. Finally, tissue sections were evaluated under light microscopy (Olympus BX53, Olympus, Japan).

Transcriptional reporter gene assay
Cells overexpressing MCL-1 and control cells were cultured in 96-well plates at a concentration of 5000 cells per well. Two hundred nanograms of TCF reporter plasmid, GLI reporter plasmid or RBP-JK reporter plasmid (Millipore) and 10 ng of pRL-TK (Renilla-TK-luciferase vector, Promega) were cotransfected into the cells. After 48 h, a Dual-Glo Luciferase reporter Assay System (E1910, Promega) was used to measure the luciferase activities following the manufacturer's instructions.

Coimmunoprecipitation assay and coculture immunoprecipitation assay
For the coimmunoprecipitation assay, MB-231 cell lysates transfected with Flag-tagged MCL-1 or a vector control were subjected to immunoprecipitation with Flag antibody (1:1000) or control IgG for 2 h at 4 °C. All immunoprecipitations were performed with protein A/G sepharose (Santa Cruz Biotechnology) while rotating at 4 °C overnight. The beads were collected by centrifugation at 3000 rpm and then were washed three times with lysis buffer. The immunoprecipitants were subjected to WB.
For the coculture IP assay, 293T cells were cultured in 6-well plates and individually transfected with 0.5 µg of MCL-1-Flag expression plasmids or LRP6-HA, Wnt3a-myc or Wnt10b-His plasmids (Thermo Fisher Scienti c, Shanghai) using Lipofectamine 3000. After 24 h of coculture, cell lysates with LRP6 receptors and cell lysates with Wnt3a or Wnt10b ligands were prepared at a ratio of 1:2:2 to detect the interaction of the LRP6 receptor and two ligands. The subsequent steps were the same as the IP assay mentioned above.

Statistical analysis
Data are presented as the means ± standard deviation, and GraphPad Prism (version 8, San Diego, CA, USA) was used to conduct statistical analyses. The signi cance of a difference between the groups was tested using Student's t-test for the data. A P-value < 0.05 was considered statistically signi cant.

MCL-1 is associated with resistance to taxol in breast cancer patients
Three breast cancer patients, one with resistance to chemotherapy and the others with sensitivity, were tested with Foundation One CDx assays. The copy number of MCL-1 in Taxol-resistant breast cancer patients was 32, which was much higher than that in sensitive breast cancer patients (Supplementary Excel 1). Overexpression of MCL-1 has been reported to be associated with poor prognosis in cancers [19].  . 1a). Furthermore, we examined the expression of MCL-1 in the tumors of 168 patients for neoadjuvant therapy by using IHC. IHC staining revealed that MCL-1 was remarkably elevated in neoadjuvant chemotherapy resistance (NAC-R) compared to that of neoadjuvant chemotherapy sensitivity (NAC-S) (Fig. 1b, c and d). Furthermore, to con rm that chemotherapy drug resistance could upregulate MCL-1 expression, we detected MCL-1 expression by ICH before and after neoadjuvant chemotherapy. We randomly chose 20 of the 168 patients to measure MCL-1 expression in the puncture tissue samples before neoadjuvant therapy and after. The results showed that MCL-1 expression in the tumor tissues signi cantly increased after receiving neoadjuvant therapy (Fig. 1e, f and g). Collectively, our results reveal that MCL-1 is upregulated in chemoresistant breast cancer. In addition, we also con rmed by immuno uorescent analysis that the protein level of MCL-1 was strongly upregulated in breast cancer mammospheres (Fig. 1h) derived from 2 independent patients. Similarly, western blot analysis revealed that compared to control cells, the mammospheres exhibited higher expression of Nanog, OCT4 and SOX2, which are transcription factors involved in the maintenance of stem cell properties [20,21].

Chemotherapies promote MCL-1 overexpression in breast cancer cell lines.
To investigate the effect of chemotherapy drugs on MCL-1 expression in breast cancer, we used western blotting to analyze MCL-1 expression in the multidrugresistant cell line 231/MDR and in wild-type MDA-MB-231 cells (231). The results showed that MCL-1 expression in 231/MDR cells was much higher than it was in 231 cells (Fig. 2b). Interestingly, we found that 231/MDR had a stronger ability to induce cancer stem cell colony formation (Fig. 2a) and higher expression of Nanog, OCT4 and SOX2 (Fig. 2b). Meanwhile, real-time RT-PCR and western blotting were used to analyze MCL-1 expression in MB-213 and MCF-7 cells treated with different concentrations of taxol (1 nM, 5 nM, and 10 nM), which is the most common drug in chemotherapy regimens. Real-time RT-PCR and western blot results revealed that cells treated with higher taxol concentrations had higher levels of MCL-1 mRNA and protein ( Fig. 2c and d). All the results above demonstrate MCL-1 overexpression in breast cancer cell lines after chemotherapy.
3.3 MCL-1 depresses sensitivity to taxol in breast cancer cell lines to regulate stem cell transformation.
To elucidate the mechanism of MCL-1 in taxol resistance, we overexpressed MCL-1 in two breast cancer cell lines, MDA-MB-231 (MB231) and MCF-7, using a lentivirus-based expression system. The western blot results showed signi cant overexpression of MCL-1 in cells transfected with MCL-1-expressing lentivirus compared with the control vector (Fig. 3a). To con rm the anti-apoptotic effect of MCL-1 on breast cancer cells, we conducted apoptosis analysis, cell viability assays and MTT assays. MTT assay results also con rmed that MCL-1 could inhibit taxol-induced cell death. MTT assays showed that the IC 50 values of MCL-1-overexpressing cells were dramatically increased compared with those of the control vector ( Fig. 3b and c). Similar ndings were observed in ow cytometry results, which showed that overexpression of MCL-1 had an anti-apoptotic effect in MB231 and MCF-7 cells treated with different concentrations of taxol (1, 5 or 10 nM) (Fig. 3d).
We hypothesized that MCL-1 would induce cellular drug resistance by increasing breast cancer cell stemness. To test this hypothesis, we rst conducted a mammosphere formation assay to investigate the relationship between MCL-1 expression and mammosphere formation ability. The results show that MCL-1 overexpression dramatically increased mammosphere formation by MB231 and MCF-7 cells in number and diameter (Fig. 3e-f). Then, we used ow cytometry to test the expression of CD44 and CD24. CD44 + /CD24 − cells in breast tumors have cancer stem cell characteristics [22].
In addition, we examined the mRNA and protein expression of CD44 and three other typical stemness cell markers, OCT4, NANOG and SOX2. They were all upregulated in both MB231 and MCF-7 cells transfected with MCL-1 vectors, as assessed by RT-PCR and western blot (Fig. 4a b). CD44 is a glycoprotein that resides at the cell surface. MCL-1 is located at the outer mitochondrial membrane and mitochondrial matrix in pluripotent stemness cells [23].
Immuno uorescence staining revealed that the expression of MCL-1 (green) correlated with CD44 (red) in cells (Fig. 4c). We have proven that the expression level of MCL-1 in resistant 231/MDR cells is much higher than it is in MB-231 cells (Fig. 1a). To comprehensively clarify the effect of MCL-1 on taxol resistance, we blocked MCL-1 expression in MDR cells by treating them with the MCL-1 inhibitor M-i. MTT assays revealed that MDR-231 cells treated with the MCL-1 inhibitor M-i regained sensitivity to taxol (Fig. 5a, b). As expected, the mammosphere forming ability of the MCL-1 blocking cells was weakened (Fig. 5c). The expression levels of SOX2, Nanog, OCT4 and CD44 were decreased with the inhibition of MCL-1 by western blot and immuno uorescence (Fig. 5d, e), suggesting that cancer cell stems were eliminated. Taken together, these data demonstrated that overexpression of MCL-1 is greatly increased in cancer stem cells and depresses taxol sensitivity in breast cancer cells.

MCL-1 activates the WNT/β-catenin pathway
Having shown the effects of MCL-1 on chemotherapy resistance and cancer stem cells. Next, we investigated the signaling pathway by which MCL-1 alters the function of cancer stem cells. WNT/β-catenin, Notch and Hedgehog have been reported to play essential roles in cancer stem cells [24,25]. Herein, we examined which signaling pathway is involved in MCL-1 in breast cancer by a dual-luciferase reporter gene assay. The MB231 cells overexpressing Mcl-1 and MCF-7 cells were transfected into Renilla and one of the following signal pathway plasmids: TCF/catenin plasmid (WNT/β-catenin reporter plasmid), RBP plasmid (Notch reporter plasmid) or GLI plasmid (Hedgehog reporter plasmid). The results showed that WNT/β-catenin signaling was signi cantly activated, while the other signaling pathways experienced no signi cant change (Fig. 6a-c). We further con rmed that WNT/β-catenin signaling is inhibited by silencing MCL-1 (Fig. 6d). Staining of β-catenin showed that nuclear localization of β-catenin in MCL-1-overexpressing cells was more evident than it was in control cells, as shown by immuno uorescence data (Fig. 6e). Western blotting also revealed similar results, showing that β-catenin was increased in MCL-1overexpressing cells (Fig. 6f). To further clarify that MCL-1 activates WNT/β-catenin signaling in breast cancer, small molecule inhibitors of WNT/β-catenin (CWP232228; HY-18959 MCE) [26], Notch inhibitor 1 (HY-12860 MCE) and Hedgehog inhibitor (CUR61414; HY-113965 MCE) were used for our experiments. We treated MB231 and MCF-7 cells with 5 nM concentrations of the inhibitors and different concentrations of taxol at the same time. The data showed that upregulated MCL-1 increased drug resistance and that the WNT/β-catenin inhibitor reversed it (Table 3). These results have shown that MCL-1 activates the WNT/β-catenin signaling pathway and promotes the nuclear translocation of β-catenin. Blocking the WNT/β-catenin signaling pathway reversed the effects of MCL-1 on breast cancer cell drug resistance. Next, to investigate how MCL-1 activates the WNT/β-catenin signaling pathway, we used coimmunoprecipitation (Co-IP) to examine the interaction between MCL-1 and WNT proteins. The results showed that MCL-1 had no interaction with the breast cancer WNT proteins wnt 3a, wnt 10b and Frizzld 6 (Fzd 6), but there was an interaction between MCL-1 and the WNT pathway receptor LRP6 (Fig. 6g and 6 h). To con rm the interaction, we silenced LRP6 and used different concentrations of taxol before analyzing the cells by MTT assay. The data showed that silencing LRP6 reversed taxol resistance (Table 3). Furthermore, we investigated by IP how MCL-1 modulates the WNT receptor interaction. Cocultured 293T cells were transfected with MCL-1-Flag, LRP6-HA, Wnt 3a-myc and Wnt 10b-His plasmids ( Fig. 6i and 6j). Cell lysates were immunoprecipitated with an HA antibody and immunoblotted with anti-Flag, anti-HA, anti-Myc or anti-His antibodies. The results con rmed that LRP6 coimmunoprecipitated with MCL-1, Wnt 3a, Wnt10b and LRP6. Moreover, upregulated MCL-1 enhances the binding between Wnt 10b and LRP6. These results suggest that MCL-1 directly interacts with LRP6 and increases the binding between the WNT receptor and ligands in cells.

MCL-1 overexpression is associated with tumorigenesis and taxol resistance in vivo.
We have demonstrated that MCL-1 overexpression induces taxol resistance in vitro. Furthermore, we conducted tumorigenicity assays and xenograft tumor growth assays. Tumorigenicity assays revealed that the tumorigenicity of MCL-1-overexpressing cells was remarkably stronger than that of control cells (P < 0.0001; Fig. 7a). Moreover, the overexpression MCL-1 xenograft nude mouse group had a poor prognosis (Fig. 7b). In the xenograft tumor growth assay, both the tumor size and weight were remarkably reduced in the MCL-1 inhibitor M-i + taxol group compared with the DMSO group, the MCL-1 overexpression + taxol group and taxol group (Fig. 7c-g). Taken together, our data revealed that MCL-1 plays a very important role in taxol resistance and the characteristics of cancer stem cells in breast cancer. 3.6 A high level of MCL-1 expression predicts chemoresistance in breast cancer.
Identifying markers of breast cancer stem cells and chemoresistance is important for e cacy strati cation in the clinic. To con rm the correlation between MCL-1 expression and poor prognosis in breast cancer patients, we classi ed the 168 breast cancer patients used in this study into 4 subtypes according to their MCL-1 staining intensity of and followed up. The patients with high MCL-1 expression had a poor prognosis compared with the patients with low MCL-1 expression (P < 0.05; Fig. 8a). To further clarify the correlation between MCL-1 and cancer stem cell relevant genes (CD44 and LRP6), we performed qRT-PCR on biopsy tissues from the 168 breast cancer patients. There was a signi cant positive correlation between MCL-1 and CD44 (R 2 = 0.3603, P < 0.0001) and LRP6 (R 2 = 0.4356, P < 0.0001) (Fig. 8b). These results implied that upregulated MCL-1 enhances chemoresistance by modulating stemness properties.

Discussion
Breast cancer is the most common malignant tumor in women and the leading cancer that threatens women's lives. It is well known that the recurrence and metastasis of breast cancer after chemotherapy is essentially caused by the residual resistance of breast cancer cells to chemotherapy. Therefore, it is of great signi cance to study the mechanism of chemoresistance in breast cancer, explore how to reduce chemoresistance and increase chemosensitivity to prevent recurrence and metastasis of breast cancer and reduce the risk of death of breast cancer. Here, we identi ed MCL-1 as a novel signature gene of breast cancer stem cells. First, we found that MCL-1 gene expression is higher in NAC-R patients than it is in NAC-S patients and that MCL-1 expression is greatly increased in mammospheres compared to monolayers in primary 3D culture. Second, we used the multidrug resistance model 231/MDR to con rm the correlation of MCL-1 expression and taxol sensitivity. Third, overexpressing MCL-1 enhances mammosphere formation and upregulates the levels of stemness-associated genes SOX2, Nanog and OCT4 in breast cancer cells. Fourth, we found that the expression of MCL-1 was positively correlated with CD44 by immunohistochemical staining. Fifth, a high level of MCL-1 expression predicts poor prognosis in breast cancer.
MCL-1 encodes an anti-apoptotic protein that is a member of the Bcl-2 family [27]. Two research teams, those of Kelly, G. L and Gasca, J, reported that MCL-1 and other Bcl-2 family proteins were altered in response to chemotherapy [28][29]. MCL-1 and WNT/β-catenin were both listed as potential therapeutic target genes in liver cancer [30]. In our study, we con rmed that MCL-1 activated WNT/β-catenin signaling in breast cancer by following evidence. First, overexpressing and silencing MCL-1 promoted and reduced the activation of WNT/β-catenin signaling, as shown by luciferase reporter assay. Second, the nuclear translocation of β-catenin was found by western blot and immuno uorescence staining assays to be increased following MCL-1 upregulation. Moreover, IP assays found that MCL-1 interacts with LRP6, and Co-IP assay discovered that MCL-1 may help LRP6 proteins interact with the WNT/β-catenin ligand receptor Wnt 10b. Taken together, we conclude that MCL-1 plays an important role in the progression of breast cancer though the WNT/β-catenin signaling pathway (Fig. 9).
The WNT/β-catenin signaling pathway has been implicated in oncogenesis and in several developmental processes, including regulation of cell fate and patterning during embryogenesis [31]. This pathway supports stem and cancer stem cells [32]. When the concentration of β-catenin in the cytoplasm reaches a certain level, it can translocate to the nucleus and combine with TCF/lefs, which lead to cell proliferation, differentiation and maturation [33]. Several studies in different tumors have con rmed that the WNT/β-catenin signaling pathway mediates stem cell renewal and drug resistance by promoting translocation to the nucleus of β-catenin [34][35][36]. Our data provide evidence that MCL-1 promotes cancer stem cell characteristics and drug resistance though the WNT/βcatenin signaling pathway, which offers a novel strategy for targeted therapy in breast cancer.
Our research nds that MCL-1 is a stem cell signature gene in breast cancer and is a novel regulator of the WNT/β-catenin signaling pathway. However, a limitation of our research is the number of samples and the retrospective analysis. Prospective, multicenter, large sample clinical trials are needed to further con rm the predictive value of MCL-1.

Conclusion
Our team determined a new role of MCL-1 in chemotherapy resistance and prognosis for breast cancer patients. We revealed that MCL-1 mediates cancer stemness cell and taxol resistance both in vitro and in vivo. The effect is associated with the activation of the WNT/β-catenin signaling pathway, which directly interacted with LRP6 to regulate its expression. These ndings provide evidence that the MCL-1-WNT/β-catenin axis may be a target for breast cancer therapy, which offers novel strategies for the precise treatment of breast cancer.

Declarations
Availability of data and materials All data in this study are included in this publication and related data les.      t-test, *P 0.05, **P 0.01, *** P 0.001).