STAT3 regulates 5‐Fu resistance in human colorectal cancer cells by promoting Mcl‐1–dependent cytoprotective autophagy

Abstract Chemoresistance to 5‐fluorouracil (5‐Fu)‐based chemotherapy is one of the primary reasons for the failure of colorectal cancer (CRC) management. STAT3 can mediate tumor drug resistance through a variety of diverse mechanisms. Nonetheless, the underlying mechanisms of STAT3‐induced 5‐Fu resistance in CRC are still poorly understood. Here, we aimed to investigate the potential mechanism(s) of STAT3‐induced 5‐Fu resistance in CRC. Quantitative RT‐PCR and Western blot were used to test the expression of STAT3 and Mcl‐1 in chemosensitive and chemoresistant CRC tissues and cell lines. After overexpression or knockdown of STAT3 or Mcl‐1, and/or treatment with or without 5‐Fu or chloroquine (CQ), we tested cell viability, inhibitory concentration 50% (IC50) value of 5‐FU, cell apoptosis, proliferation, migration, and autophagy. STAT3 and Mcl‐1 were significantly upregulated in the chemoresistant CRC tissues and cell lines, and STAT3 positively regulated Mcl‐1. Functional studies demonstrated that STAT3 promoted 5‐Fu resistance in CRC. Mechanistically, STAT3 triggered autophagy via Mcl‐1 to induce cancer chemoresistance. Our results show that STAT3 regulates 5‐Fu resistance in CRC by promoting Mcl‐1–dependent cytoprotective autophagy. Our results provide a novel role of STAT3 and may offer a new approach for managing CRC 5‐Fu resistance.

drugs for CRC in palliative and adjuvant treatment. In the last few decades, several adjustments to the strategies of 5-Fu-based combination regimens have been established. 5 However, 5-Fu resistance still remains common in the clinical setting. Therefore, it is essential to investigate the therapeutic targets of 5-Fu resistance in CRC.
Autophagy is a bulk lysosomal degradation response to cellular stress, such as nutrient starvation or metabolic stress. 6 The main function of autophagy is to maintain intracellular metabolic homeostasis by degrading unfolded, damaged or aggregated proteins and organelles. In recent years, autophagy has been shown as a novel therapeutic target in anticancer drug resistance. 7 Inhibition of autophagy has been revealed to improve the efficacy of chemotherapy for many tumors. For instance, chloroquine (CQ), an important autophagy inhibitor, has been demonstrated in application prospects in nasopharyngeal carcinoma 8 and gallbladder cancer. 9 Accordingly, targeting autophagy may be a promising approach for enhancing the chemotherapy effect in CRC.
Signal transducer and activator of transcription (STAT) 3 is an essential member of the STATs family, which exists in the cytosol and can be transferred to the nucleus to bind DNA with or without phosphorylation. 10,11 STAT3 is abnormally expressed in a variety of tumor tissues and cell lines and is closely related to tumor proliferation, differentiation and cell apoptosis. 12,13 Furthermore, STAT3 has been recently reported to be involved in tumor drug resistance. 14,15 STAT3 can mediate tumor drug resistance in a variety of ways. A previous study revealed that exosomal transfer of p-STAT3 contributes to 5-Fu resistance in CRC. 16 Therefore, STAT3 has attracted much attention in antitumor research in recent years and has become a good target for cancer treatment. 17 However, the mechanism of STAT3 mediating chemotherapy drug resistance is still unclear and needs further study.
Myeloid cell leukemia-1 (Mcl-1) is a member of the BCL-2 family and has been reported to be highly expressed in various tumor tissues and cell lines, including CRC. 18 The expression of Mcl-1 is regulated by different cytokines and related signaling pathways, such as the JAK/STAT pathway. 19,20 It has been revealed that the activated STAT3 monomer forms a dimer, transfers from the cytosol to the nucleus, binds to the promoter of the target gene(s), and mediates the expression of Mcl-1. 21 Recently, Mcl-1 was described to participate in tumor cells resistant to chemotherapy drugs. 22 Blocking the expression of Mcl-1 can effectively induce cell apoptosis and cell autophagy and improve the sensitivity of tumor cells to chemotherapy, which provides a new target for drug intervention and a new way for tumor treatment. 23 However, there is a little report on whether STAT3 participates in chemotherapeutic drug resistance by regulating Mcl-1-induced autophagy.
Therefore, in this paper, we investigated the mechanism of the STAT3-Mcl-1 axis in chemotherapy resistance induced by autophagy. Our study may provide a basis for targeted treatment of CRC and improve the sensitivity of CRC to chemotherapy drugs.

| Patient samples
Chemosensitive (n = 45) and chemoresistant (n = 45) CRC tissues were obtained from the Department of Gastroenterology, Shengjing Hospital of China Medical University. All subjects agreed to participate in the initial operation and received adjuvant chemotherapy based on 5-FU. The follow-up deadline was June 30, 2018. 5-FU resistance was defined as patients with disease progression during initial chemotherapy or recurrence within 6 months after completion of initial chemotherapy, and 5-FU sensitivity was defined as patients with recurrence for more than 6 months or no recurrence. 24 Recurrence was monitored by chest X-ray, computerized tomography (CT), and/or endoscopic biopsies of the gastrointestinal tract.
The collected samples were immediately frozen in liquid nitrogen.

| Cell lines
The human colon fibroblast cell line CCD-18co, 5-Fu-resistant CRC cell line HCT8Fu and its parental sensitive cell line HCT8 were obtained from the National Collection of Authenticated Cell Cultures.
HCT8Fu cells were maintained in a medium containing 0.5 μg/mL 5-Fu (Sigma-Aldrich). These cells were cultured at 37°C with 5% CO 2 and humidified atmosphere. The culture media were changed every 1-2 days.

| Colony formation assay
The cells were seeded in a six-well plate at a density of 500 cells/ well. After treatment with different concentrations of 5-Fu or transfection with pcDNA-STAT3, si-STAT3, si-Mcl-1#1, or si-Mcl-1#2, the cells were cultured for further 14 days until colonies formed.
Thereafter, the cell colonies were fixed with 4% paraformaldehyde (Sigma-Aldrich) for 1 hour and stained with 0.5% crystal violet (Sangon Biotech Co., Ltd) staining solution for 10 minutes. Then, the cell colonies were photographed and counted.

| Transwell assay
Cell migration was determined using the transwell assay. Briefly, after the cells were treated with PBS or 5-Fu and transfected with pcDNA-STAT3 or si-STAT3, the cells (2 × 10 5 per well) were seeded into the upper chamber (8.0μm pore size, Corning) with serumfree medium according to the manufacturer's commendations, while 600 μL of complete medium containing 20% FBS was added into the lower chamber. Thereafter, the chambers were incubated at 37°C with 5% CO 2 for 48 hours. The nonmigrated cells in the upper chamber were then removed and fixed with 4% paraformaldehyde for 30 minutes. The migrated cells were stained with 0.1% of crystal violet for 20 minutes and then counted under an inverted microscope.

| Transmission electron microscopy
After the cells were treated with PBS or 5-Fu and transfected with pcDNA-STAT3 or si-STAT3, or co-transfected with pcDNA-STAT3 were imaged with TECNAI 10 electron microscope (Philips) at 40-120 kV. The number of autophagosomes was quantified by analyzing 10 fields in each sample.

| Immunofluorescence
The autophagic activity was assessed using immunofluorescence for LC3-II as previously demonstrated. 25 Briefly, the treated cells were

| Quantitative real-time PCR
Total RNA was extracted from the tissues and cells using TRIzol rea-

| Statistical analysis
Data are expressed as mean ± standard deviation (SD) for at least three individual experiments, and statistically analyzed by GraphPad software Prism 8 (GraphPad Software, Inc.). Differences between two groups were compared by Student's t test, and differences among three or more than three groups were performed by one-way analysis of variance (ANOVA). P < 0.05 was considered statistically significant.

| STAT3 is elevated in 5-Fu-resistant CRC tissues and cell lines
The role of STAT3 in 5-Fu chemoresistance has not fully been clarified. To evaluate the functional role of STAT3 in 5-Fu chemoresistance in CRC, we measured the mRNA and protein expression of STAT3 in chemosensitive and chemoresistant tissues from patient samples, as well as in cell lines. As revealed in Figure 1A Figure 1C).  Figure 3A,B). These results implied that STAT3 promoted 5-Fu resistance by reducing cell apoptosis in CRC cells.

| STAT3 induces CRC 5-Fu resistance via autophagy
Our above findings showed that STAT3 itself had no effects on the regulation of cell apoptosis, reminding us of the potential protective with pcDNA-STAT3 (p < 0.01, Figure 4B). Immunofluorescence assay also disclosed that LC3 aggregation was weakened by si-STAT3 in HCT8Fu cells and stimulated by pcDNA-STAT3 in HCT8 cells

| STAT3 positively regulates Mcl-1expression in CRC cells
We further explored the potential regulatory machinima of STAT3 in CRC 5-Fu resistance. Mcl-1, a key target of STAT3, 29 is an indispensable factor in malignant cell growth, apoptosis, autophagy, and drug resistance. 30   (p < 0.05, Figure 6D). Cell apoptosis results demonstrated that the percentage of apoptotic cells was significantly decreased by overexpression of STAT3 in the presence of 5-Fu (p < 0.001). However, these effects could be reversed by si-Mcl-1#1 or si-Mcl-1#2 in HCT8 cells (p < 0.001, Figure 6E). The apoptosis-related proteins showed consistent results, including cleaved PARP and cleaved caspase3 ( Figure 6F). These results indicated that STAT3 regulated 5-Fu resistance in CRC cells, at least by promoting Mcl-1-dependent cytoprotective autophagy (Figure 7). F I G U R E 4 STAT3 induces colorectal cancer (CRC) 5-Fu resistance via autophagy. (A) Protein expression of autophagy-related proteins after transfection with si-STAT3 or pcDNA-STAT3 in HCT8Fu cells and HCT8 cells. The protein expression was normalized to tubulin as a loading control. Representative immunoblots from each group (n = 4 per group). (B) The formation of autophagic vesicles in the HCT8Fu cells transfected with si-STAT3 and HCT8 cells transfected with pcDNA-STAT3. (C) LC3 aggregation in the HCT8Fu cells transfected with si-STAT3 and HCT8 cells transfected with pcDNA-STAT3. (D) Cell viability of HCT8 cells transfected with pcDNA-STAT3 in the presence or absence of CQ. (E and F) Cell apoptosis and protein expression of apoptosis-related proteins in HCT8 cells transfected with pcDNA-STAT3 in the presence or absence of CQ under 5-Fu treatment. The protein expression was normalized to tubulin as a loading control. Representative immunoblots from each group (n = 4 per group). (G) Protein expression of apoptosis-related proteins in HCT8 cells transfected with pcDNA-STAT3 in the presence or absence of CQ without 5-Fu treatment. The protein expression was normalized to tubulin as a loading control. Representative immunoblots from each group (n = 4 per group). Data are expressed as mean ± standard deviation (SD). */# p < 0.05, **/## p < 0.01, ***/### p < 0.001

| DISCUSS ION
The 5-year survival of CRC is poor, mainly due to recurrence and resistance to chemotherapy. 5 Therefore, it is essential to understand the mechanisms of chemoresistance in CRC to pursue more efficient intervention approaches for CRC. It has been well acknowledged that constitutive activation of STAT3 is reported in several malignant tumors, and interference with the STAT3 pathway might reestablish the susceptibility to anticancer drugs. 14,32 A previous study revealed that exosomal transfer of p-STAT3 was responsible for acquired 5-FU resistance in CRC. 16 Nevertheless, the mechanism by which STAT3 regulates downstream factors and its relationship with autophagy is still unclear and needs to be further studied. Our study focused on the downstream factors of STAT3 and autophagy-related mechanisms on acquired 5-FU resistance in CRC.
To confirm the role of STAT3 in the drug resistance of CRC, we first tested the expression of STAT3 in the chemosensitive and chemoresistant CRC tissues from patient samples and cell lines. In the present study, HCT8Fu cells were used as a chemoresistant cell line, which was established by utilizing the incremental dose of 5-Fu. HCT8 cells were used as a homologous chemosensitive cell line.
In line with previous studies, 31,33 we also found that the expression of STAT3 was significantly increased in the chemoresistant cancer tissues and cell lines. In addition, we observed that the expression of STAT3 was altered by 5-Fu treatment and after replacement with medium in the HCT8 cells but not in the HCT8Fu cells, implying that Autophagy is a dynamic, sequential, multistep process of degradation of cytoplasmic proteins and organelles. 34 It is a significantly complex process with opposite functions, namely cytoprotective autophagy and lethal autophagy, in tumors. 35,36 Cytoprotective autophagy can prevent cancer cell apoptosis induced by anticarcinogen, while lethal autophagy can prompt cell apoptosis combined with anticarcinogen. 37 There is a cross-talk relationship between autophagy and apoptosis in antitumor therapy, and they exert synergistic and antagonistic effects in cancer. 36 Autophagy usually occurs before apoptosis. 38 An increasing body of evidence suggests that cytoprotective autophagy not only increases the survival of cancer cells but also enhances the drug resistance of tumors in several cancer types. [39][40][41] The potential underlying mechanisms might be that autophagy induced by therapeutic agents maintains cancer cell metabolism by recycling the damaged proteins and organelles and then inhibits DNA damage, leading to drug resistance. 42 In addition, it has been reported that cytoprotective autophagy promotes drug resistance through the inhibition of apoptosis in cancer cells. 38,43 In our study, we found that STAT3-induced 5-Fu resistance was elimi- also an imperative molecular bridge connected to autophagy and cell apoptosis. 44 Mcl-1 was reported to prevent autophagy by interacting with the autophagy protein Beclin-1 and inhibiting the formation of autophagosomes. [45][46][47] However, a previous study found that overexpression of Mcl-1 significantly promoted mitophagy, and Mcl-1 acted as an LC3-interacting mitophagy receptor that can be targeted to induce mitophagy. 48 As an established BH3-mimetic for Mcl-1, UMI-77-mediated mitophagy was dependent on the LIR motif but Representative immunoblots from each group (n = 4 per group). Data are expressed as mean ± standard deviation (SD). #/& p < 0.05, **p < 0.01, ***p < 0.001 not on the anti-apoptotic function of Mcl-1. 48 The differences might be due to the different diseases, but it is more likely that it is a result of the complex regulatory relationship between autophagy and apoptosis. Furthermore, previous studies revealed that inhibition of Mcl-1 can overcome intrinsic and acquired regorafenib resistance in CRC by renovating cell apoptosis 23 and conquer lapatinib resistance via BAK-dependent autophagy in CRC cells. 49 Therefore, targeting Mcl-1 might be a promising technique for overcoming chemoresistance in CRC.
In the present study, we measured the expression of Mcl-1 in both chemosensitive and chemoresistant CRC tissues and cell lines. Similar to previous studies, 31,50 we also confirmed that