Circular RNA circ_0006168 enhances Taxol resistance in esophageal squamous cell carcinoma by regulating miR-194-5p/JMJD1C axis

Chemoresistance is one of the major obstacles for cancer therapy in the clinic. Circular RNAs (circRNAs) are involved in the pathogenesis of esophageal squamous cell carcinoma (ESCC) and chemoresistance. This study aimed to explore the role and molecular mechanism of circ_0006168 in Taxol resistance of ESCC. The expression levels of circ_0006168, microRNA-194-5p (miR-194-5p) and jumonji domain containing 1C (JMJD1C) were measured by quantitative real-time polymerase chain reaction (qRT-PCR) or western blot. The half-inhibition concentration (IC50) value of Taxol was evaluated by Cell Counting Kit-8 (CCK-8) assay. Cell proliferation was evaluated by CCK-8 and colony formation assays. Cell migration and invasion were detected by transwell assay. Cell apoptosis was determined by flow cytometry. The interaction between miR-194-5p and circ_0006168 or JMJD1C was predicted by bioinformatics analysis (Circinteractome and TargetScan) and verified by dual-luciferase reporter and RNA Immunoprecipitation (RIP) and RNA pull-down assays. The mice xenograft model was established to investigate the roles of circ_0006168 in vivo. Circ_0006168 and JMJD1C were upregulated and miR-194-5p was downregulated in ESCC tissues, ESCC cells, and Taxol-resistant cells. Functionally, knockdown of circ_0006168 or JMJD1C increased Taxol sensitivity of ESCC in vitro via inhibiting cell proliferation, migration and invasion, and promoting apoptosis. Moreover, circ_0006168 could directly bind to miR-194-5p and JMJD1C was verified as a direct target of miR-194-5p. Mechanically, circ_0006168 was a sponge of miR-194-5p to regulate JMJD1C expression in ESCC cells. Furthermore, JMJD1C overexpression reversed the promotive effect of circ_0006168 knockdown on Taxol sensitivity. Besides, circ_0006168 silence suppressed tumor growth in vivo. Circ_0006168 facilitated Taxol resistance in ESCC by regulating miR-194-5p/JMJD1C axis, providing a promising therapeutic target for ESCC chemotherapy.


Introduction
Esophageal cancer (EC) is a common malignant tumor worldwide, ranking the 7th in incidence (approximately 572,000 new cases) and 6th in mortality (approximately 509,000 deaths) [1,2]. Esophageal squamous cell carcinoma (ESCC) is the most common histological type in EC and nearly half of these ESCC cases have occurred in China [3]. Although surgery, radiation therapy, chemotherapy and molecular targeted therapy have all made significant contributions to treatment, the 5 year survival rate for ESCC patients is still poor because patients are frequently diagnosed at advanced stages [4,5]. Besides, chemotherapy resistance (including Taxol resistance) is also an essential factor leading to poor prognosis in ESCC patients [6,7]. Therefore, it is essential to explore the mechanism of chemotherapy resistance or sensitivity for improving the prognosis of ESCC.
Circular RNA (circRNA) are a novel type of noncoding RNAs (ncRNAs) with covalently closed-loop structures lacking 5'-cap and 3'-end poly A tail [8,9]. Due to the special circular structures, circRNAs are not influenced by RNA excision enzymes and more stable than linear RNAs [10]. In recent years, growing evidence has indicated that circRNAs can function as pivotal mediators in modulating the progression of diverse cancers, including ESCC [11]. For instance, circRNA circ-TTC17 acted as a tumor suppressor through repressing ESCC cell growth and migration [12]. Moreover, circRNA hsa_ circ_0000337 served as a tumor promoter in ESCC via promoting cell proliferation, migration and invasion [13]. In addition, circRNAs have been reported to play critical roles in Taxol resistance [14,15]. Hsa_circ_0006168 (circ_0006168) is derived from CCR4-NOT transcription complex subunit 6 like (CNOT6L) gene and located at chr4:78,694,234-78,697,546, and it has been reported to act as an oncogene in ESCC [16]. Moreover, hsa_ circ_0030998 suppressed lung cancer tumorigenesis and Taxol resistance by sponging miR-558 [17]. In addition, hsa_circ_0002483 inhibited the progression and enhanced the Taxol sensitivity of non-small cell lung cancer via regulating miR-182-5p [18]. Nevertheless, the exact roles and regulatory mechanism of circ_0006168 in Taxol resistance of ESCC remain largely unknown.
CircRNAs are identified as competing endogenous RNAs (ceRNAs) or microRNA (miRNA) sponges/decoys in many cancers [19]. , an endogenous miRNA with low expression in many cancers, was shown to suppress ESCC cell progression [20,21]. Moreover, miR-194-5p downregulation induced Taxol resistance in ovarian cancer cells via affecting MDM2 expression [22]. Jumonji domain containing 1C (JMJD1C) has been identified as a DNA-damage response (DDR) component to regulate the cellular behaviors of cancer cells [23]. A previous research has demonstrated that JMJD1C can promote EC cell proliferation [24]. Base on the analysis of bioinformatics software (Circinteractome and Tar-getScan), we found that circ_0006168 and JMJD1C had complementary binding sequence for miR-194-5p, which stimulated us to assume the ceRNA regulatory network of circ_0006168/ miR-194-5p/JMJD1C in Taxol resistance of ESCC.
In this work, the abundance of circ_0006168, miR-194-5p and JMJD1C in ESCC tissues, ESCC cells, and Taxol-resistant cells was determined. Then we investigated the biological roles of circ_0006168 in Taxolresistant cell growth, migration, invasion, and apoptosis. Additionally, the function of circ_0006168 in vivo was investigated via xenograft tumor model. The purpose of this research was to offer a promising therapy strategy for ESCC with Taxol resistance.

Specimen collection
In this study, ESCC tissues (n = 40) and adjacent normal tissues (n = 40) were acquired from patients who had undergone surgery at Yantai Affiliated Hospital of Binzhou Medical University. None of the patients with ESCC received systemic or local treatment before the operation. These tissues were fast-frozen in liquid nitrogen and preserved at -80℃ until required. Informed consents have been obtained from all patients. And the protocols have been approved by the Research Ethics Committee of Yantai Affiliated Hospital of Binzhou Medical University.

RNase R treatment
To determine the circular characteristic of circ_0006168, total RNA (2 µg) was incubated with or without 6 units of RNase R (Geneseed Biotech, Guangzhou, China) for 0.5 h at 37℃. Subsequently, the expression of circ_0006168 and JMJD1C mRNA was tested via qRT-PCR analysis.

Cell counting Kit-8 (CCK-8) assay
CCK-8 purchased from Boster (Wuhan, China) was used for analyzing cell viability and half-inhibition concentration (IC 50 ) value of Taxol. In brief, 100 uL cell suspension (2 × 10 3 cells) was added into a 96-well plate. CCK-8 reagent (10 µL) was added to per well after respective treatment. After incubation for 2-3 h, a microplate reader (Bio-Rad) was employed to examine the absorbance of per well at 450 nm. IC 50 value of Taxol was calculated using Graphpad Prism (GraphPad Software, San Diego, CA, USA).

Colony formation assay
Briefly, transfected cells (Eca109/Taxol and KYSE150/ Taxol) were inoculated in a six-well plate and the medium was updated every 3 days. After incubation for 2 weeks, the cells were fixed using the paraformaldehyde (4%, Beyotime, Jiangsu, China), followed by staining with crystal violet (0.1%, Sinopharm Chemical Reagent, Beijing, China). Next, microscope (Leica, Wetzlar, Germany) was used to count the number of colonies (> 50 cells per colony).

Transwell assay
Transwell assay was used for measuring migration and invasion of Eca109/Taxol and KYSE150/Taxol cells. For migration, transfected cells (Eca109/Taxol and KYSE150/ Taxol) suspended in serum-free RPMI-1640 (0.2 mL) were seeded into the top chamber of the 24-well transwell plates (Costar, Corning, NY, USA). The bottom chamber was filled with medium containing 10% FBS (0.6 mL). After 24 h transfection, the migrated cells fixed using the paraformaldehyde (4%, Beyotime) and stained by crystal violet (0.1%, Sinopharm Chemical Reagent). The images were captured under a microscope (Leica) at × 100 magnification. Similar to the migration assay, excluded cells were seeded into a Matrigel-coated (BD Biosciences, Franklin, NJ, USA) top chamber for the invasion assay.

Cell apoptosis assay
Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (Keygen, Nanjing, China) was applied for detecting cell apoptosis. In short, Eca109/Taxol and KYSE150/Taxo cells were collected following transfection for 48 h, and then dyed with Annexin V-FITC and PI. After incubation for 20 min in the darkness, the percentage of apoptotic cells was examined via a flow cytometer (BD Biosciences).

RNA immunoprecipitation (RIP) assay
RIP assay was carried out using an EZ-Magna RIP kit (Millipore, Billerica, MA, USA). Briefly, RIP lysis buffer was employed to lyse Eca109/Taxol and KYSE150/Taxol cells, and RIP buffer that contained magnetic beads conjugated with human antibody against anti-Ago2 or anti-IgG was utilized to incubate the lysate products. Next, the beads were digested with protease K, and circ_0006168, miR-194-5p and JMJD1C levels were detected using qRT-PCR analysis.

Tumor formation assay in vivo
BALB/c nude mice (n = 20, male, 5-6 weeks old, weighing 18-25 g) were provided by Vital River Laboratory Animal Technology Co., Ltd (Beijing, China). Eca109/ Taxol and KYSE150/Taxol cells stably transfected with downregulating lentivirus (sh-circ_0006168) or control lentivirus (sh-NC) were subcutaneously injected into BALB/c nude mice (n = 5/group). The tumor volume was tested with a caliper every week and calculated as follows: 0.5 × length × width 2 . All mice would be sacrificed after injection for 5 weeks, and the tumor tissues were excised, weighed and harvested for detecting the abundance of circ_0006168, miR-194-5p and JMJD1C. This in vivo experiments were permitted by the Animal Care and Use Committee of Yantai Affiliated Hospital of Binzhou Medical University.

Statistical analysis
Experimental data were presented by mean ± standard deviation (SD). Every test was executed at least three times. GraphPad Prism 7.0 (GraphPad Software Inc, La Jolla, CA, USA) was used to display the analysis of results. Student's t-test and one-way analysis of variance (ANOVA) were performed to analyze significant differences between the groups. Statistical significance was considered when P < 0.05. *P < 0.05, **P < 0.001, ***P < 0.0001.

Circ_0006168 was upregulated in ESCC tissues and cells, and Taxol-resistant cells
To explore the potential roles of circ_0006168 in ESCC, the expression of circ_0006168 was detected by qRT-PCR in ESCC tissues and cells. Our study found that circ_0006168 expression in ESCC tissues was obviously higher than that in adjacent normal tissues (Fig. 1A). Likewise, compared with HET-1A cells, the expression of circ_0006168 level was also increased in ESCC cells (Eca109, KYSE450, KYSE150, and KYSE30), especially in Eca109 and KYSE150 cells (Fig. 1B). Thus, we chose Eca109 and KYSE150 cells for further experiments. Next, we tested the expression of circ_0006168 in parental cells and Taxol-resistant cells. As presented in Fig. 1C, circ_0006168 level in Eca109/Taxol and KYSE150/Taxol cells was higher than that in HET-1A cells and parental cells. Moreover, RNase R treatment assay showed that the expression level of CNOT6L mRNA was significantly reduced, while that of the circ_0006168 showed no obvious change (Fig. 1D), implying the cyclic structure of circ_0006168. These results suggested that circ_0006168 might play an important role in Taxol resistance.

Knockdown of circ_0006168 increased Taxol sensitivity in Taxol-resistant ESCC cells
To investigate the effects of circ_0006168 on Taxol sensitivity, Taxol-resistant cells (Eca109/Taxol and KYSE150/Taxol) were transfected with si-control or si-circ_0006168. Knockdown efficiency of circ_0006168 was detected by qRT-PCR. Transfection of si-circ_0006168 could specifically knock down the expression of circ_0006168 in Eca109/Taxol and KYSE150/ Taxol ( Fig. 2A). CCK-8 assay showed that cell viability and IC 50 value of Taxol were reduced by downregulating circ_0006168 in Eca109/Taxol and KYSE150/Taxol cells as well as parental cells (Eca109 and KYSE150) (Fig. 2B, Additional file 1: Fig. 1A-D). Next we explored Transwell assay was used to determine cell migration and invasion abilities (× 100). G Cell apoptosis was examined using flow cytometry analysis. *P < 0.05, **P < 0.001, ***P < 0.0001 the effect of circ_0006168 overexpression on Taxol toxicity in parental cells. Overexpression efficiency of circ_0006168 was presented in Additional file 2: Fig. 2A, B. We found that circ_0006168 overexpression increased the IC 50 value of Taxol in Eca109 and KYSE150 cells (Additional file 2: Fig. 2C-F). Meanwhile, CCK-8 assay and colony formation assay indicated that cell viability and colony-forming ability were was impaired by silencing circ_0006168 in Eca109/Taxol and KYSE150/Taxol cells (Fig. 2C, D), suggesting that circ_0006168 silence suppressed proliferation of Eca109/Taxol and KYSE150/ Taxol cells. Transwell assay demonstrated that interference of circ_0006168 could reduce migration and invasion of Eca109/Taxol and KYSE150/Taxol cells (Fig. 2E,  F). In addition, flow cytometry analysis indicated that circ_0006168 downregulation prominently increased apoptosis of Eca109/Taxol and KYSE150/Taxol cells (Fig. 2G). These data illustrated that circ_0006168 sensitized Eca109/Taxol and KYSE150/Taxol cells to Taxol.

Knockdown of JMJD1C enhanced Taxol sensitivity in Taxol-resistant cells
To investigate the potential roles of JMJD1C in Taxol resistance, the expression of JMJD1C was detected by qRT-PCR or western blot analysis. The results showed that JMJD1C mRNA expression was increased in ESCC tissues compared to normal tissues (Fig. 4A). Moreover, JMJD1C protein expression was enhanced in Eca109 and KYSE150 cells compared with HET-1A cells, and its expression was markedly higher in Eca109/Taxol and KYSE150/Taxol than that in HET-1A cells (Fig. 4A). JMJD1C protein expression was inhibited by transfection with si-JMJD1C, indicating a high transfection efficiency (Fig. 4B). CCK-8 results showed that JMJD1C knockdown decreased the IC 50 value of Taxol in Eca109/ Taxol and KYSE150/Taxol cells (Fig. 4C, D). Moreover, downregulation of JMJD1C repressed Eca109/Taxol and KYSE150/Taxol cell viability and colony formation ability (Fig. 4E, F). In addition, JMJD1C knockdown suppressed cell migration and invasion and increased apoptosis of Eca109/Taxol and KYSE150/Taxol cells (Fig. 4G-I). All the data suggested that JMJD1C had similar role with circ_0006168 in Taxol-resistant cells.

Silencing circ_0006168 promoted Taxol sensitivity in Taxol-resistant cells by downregulating JMJD1C
Rescue experiments were conducted to evaluate the regulatory effects of circ_0006168/JMJD1C axis on Taxol sensitivity. Western blot assay indicted that circ_0006168 knockdown reduced the protein expression of JMJD1C, which was rescued by upregulating JMJD1C (Fig. 5A). Moreover, the effect of circ_0006168 downregulation on reduction of IC 50 value of Taxol was abolished by elevating JMJD1C in Eca109/Taxol and KYSE150/Taxol cells (Fig. 5B, C). In addition, overexpression of JMJD1C reversed the inhibitory effects of circ_0006168 silence on cell viability, colony formation, migration, and invasion in Eca109/Taxol and KYSE150/ Taxol cells (Fig. 5D-G). Furthermore, JMJD1C upregulation attenuated si-circ_0006168-induced apoptosis in Eca109/Taxol and KYSE150/Taxol cells (Fig. 5H). These results indicated that circ_0006168 exerted its functions by regulating JMJD1C in ESCC cells.

Discussion
ESCC is a common malignant tumor of the digestive system [26]. Chemotherapy, including Taxol, is a promising treatment for patients with advanced ESCC who cannot undergo resection [22]. The development of chemoresistance has become a major obstacle limiting the efficacy of chemotherapy [27]. Growing evidence has certified that circRNAs function as crucial regulators in different cellular physiological processes [28,29]. Compared to miRNAs and lncRNAs, cir-cRNAs are highly abundant, stable, and specifically expressed [30,31]. Thus, circRNAs have become a hot topic in carcinoma research. Recently, although more and more reports have suggested that some circRNAs are involved in the progression of multiple cancers and chemoresistance [32,33], the functions of the majority of circRNAs remain unclear. Herein, we investigated the roles and regulatory mechanism of circ_0006168 in Taxol resistance of ESCC.
Previous studies have shown that some circRNAs are associated with the progression and development of ESCC [34][35][36]. More importantly, Xie et al. declared that circ_0006168 was overexpressed in EC tumors and cells, and inhibition of circ_0006168 attenuated cell growth migration, invasion, and glycolysis in EC via modulating miR-384/RBBP7 axis [37]. Moreover, Shi et al. The protein expression of JMJD1C was detected by western blot assay. C, D CCK-8 assay was employed to detect cell viability and IC 50 value of Taxol. E Cell viability was measured using CCK-8 assay. F Colony formation assay was utilized to assess cell colony formation ability. G, H The migration and invasion capacities were evaluated using transwell assay (× 100). I Flow cytometry analysis was utilized to determine cell apoptosis. *P < 0.05, **P < 0.001, ***P < 0.0001 also declared that high level of circ_0006168 level was positively correlated with TNM stage and lymph node metastasis in the patients with ESCC, and circ_0006168 facilitated the growth and metastasis of ESCC cells via sponging miR-100 and regulating mTOR [16]. Nevertheless, the influence of circ_0006168 on Taxol resistance of ESCC has not been clarified. In this research, we observed an obvious increase of circ_0006168 expression not only in ESCC tissues and cells but also in Taxolresistant cells. In vitro functional experiments revealed that circ_0006168 downregulation repressed cell proliferation, invasion and migration as well as accelerated apoptosis in Taxol-resistant cells. Taken together, these findings implicated that circ_0006168 promoted Taxol resistance in ESCC.
Mechanistically, circRNAs can exert their functions in human malignant tumors via competitively binding to miRNAs and regulating the expression of target genes [38]. For instance, hsa_circ_0030018 served as a miR-599 sponge to facilitate EC progression through upregulating ENAH expression [39]. CircRNA circ-Foxo3 inhibited ESCC progression via sponging miR-23a and upregulating PTEN [40]. In this research, we confirmed that miR-194-5p directly bound to circ_0006168, which was a key miRNA that targeted multiple genes. MiR-194-5p was reported to be involved in regulating chemoresistance, including Taxol resistance [22,41]. Previous studies have proven that miR-194-5p payed a tumor-suppressive role and downregulated in many cancers, including ESCC [21,42,43]. Consistent with these reports, we uncovered that miR-194-5p was obviously reduced in ESCC tissues, ESCC cells, and Taxol-resistant cells. Next, TargetScan database was utilized to search for the downstream targets of miR-194-5p. We revealed that JMJD1C acted as a downstream target of miR-194-5p, and circ_0006168 could positively modulate JMJD1C expression via sponging miR-194-5p. JMJD1C was reported to function as a tumor facilitator in some cancers by promoting cancer proliferation and inhibiting apoptosis [44,45]. Ek et al. demonstrated that JMJD1C might be associated Circ_0006168 regulated Taxol resistance by affecting JMJD1C expression. Eca109/Taxol and KYSE150/Taxol cells were transfected with si-control, si-circ_0006168, si-circ_0006168 + pcDNA, or si-circ_0006168 + pcDNA-JMJD1C. A Western blot assay was used to analyze the protein abundance of JMJD1C. B, C Cell viability and IC 50 value of Taxol were determined by CCK-8 analysis. D, E CCK-8 assay and colony formation assay were employed to evaluate cell proliferation. F, G Cell migration and invasion abilities were assessed by transwell assay. H Flow cytometry analysis was applied to detect cell apoptosis. *P < 0.05, **P < 0.001, ***P < 0.0001 with the risk of EC [46]. Cai et al. proved that JMJD1C level was positively correlated with the TNM stage, and knockdown of JMJD1C inhibited EC cell proliferation by regulating YAP1 signaling [24]. However, the biological functions of JMJD1C in Taxol resistance of ESCC have not been illustrated. In our study, we confirmed that JMJD1C showed higher level in ESCC tissues, ESCC cells, and Taxol-resistant cells. Moreover, JMJD1C interference inhibited cell growth, invasion and migration and facilitated apoptosis in Taxol-resistant cells, indicating that JMJD1C knockdown could attenuate Taxol resistance. Rescue assays proved that JMJD1C overexpression countervailed the promotive role of Taxol sensitivity caused by circ_0006168 knockdown, suggesting that circ_0006168 contributed to Taxol resistance of ESCC cells by upregulating JMJD1C. In addition, in vivo experiments revealed that circ_0006168 deficiency suppressed tumor growth by increasing miR-194-5p and decreasing JMJD1C, implying that circ_0006168 silence might improve Taxol sensitivity in vivo. Although circ_0006168 knockdown inhibits tumor growth in vivo, this effect does not necessarily translate into meaningful clinical benefits. Therefore, it is necessary to further study the role of circ_0006168 in clinical practice. Moreover, the cir-cRNAs/miRNAs/mRNAs regulatory networks are very complicated, thus, the detailed physiological mechanisms for the effects of circ_0006168 in ESCC and Taxol resistance need further exploration.
In conclusion, interference of circ_0006168 restrained cell proliferation, invasion and migration and accelerated apoptosis of Taxol-resistant cells in vitro, and inhibited tumor growth in vivo via sponging miR-194-5p and regulating JMJD1C (Fig. 7). It provided experimental basis for studying tumor resistance and further targeted therapy. Fig. 6 Circ_0006168 knockdown repressed tumor growth in vivo. Eca109/Taxol and KYSE150/Taxol cells transfected with sh-control or sh-circ_0006168 were injected into nude mice to establish mice xenograft model. A, B Tumor volume and weight were measured. C, D The expression levels of circ_0006168 and miR-194-5p were measured by qRT-PCR in the collected tissues. E Western blot assay was used to analyze the protein expression of JMJD1C in the collected tissues. **P < 0.001, ***P < 0.0001 Fig. 7 Schematic diagram of the mechanism by which the circ_0006168/miR-194-5p/JMJD1C regulates cell proliferation, migration, invasion, and Taxol resistance in Taxol-resistant ESCC cells