FR 180204

Down-regulation of autophagy-associated protein increased acquired radio-resistance bladder cancer cells sensitivity to taxol

Xiangli Ma, Guangmin Mao, Rulve Chang, Fang Wang, Xiangyan Zhang & Zhaolu Kong

ABSTRACT

Background: As a bladder-preserving therapy, radiation therapy (RT) has been widely used in the treatment of bladder cancer (BCa) and made great progress in the past few decades. However, some BCa patients have low RT responsiveness and local recurrence rate after RT could reach 50%. Acquired radio-resistance (ARR) is one of the important reasons for the failure of RT. Unfortunately, these ARR cells also lack sensitivity to chemotherapy and cause tumor recurrence and metastasis.
Purpose: To build ARR-phenotype BCa cell model, discuss the possible molecular mechanism of ARR and find effective target molecules to overcome ARR.
Materials and methods: Five thousand six hundred and thirty-seven cells were subjected 30 times to 2 Gy of c-rays and the surviving cells were called 5637R. Colony formation and MTT assay were applied to evaluate cells sensitivity to ionizing radiation (IR) and anti-neoplastic agents, respectively. Cells abilities of migration and invasion were determined using transwell method. Quantitative real-time polymerase chain reaction (RT-qPCR) and western blot (WB) were respect- ively utilized to compare the difference of gene and protein expression between 5637 and 5637R cells. Molecule inhibitors and small interfering RNA (siRNA) systems were employed to decrease the expression of target proteins, respectively.
Results: BCa cells survived from fractionated irradiation (FI) exhibited tolerance to both IR and chemotherapy drugs. These ARR cells (5637R) had elevated migration and invasion abilities, accompanied by increased expression of epithelial mesenchymal transition (EMT)-related transcrip- tion factors (ZEB1/Snail/Twist). Moreover, 5637R cells showed enhanced cancer stem cell (CSC)-like characteristics with activated KMT1A–GATA3–STAT3 circuit, a newly reported self-renewal pathway of human bladder cancer stem cell (BCSC). Combined with Kaplan–Meier’s analysis, we speculated that GATA3/MMP9/STAT3 could be an effective molecular panel predicting poor prognosis of BCa. In order to enhance the sensitivity of resistant cells to radiation, we introduced ERK inhibitor (FR 180204) and STAT3 inhibitor (S3I-201). However, both of them could not enhance ARR cells response to IR. On the other hand, siRNAs were respectively implemented to inhibit the expres- sion of endogenous Beclin1 and Atg5, two important autophagy-related genes, in BCa cells, which significantly increased 5637R cells death upon taxol exposing. Similarly, chloroquine (CQ), a classic autophagy inhibitor, enhanced the cytotoxicity of taxol only on 5637R cells.
Conclusions: Long-term FI treatment is an effective method to establish the ARR-phenotype BCa cell model, by enriching BCSCs and enhancing cells migration and invasion. Both inhibiting the expression of autophagy-related proteins and using autophagy inhibitor can increase the sensitiv- ity of ARR cells to taxol, suggesting that autophagy may play an important role in ARR cells chem- ical tolerance.

KEYWORDS
Acquired radio-resistance; radiosensitivity; autophagy; chemoresistance

Introduction

Bladder cancer (BCa) is one of the most common tumors with the highest incidence in the urinary system. In 2020, the total number of new BCa cases occurred in the United States was estimated to 81,400, including 62,100 males and 19,300 females. Meanwhile, an expected 17,980 Americans will die from BCa, including 13,050 males and 4930 females (Siegel et al. 2020). Similarly, an epidemiological study in China found that BCa was also the malignant tumor with the highest incidence of the male urinary system. Moreover, the incidence of BCa in China shows an increasing trend year by year, which is independent on gender and lifestyle (Han et al. 2013). Recent studies have reported that some molecular markers in urine, such as NMP22 (Miyake et al. 2017) and survivin (Chang et al. 2017), can be used to screen high-risk groups of BCa. In addition, more and more data supported that smoking (Ogihara et al. 2016) and hor- monal disorders (Gil et al. 2019) are susceptible factors for BCa.
In clinical practice, around 90% BCa cases are urothelial carcinoma and could be divided into two distinct forms: non-muscle-invasive bladder cancer (NMIBC) and muscle- invasive bladder cancer (MIBC), with different prognosis and treatment options (Alfred et al. 2017; Babjuk et al. 2017). Compared to NMIBC, MIBC is a more life-threaten- ing disease for distant metastases (Kaufman et al. 2009). Radical cystectomy (RC) and curative-intent radiotherapy (RT) are two definitive treatments for MIBC (Ghate et al. 2018). Although RC is the main method in the clinical treat- ment of BCa, large area tissue removal and urinary tract diversion not only reduce the patient’s quality of life, but also cause the patient facing certain psychological and men- tal pressure. Therefore, as a bladder-preserving strategy, RT has been widely used in treating MIBC in western countries (Soloway 2013). However, MIBC has low RT responsiveness and the local recurrence rate can be as high as 50% (El-Taji et al. 2016). Acquired radio-resistance (ARR), cells show tol- erance to ionizing radiation (IR) after primary RT, is one of the important reasons for the failure of clinical RT (Shimura et al. 2010). Until now, the exact mechanism of ARR remains unclear and lacks effective countermeasures (Shimura 2011). In addition, those ARR cells also show resistance to chemotherapy after recurrence.
To elucidate the possible molecular mechanism of ARR, we established long-term fractionated irradiation (FI)-treated BCa cells and found that accompanied by radiation resist- ance, ARR cells exhibited an increased distribution of S phase cells, enhanced migration ability and a constitutive activation of anti-extracellular signal-regulated kinase (ERK)/signal transducer and activator of transcription 3 (STAT3) pathway (Mao et al. 2018). In this study, we dis- closed that these ARR cells showed elevated migration and invasion ability, and increased expression of epithelial mes- enchymal transition (EMT)-related transcription factors (ZEB1/Snail/Twist) (Yun et al. 2015). They exhibited enhanced cancer stem cell (CSC)-like characteristics with activated KMT1A–GATA3–STAT3 (Yang et al. 2017) circuit. Combined with Kaplan–Meier’s analysis, we speculate that GATA3/MMP9/STAT3 could be a potential molecular panel predicting poor prognosis of BCa. We further introduced ERK inhibitor (FR 180204) (Ohori et al. 2007) and STAT3 inhibitor (S3I-201) (Jaganathan et al. 2010), respectively, to discuss feasible strategies for overcoming ARR. However, our data suggested that neither FR 180204 nor S3I-201 could enhance ARR cells sensitivity to IR. Finally, we compared ARR cells and their parent cells’ sensitivity to a series of anti-tumor drugs commonly used in BCa treatment. It was found that autophagy may play an important role in ARR cells chemo-resistance. Hopefully, this work can benefit the treatment of relapsed BCa.

Methods and materials

Cell culture and irradiation

Urinary bladder transitional cell line 5637 and its long-term FI-treated counterpart –5637R (Mao et al. 2018) were main- tained in Dulbecco’s modified Eagle’s medium (DMEM; high glucose; Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA) containing 10% fetal bovine serum (FBS; Invitrogen, Waltham, MA), 100 U/mL penicillin and 100 U/ mL streptomycin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) under a humidified atmosphere with 5% carbon dioxide at 37 ◦C. Cells were irradiated at room temperature (RT) in ambient air using a 137Cs source (c-ray, MDS Nordion, Toronto, Canada) at a dose rate of 0.77 Gy/min.

Cell treatment and chemicals

ERK inhibitor (FR 180204; sc-203945; work concentration: 10 lM) and STAT3 inhibitor (S3I-201; sc-204304; work con- centration: 2 lM) were purchased from Santa Cruz Biotechnology (Dallas, TX). Both 5637 and 5637R cells were planted in 60 mm dishes, 5 105 cells per dish, the inhibitor was added into the medium one hour before irradiation. Un-irradiated cells served as a control group. For anti-neo- plastic agents, cisplatin (PT, P4394) was from Sigma- Aldrich, paclitaxel (taxol, S1150), gemcitabine (GEM, S1714), mitomycin C (MMC, S8146) and 5-fluoracil (5-FU, S1209) were got from Selleck Chemicals (Houston, TX). Autophagy inhibitor (chloroquine (CQ); S1150; work con- centration: 20 lM) was purchased from Selleck Chemicals (Houston, TX) and added to media 24 hours before taxol treatment.

Migration and invasion assay

The migration and invasion assay was performed by using 24-well plate with chamber inserts (8-mm pore size, BD Biosciences, San Jose, CA) as described previously (Shen et al. 2014). Generally, we seeded 2 105 cells in the upper chambers and added 1300 mL DMEM with 10% FBS to the lower chambers. In invasion assay, we coated the upper chamber with Matrigel (BD Biosciences, San Jose, CA).
Following incubation for 16 hours at 37 ◦C, the upper cham- bers were removed, fixed and stained. After dry, cells were photographed and counted under light microscope (DFC450-C Leica, Wetzlar, Germany) (magnifica- tion, ×100).

Sphere formation assay

Cells were trypsinized and washed in DMEM. 2.0 105 cells were seeded in ultra-low attachment surface six-well plate (Corning, Inc., Corning, NY) and maintained in the serum- free medium (SFM) supplemented with 20 ng/mL EGF, 20 ng/mL bFGF and 2% B27 (Invitrogen, Waltham, MA). Cell spheres were imaged by microscope in five independent fields and spheres with diameters greater than 50 mm were counted after 8–10 days.

Soft agar colony anchoring assay

Low melting point agarose (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was dissolved in ultrapure H2O to final concentration of 0.8% and 1.2%. Sterile 2X DMEM, which contains 2% PS and 30% FBS, were mixed with agar- ose solution according to the formula: (1) 1.2% agarose: 2 DMEM 1:1, adding 3 mL of the mixture to 60 mm dishes in the bottom; (2) 0.8% agar: 2 DMEM 1:1, cells were suspended in medium and seeded at 500 cells/dish in triplicate in 60 mm dishes. After the upper layer of agarose fully solidified, incubated at 37 ◦C for two weeks. Cells were stained with 0.2% crystal violet (Sangon Biotech Co., Ltd., Shanghai, China) for 15–30 min at RT. Finally, the colonies were photographed and counted.

Quantitative real-time polymerase chain reaction (RT- qPCR) analysis

Total RNA was extracted and analyzed as described formerly (Zhang et al. 2015). The primers of target genes designed by GenScript (Nanjing, China) as mentioned previously (Wu et al. 2013; Yun et al. 2015; Yang et al. 2016). Primers of P62 were as follows: P62 forward, 50-CTGGGACTGAGA AGGCTCAC-30 and reverse, 50-GCAGCTGATGGTTTGGA AAT-30. Each sample was examined in triplicate and the amount of product was normalized relative to that of GAPDH. DDCt method was applied to calculate the gene expression.

RNA interference

Transient inhibition of human Beclin1 and Atg5 was carried out by transfection with 20 nM small interfering RNA (siRNA) (siBeclin1, 50-CAGUUUGGCACAAUCAAUA-30; siAtg5, 50-GGATGAGATAACTGAAAGG-30) with Lipofectamine 2000 (Invitrogen, Waltham, MA), respectively, according to the manufacturer’s instructions. All siRNA oligonucleotides, including siControl (50-CTGGACT TCCAGAAGAACA-30), were synthesized by Genepharma (Shanghai, China).

Western blot (WB) analysis

Cellular proteins were extracted two hours post-IR with M- PERTM protein extraction kit (Thermo Fisher Scientific, Waltham, MA, cat. no. 78501). Proteins were quantified (BCA assay kit; cat. no. 23227; Thermo Fisher Scientific, Waltham, MA) and 40 lg total protein was loaded on 10% SDS-PAGE gels. Proteins were then fractionated and trans- ferred to polyvinylidene difluoride (PVDF) membranes (0.45 lm; Merck KGaA, Darmstadt, Germany), blocked with 5% bovine serum albumin (BSA, AR2440, Sangon Biotech Co., Shanghai, China) for 1 h at RT. Subsequently, the mem- branes were incubated with primary antibodies at 4 ◦C over- night, and then incubated with secondary antibodies, horseradish peroxidase-conjugated secondary antibodies against mouse (1:1000; Beyotime Institute of Biotechnology, Nanjing, China) or rabbit (1:1000; Beyotime Institute of Biotechnology, Nanjing, China) for one hour at RT. Finally, the bands were visualized with the ChemiDoc XRS system (Bio-Rad Laboratories, Hercules, CA). Primary antibodies: anti-SUV39H1, anti-GATA-3, anti-MMP-9, anti-MMP-2, anti-ZEB1, anti-Snail, anti-Twist, anti-b-catenin, anti-Oct-3/ 4, anti-Vimentin, anti-E-cadherin (1:1000; Santa Cruz Biotechnology, Dallas, TX), anti-p-ATM, p-DNA-PKcs, anti- p-ATR, anti-Atg5, anti-Beclin1 (1:1000; Abcam PLC, Cambridge, UK), anti-p-ERK, anti-LC3B (1:1000; Cell Signaling Technology, Danvers, MA); anti-ERK, anti-STAT3, anti-p-STAT3 (Ser-705), anti-Bax, anti-P62, anti-Bcl-2 (1:1000; HuaAn Biotechnology, Hangzhou, China) and anti- GADPH (1:1000; Hangzhou Goodhere Biotechnology Co., Ltd., Hangzhou, China).

Colony formation assay (CFA)

Four hours post IR of 2 Gy, cells under different treatment (FR180204, S3I-201, IR, FR 180204 IR, S3I-201 IR) and control cells were trypsinized to a single-cell suspension and 200 cells were counted and seeded into 6-cm dishes. After 2 weeks of incubation, the colonies were fixed with 100% methyl alcohol at RT for 15 min, stained with 0.2% crystal violet for 1 h at RT, counted after sufficiently dried and the surviving fraction was calculated. For cells treated with FR180204 and S3I-201, inhibitors were added to the medium one hour ahead of IR, respectively.

Cell proliferation assay

Log-phase cells were seeded in 96-well plates (1 104 cells/ mL) with 200 lL cell suspension in each well for 24 hours. Anti-neoplastic agents were followed added into the media and co-incubated with cells for three days. Then, 20 lL thia- zolyl blue tetrazolium bromide (MTT, 0.5 mg/mL; Promega, Madison, WI) was added in the well and the plates were incubated for 4 h at 37 ◦C. Removed medium and added 150 lL DMSO to solve formazan crystals. The absorbance value was measured at 570 nm with a microplate reader (BioTek Instruments, Inc., Winooski, VT). IC50, half max- imal inhibitory concentration, was calculated to compare cells’ chemo-sensitivity.

Cell cycle analysis

1 107 log-phase cells were seeded in 10-cm dishes were cultured at 37 ◦C overnight. Cells were subsequently sub- jected to 0.6 lg/mL taxol and untreated cells were used as control. Cells were collected and fixed with pre-chilled 70% ethanol 0, 8, 24, and 48 hours later. DNA content was detected with flow cytometry as previously described (Zhang et al. 2015). The proportion of cells in various phases was calculated with Modfit LT 3.1 (Verity Software House, Inc., Topsham, ME).

Statistical analysis

Data were presented as the mean ± standard error of at least three independent experiments. The results were tested for significance using an unpaired Student’s t-test. Statistical analysis was conducted using STATA 10.0 statistics software (StataCorp LP, College Station, TX) and p < .05 was consid- ered to indicate a statistically significant difference. Results FI enhanced BCa cells ability of migration, invasion and expression of EMT-related transcription factors In order to study the mechanism of BCa recurrence and treatment tolerance after clinical RT, we established a radi- ation resistant (RR) cell model, 5637R, by subjected a human BCa cell line 5637 to long-term IR as described pre- viously (Mao et al. 2018). Compared to 5637, 5637R cells showed obvious morphological changes (Figure 1(A), top). Meanwhile, 5637R cells exhibited higher migration (Figure 1(A), middle) and invasion ability (Figure 1(A), bottom) than 5637. The migration numbers were 257.00 ± 23.43 vs. 477.90 ± 92.32, (5637 vs. 5637R, p .016) and the invasion numbers were 106.20 ± 38.49 vs. 379.88 ± 36.53 (5637 vs. 5637R, p < .001) (Figure 1(B)). We further detected the expression of a series of EMT markers, including b-catenin, vimentin and E-cadherin (Xie et al. 2010). Figure 1(C) shows that both 5637 and 5637R cells had high abundance of b-catenin and E-cadherin protein and very low expression of vimentin, which was consistent with the characteristics of 5637 cells as low-grade malignant cells. On the other hand, a higher level of MMP2 and MMP9 proteins, which were closely associated with BCa metastasis (Gao et al. 2015), were observed in 5637R cells than that in 5637. We further determined the expression of EMT-related transcription fac- tors in both 5637 and 5637R cells (Figure 1(D)). Significantly, the expression of ZEB1, Snail and Twist is greatly increased in RR cells compared to their parent coun- terparts at both mRNA (Figure 1(D)) and protein levels (Figure 1(E)), suggesting that FI did induce EMT in 5637R cells. FI enriched stem cell-like tumor cells in BCa with activated KMT1A–GATA3–STAT3 circuit Considering that CSC population in the tumor are radio- resistant, we next discussed the possibility of FI inducing CSC-like phenotype changes in BCa cells. With tumor microsphere culture and soft agar colony anchoring assay, we confirmed that 5637R exhibited enhanced CSC proper- ties compared to 5637. It was found that 5637R could form more spheres and colonies than 5637 in the same cultivation conditions. 5637 vs. 5637R, 1.75 ± 0.66 vs. 6.00 ± 1.50, p < .001 (spheres, Figure 2(A), top) and 30.30 ± 5.25 vs. 40.30 ± 4.70, p .039 (clones, Figure 2(A), bottom). Correspondingly, the mRNA expression of CSC-related markers, including OCT3/4, CD44, CD24, and CD117 (Yun et al. 2015) was increased in 5637R cells compared to 5637 cells (Figure 2(B), left). WB data also supported that OCT3/ 4 was overexpressed in 5637R cells. It was also found that a novel self-renewal signaling pathway, KMT1A-GATA3- STAT3, was highly active in 5637R cells (Figure 2(B), right). Based on an online Kaplan–Meier plotter tool (http:// kmplot.com/analysis/), we analyzed the relationship between the mRNA expression of a series of EMT and CSC-related genes mentioned in this study and BCa patient prognosis, using the overall survival (OS) as the evaluation index. The results are listed in Table 1. We noticed that the expression of MMP-9 (p .002) and STAT3 (p .023) were correlated with poor prognosis in BCa patients, while GATA3 (p .002) was related to good prognosis. The Kaplan–Meier plot of MMP-9, STAT3 and GATA3 is paneled in Figure 2(C). Neither ERK nor STAT3 inhibitors could increase ARR- BCa cells radio-sensitivity In order to clarify the mechanism of FI exposure induced radiation tolerance on BCa cells, we further monitored the expression of the biomolecules relevant to DNA damage repair and cell growth in the RR cells (5637R) and the par- ental cells (5637). Figure 3(A) shows that 5637R was more radio-resistant than 5637 (left), with slightly increased DNA- PKcs phosphorylation (left, inset) and transcription (right). Meanwhile, 5637R cells have higher levels of p-STAT3Tyr705 than that in 5637. When subjected to IR, ERK activation was greatly increased in 5637R cells but not in 5637 cells (Figure 3(B)). Therefore, we speculated that the ERK and STAT3 pathways may be associated with ARR cells resistant to IR. In the present study, the effects of FR180204 (an ERK inhibitor) and S3I201 (a STAT3 inhibitor) on ARR cells radio-sensitivity were respectively determined. It was found that FR180204 greatly decreased IR-induced ERK phosphor- ylation in 5637R cells (Figure 3(C), top), but it did not enhance 5637R cells sensitivity to IR (Figure 3(C), bottom). For 5637R cells, the survival fraction of 2 Gy (SF2) with 10 lM FR180204 and without FR180204 was 0.68 ± 0.08 and 0.66 ± 0.12 (p .082), respectively. On the other hand, S3I- 201, which inhibited STAT3 activation in BCa cells (Figure 3(D), top), partially increased 5637R cells sensitivity to radi- ation (Figure 3(D), bottom). The SF2 of 5637R cells co- treated with 2 lM S3I-201 was reduced to 0.49 ± 0.05. However, the treatment of 2 lM S3I-201 alone caused about 68%±4% 5637R cells death, suggested there is an antagonis- tic effect of S3I-201 combined with IR. Our data showed that neither ERK nor STAT3 is an effective target for increasing the radio-sensitivity of ARR cells. ARR-BCa cells exhibited enhanced tolerance to taxol We next determined the chemo-sensitivity of 5637 and 5637R cells by MTT assay. The IC50 values of 5637 and 5637R cells were 0.31 ± 0.02 lg/mL vs. 4.57 ± 0.12 lg/mL (p < .001, taxol), 0.13 ± 0.03 lg/mL vs. 0.30 ± 0.02 lg/mL (p ¼ .001, GEM), 1.42 ± 0.30 lg/mL vs. 2.17 ± 0.2 lg/mL (p ¼ .023, PT), 0.50 ± 0.03 lg/mL vs. 0.71 ± 0.10 lg/mL (p ¼ .025, 5-FU) and 0.11 ± 0.02 lg/mL vs. 0.10 ± 0.02 lg/mL (p ¼ .573, MMC), respectively. Here, we introduced an item of the resistance index (RI IC50 (5637R)/IC50 (5637)) to quan- tify the chemo-sensitivity differences between 5637 and 5637R cells. As indicated in Figure 4(A), 5637R cells were more resistant to most chemotherapy drugs except MMC than 5637 cells. Furthermore, a 14-fold increase of taxol IC50 value in the 5637R cells vs. 5637 cells was noticed. To explore the mechanism of 5637R cells resistant to taxol, both 5637 and 5637R cells were subjected to taxol (0.6 lg/ mL) and the cell cycle distribution was monitored with a flow cytometer. We observed that taxol exposure caused an early G2 block on 5637 cells (eight hours, 15.78%±1.25% to 45.25%±4.32%), but a late G2 arrest on 5637R cells (24 hours, 18.71%±4.62% to 34.87%±2.51%) (Figure 4(B)). Moreover, there was no G1 phase 5637 cells could be detected 48 hours after drug treatment. Meanwhile, around 10.17%±1.37% cells entered G1 phase was observed in 5637R, which could partially explain why 5637R cells are resistant to taxol. Down-regulation of autophagy could increase ARR-BCa cells sensitivity to taxol It is known that taxol is a microtubule-stabilizing drug which induced mitotic arrest followed by cell death (Weaver 2014), as observed in Figure 4(B). Compelling evidence has indicated that taxol kills tumor cells through the inducing of autophagy and apoptosis (Xi et al. 2011). In order to eluci- date the mechanisms why RR cells were resistance to taxol, both 5637 and 5637R cells were subjected to taxol solution at different concentration (0, 0.06 lg/mL, 0.6 lg/mL, 6 lg/ mL). The proteins related to apoptosis and autophagy were detected by WB (Figure 4(C), left). Using ratios of Bcl-2/ BAX (Liao et al. 2018) and LC3B II/I (Yu et al. 2012) to respectively evaluate the levels of apoptosis and autophagy, we found that the both 5637 and 5637R have roughly equivalent apoptosis levels in response to taxol exposure. Moreover, dose-dependent Bcl-2/BAX ratio increase was only observed in 5637 cells (Figure 4(C), right top). Meanwhile, taxol induced a dose-dependent increasing of autophagy in 5637 cells. Whereas for 5637R cells, enhanced LC3B II expression could only be noticed in the high dose group (taxol: 6 lg/mL) (Figure 4(C), right bottom). Considering 5637 and 5637R cells had different response to taxol-triggered autophagy rather than apoptosis, we next observed the effect of autophagy on cells sensitivity to taxol. RNA interference was applied to knockdown the expression of two important autophagy-associated proteins, Beclin1 and Atg5 (Figure 4(D)). The cells viability was detected with MTT assay (Figure 4(E), top). Both siBeclin1 and siAtg5 treated 5637 cells showed tolerance to taxol (Figure 4(E), left bottom). The IC50 values of taxol were increased from 0.27 ± 0.05 lg/mL (5637siControl) to 0.31 ± 0.05 lg/mL (5637siAtg5) and 0.36 ± 0.02 lg/mL (siBeclin1), respectively (Table 2). Chloroquine, a classic autophagy inhibitor, was enrolled as a positive control in this study. Similarly, CQ did not affect 5637 cells response to taxol (Figure 4(E), right bottom). The IC50 values of taxol with and without CQ on 5637 cells were 0.31 ± 0.02 lg/mL and 0.30 ± 0.04 lg/mL, respectively (Table 3). In contrast, autophagy inhibition enhanced 5637R cells death upon taxol treatment. The IC50 values of taxol on 5637RsiControl, 5637RsiAtg and 5637RsiBeclin1 cells were 3.96 ± 0.36 lg/mL, 1.86 ± 0.20 lg/ mL and 1.94 ± 0.14 lg/mL, respectively (Table 2). Also, the IC50 values of taxol with and without CQ on 5637R cells were 4.57 ± 0.12 lg/mL and 2.08 ± 0.17 lg/mL, respectively (Table 3). It is worth mentioning that 5637R cells exhibited an increase in autophagy levels only when exposed to high concentrations of taxol (Figure 4(C)). Therefore, we specu- lated that autophagy is an important protective mechanism for 5637R cells in response to stress, such as radiation and cytotoxic agents, and there could be a dose threshold to ini- tiate autophagy. Discussion In order to clarify the mechanism of acquired radiation tol- erance, a BCa cell line, 5637 cells were objected to 30 FI treatments of 2 Gy/day. The survived cells were cultured and passaged, and we obtained progeny cells with typical ARR phenotype and named it 5637R. Different from picking clones to obtain stable cell lines, we believed that such hybrid cells harvested after FI and continuous passage are more similar to those ARR-cells produced after clinical RT in BCa patients. Comparing to 5637, 5637R cells have (1) increased migration and invasion ability, accompanied by up-regulation of EMT-related transcription factors (ZEB1, Snail, Twist) but no significant changes in the expression of EMT markers (b-catenin, vimentin, E-cadherin) (Figure 1); (2) enhanced microsphere formation and cloning anchoring capacities in specific media with highly expressed CSC- related markers (OCT3/4, CD44, CD24, CD117) and acti- vated KMT1A–GATA3–STAT3 circuit, a novel self-renewal pathway of bladder cancer stem cell (BCSC) (Figure 2); (3) decreased sensitivity to radiation companying by slightly p < .05 (ω5637 treatment group vs. 5637 control group; #5637R treatment group vs. 5637R control group) indicates a statistically significant difference. ATM: ataxia telangiectasia-mutated; ATR: ATM-Rad3-related; DNA-PKcs: DNA-dependent protein kinase catalytic subunit; ERK: extracellular signal-regulated kinase. increased DNA-PKcs transcription and phosphorylation. Meanwhile, we noticed that un-treated 5637R cells exhibited a constitutive phosphorylation of STAT3 while IR caused an increased activation of ERK on 5637R cells (Figure 3), and the same phenomenon did not appear on 5637 cells; (4) declined responses to a series of chemotherapy drugs, espe- cially to taxol. High dose of taxol increased autophagy- related protein expression (Beclin1, Atg5) in 5637R cells but not in 5637 cells. Autophagy inhibition caused more 5637R cells death upon taxol treatment (Figure 4). Combined with Kaplan–Meier’s analysis, we speculated that GATA3/MMP9/ STAT3 could be an effective molecular panel predicting poor prognosis of BCa. In 2017, using single cell sequencing technology, Yang first proposed that BCSC originated from BCa non-stem cells or bladder epithelial stem cells (Yang et al. 2017), and KMT1A-GATA3-STAT3 circuit is an unique self-renewal pathway of human BCSC (Yang et al. 2017). On the basis of their research, we speculate that continuous energy depos- ition during long-term FI can trigger genes changes, such as epigenetic related gene, and consequently promote the trans- formation of non-BCSC into BCSC cells. The population dominated by CSC-like cells exhibits the ARR phenotype. In this study, we noticed that combined with elevated STAT3 phosphorylation at Tyr705, 5637R cells also showed lower GATA3 expression and a higher level of KMT1A than that in 5637, similar to BCSC. It could be a useful model to screen radio/chemo-sensitizer targeted on CSCs. Autophagy is a self-cannibalization process (Rabinowitz and White 2010), which plays an important role in maintaining cell homeostasis in normal tissues (Chen et al. 2010). More and more studies support that activation of autophagy is closely related to cell survival in response to various stress, such as cytotoxic agents, IR and hypoxia (Mathew et al. 2007). Nevertheless, autophagy in cancer is very complicated and considered as a ‘double-edged sword’ in the previous literature (Janku et al. 2011). That is, autophagy have both tumor-promoting and tumor-suppress- ing properties. However, the role of autophagy in tumor cells response to anticancer drugs and RT is largely unknown. In the present study, we found that autophagy was activated under taxol treatment in BCa cells. In taxol- sensitive 5637 cells, such autophagy activation was inde- pendent of the dose of taxol. Whereas, in resistant 5637R cells, only when the working concentration of taxol was increased to 6 lg/mL, which caused more than 50% 5637R cells death, the autophagy process was triggered with increased ratios of LC3B II/I (Figure 4(C)). On the other hand, apoptosis was considered as the main way of taxol- induced cell death (Wang et al. 2000). Using ratios of Bcl2/ BAX to assess intracellular apoptosis levels, we observed that taxol triggered a dose-independent apoptosis in both 5637 and 5637R cells (Figure 4(C), right top panel). It was worth mentioning that in 5637 cells, the expression of both Bcl2 and BAX was increased in response to taxol exposure. In contrast, taxol caused Bcl2 and BAX decreasing in 5637R cells. Similarly, the expression of autophagy-related proteins, Beclin1 and Atg5, also changed conversely in taxol-treated 5637 and 5637R cells, respectively (Figure 4(C), left panel). Therefore, we believed that taxol-caused programmed-cell- death progresses, especially autophagy, could be very differ- ent in 5637 and 5637R cells. Further study of the process of autophagy in ARR cells may provide new targets to improve their therapeutic sensitivity. By enrolling autophagy inhibitor CQ (Pasquier 2016), and RNA interference (siBeclin1, siAtg5), we proved that inhibiting autophagy greatly increased 5637R cells sensitivity to taxol (Figure 4(E)). It was quite intriguing that autophagy inhibition exhibited the two-way effect on chemosensitivity of parental (5637) and radiation-tolerant cells (5637R). Does this mean that autophagy is the main cause resulted in ARR cells radio/chemo-resistance? Why do taxol-sensitive 5637 cells show tolerance after autophagy inhibition? Is this phenom- enon prevalent in BCa cells? Our subsequent research will focus on solving these problems. More BCa cell lines and their ARR counterparts will be enrolled. Autophagic flux and related signal pathways, such as PI3K/AKT/mTOR and AMPK, in response to various stress will be monitored. Identification of the key molecules involved in the above process may provide new targets to improve the efficacy of bladder-preservation treatments. References Alfred WJ, Lebret T, Comperat EM, Cowan NC, De Santis M, Bruins HM, Hernandez V, Espinos EL, Dunn J, Rouanne M, et al. 2017. Updated 2016 EAU guidelines on muscle-invasive and metastatic bladder cancer. Eur Urol. 71(3):462–475. Babjuk M, Bohle A, Burger M, Capoun O, Cohen D, Comperat EM, Hernandez V, Kaasinen E, Palou J, Roupret M, et al. 2017. EAU guidelines on non-muscle-invasive urothelial carcinoma of the blad- der: update 2016. Eur Urol. 71(3):447–461. Chang Y, Xu J, Zhang Q. 2017. Microplate magnetic chemiluminescence immunoassay for detecting urinary survivin in bladder cancer. Oncol Lett. 14(4):4043–4052. Chen S, Rehman SK, Zhang W, Wen A, Yao L, Zhang J. 2010. Autophagy is a therapeutic target in anticancer drug resistance. Biochim Biophys Acta. 1806(2):220–229. El-Taji OM, Alam S, Hussain SA. 2016. Bladder sparing approaches for muscle-invasive bladder cancers. Curr Treat Opt Oncol. 17(3):15. Gao Y, Guan Z, Chen J, Xie H, Yang Z, Fan J, Wang X, Li L. 2015. CXCL5/CXCR2 axis promotes bladder cancer cell migration and invasion by activating PI3K/AKT-induced upregulation of MMP2/ MMP9. Int J Oncol. 47(2):690–700. Ghate K, Brennan K, Karim S, Siemens DR, Mackillop WJ, Booth CM. 2018. Concurrent chemoradiotherapy for bladder cancer: practice patterns and outcomes in the general population. Radiother Oncol. 127(1):136–142. Gil D, Zarzycka M, Dulin´ska-Litewka J, Ciołczyk-Wierzbicka D, Lekka M, Laidler P. 2019. Dihydrotestosterone increases the risk of bladder cancer in men. Hum Cell. 32(3):379–389. Han S, Zhang S, Chen W. 2013. Analysis of the status and trends of bladder cancer incidence in China. Oncol Prog. 11:89–95. Jaganathan S, Yue P, Turkson J. 2010. Enhanced sensitivity of pancre- atic cancer cells to concurrent inhibition of aberrant signal trans- ducer and activator of transcription 3 and epidermal growth factor receptor or Src. J Pharmacol Exp Ther. 333(2):373–381. Janku F, McConkey DJ, Hong DS, Kurzrock R. 2011. Autophagy as a target for anticancer therapy. Nat Rev Clin Oncol. 8(9):528–539. Kaufman DS, Shipley WU, Feldman AS. 2009. Bladder cancer. Lancet. 374(9685):239–249. Liao W, Liu J, Liu B, Huang X, Yin Y, Cai D, Li M, Zhu R. 2018. JIB- 04 induces cell apoptosis via activation of the p53/Bcl-2/caspase pathway in MHCC97H and HepG2 cells. Oncol Rep. 40(6): 3812–3820. Mao G, Yao Y, Kong Z. 2018. Long term exposure to gammarays indu- ces radioresistance and enhances the migration ability of bladder cancer cells. Mol Med Rep. 18(6):5834–5840. Mathew R, Karantza-Wadsworth V, White E. 2007. Role of autophagy in cancer. Nat Rev Cancer. 7(12):961–967. Miyake M, Morizawa Y, Hori S, Tatsumi Y, Onishi S, Owari T, Iida K, Onishi K, Gotoh D, Nakai Y, et al. 2017. Diagnostic and prognostic role of urinary collagens in primary human bladder cancer. Cancer Sci. 108(11):2221–2228. Ogihara K, Kikuchi E, Yuge K, Ito Y, Tanaka N, Matsumoto K, Miyajima A, Asakura H, Oya M. 2016. Refraining from smoking for 15 years or more reduced the risk of tumor recurrence in non- muscle invasive bladder cancer patients. Ann Surg Oncol. 23(5): 1752–1759. Ohori M, Takeuchi M, Maruki R, Nakajima H, Miyake H. 2007. FR180204, a novel and selective inhibitor of extracellular signal- regulated kinase, ameliorates collagen-induced arthritis in mice. Naunyn Schmiedebergs Arch Pharmacol. 374(4):311–316. Pasquier B. 2016. Autophagy inhibitors. Cell Mol Life Sci. 73(5): 985–1001. Rabinowitz JD, White E. 2010. Autophagy and metabolism. Science. 330(6009):1344–1348. Shen YJ, Kong ZL, Wan FN, Wang HK, Bian XJ, Gan HL, Wang CF, Ye DW. 2014. Downregulation of DAB2IP results in cell prolifer- ation and invasion and contributes to unfavorable outcomes in blad- der cancer. Cancer Sci. 105(6):704–712. Shimura T. 2011. Acquired radioresistance of cancer and the AKT/ GSK3b/cyclin D1 overexpression cycle. J Radiat Res. 52(5):539–544. Shimura T, Kakuda S, Ochiai Y, Nakagawa H, Kuwahara Y, Takai Y, Kobayashi J, Komatsu K, Fukumoto M. 2010. Acquired radioresist- ance of human tumor cells by DNA-PK/AKT/GSK3beta-mediated cyclin D1 overexpression. Oncogene. 29(34):4826–4837. Siegel RL, Miller KD, Jemal A. 2020. Cancer statistics, 2020. CA A Cancer J Clin. 70(1):7–30. Soloway MS. 2013. ICUD-EAU International Consultation on Bladder Cancer 2012: recommendations on bladder cancer-progress in a can- cer that lacks the limelight. Eur Urol. 63(1):1–3. Wang TH, Wang HS, Soong YK. 2000. Paclitaxel-induced cell death: where the cell cycle and apoptosis come together. Cancer. 88(11): 2619–2628. Weaver BA. 2014. How taxol/paclitaxel kills cancer cells. Mol Biol Cell. 25(18):2677–2681. Wu K, Xie D, Zou Y, Zhang T, Pong RC, Xiao G, Fazli L, Gleave M, He D, Boothman DA, et al. 2013. The mechanism of DAB2IP in chemoresistance of prostate cancer cells. Clin Cancer Res. 19(17): 4740–4749. Xi G, Hu X, Wu B, Jiang H, Young CY, Pang Y, Yuan H. 2011. Autophagy inhibition promotes paclitaxel-induced apoptosis in can- cer cells. Cancer Lett. 307(2):141–148. Xie D, Gore C, Liu J, Pong RC, Mason R, Hao G, Long M, Kabbani W, Yu L, Zhang H, et al. 2010. Role of DAB2IP in modulating epi- thelial-to-mesenchymal transition and prostate cancer metastasis. Proc Natl Acad Sci USA. 107(6):2485–2490. Yang C, He H, Zhang T, Chen Y, Kong Z. 2016. Decreased DAB2IP gene expression, which could be induced by fractionated irradiation, is associated with resistance to c-rays and a-particles in prostate cancer cells. Mol Med Rep. 14(1):567–573. Yang Z, Li C, Fan Z, Liu H, Zhang X, Cai Z, Xu L, Luo J, Huang Y, He L, et al. 2017. Single-cell sequencing reveals variants in ARID1A, GPRC5A and MLL2 driving self-renewal of human bladder cancer stem cells. Eur Urol. 71(1):8–12. Yang Z, He L, Lin K, Zhang Y, Deng A, Liang Y, Li C, Wen T. 2017. The KMT1A–GATA3–STAT3 circuit is a novel self-renewal signal- ing of human bladder cancer stem cells. Clin Cancer Res. 23(21): 6673–6685. Yu L, Tumati V, Tseng SF, Hsu FM, Kim DN, Hong D, Hsieh JT, Jacobs C, Kapur P, Saha D. 2012. DAB2IP regulates autophagy in prostate cancer in response to combined treatment of radiation and a DNA-PKcs inhibitor. Neoplasia. 14(12):1203–1212. Yun EJ, Baek ST, Xie D, Tseng SF, Dobin T, Hernandez E, Zhou J, Zhang L, Yang J, Sun H, et al. 2015. DAB2IP regulates cancer stem cell phenotypes FR 180204 through modulating stem cell factor receptor and ZEB1. Oncogene. 34(21):2741–2752.
Zhang T, Shen Y, Chen Y, Hsieh JT, Kong Z. 2015. The ATM inhibitor KU55933 sensitizes radioresistant bladder cancer cells with DAB2IP gene defect. Int J Radiat Biol. 91(4):368–378.