The parental cisplatin-sensitive cells lines KYSE30 and ECA109 and the corresponding cisplatin-resistant cell lines KYSE30/DDP and ECA109/DDP were purchased from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences. The cells after resuscitation were cultured in high-glucose DMEM containing 10% FBS (both Sigma-Aldrich; Merck KGaA) and incubated at 37°C with 5% CO₂.

Three pairs of pathologically confirmed cisplatin-resistant and adjacent cisplatin-sensitive tissue were collected from patients with esophageal cancer at the Department of Thoracic Surgery of The First Affiliated Hospital of China Medical University between May 2017 to March 2018. The clinicopathological characteristics of the three patients are presented in Table SI. The present study was approved by the Ethics Committee of The First Affiliated Hospital of China Medical University and performed in accordance with the Declaration of Helsinki (10). Informed consent was obtained from all patients.

The circRNA expression profiles of the three pairs of cisplatin-resistant and corresponding adjacent tissues were evaluated using a circRNA chip assay (ArrayStar Co., Ltd.). The assay was performed and analyzed by Shanghai GeneChem Co., Ltd.

Total RNA was extracted using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) and reversed transcribed into cDNA using a riboSCRIPT™ reverse transcription kit (Guangzhou RiboBio Co., Ltd.). RT was conducted at 42°C for 60 min and 70°C for 10 min. The qPCR primers were designed using Primer Premier software (version 5; Premier Biosoft International) and synthesized by Shanghai GeneChem Co., Ltd. The following primers were used: circRNA_001275 forward, 5'-TCT TCT TCT CCA CTC CTG AA-3' and reverse, 5'-GAG CAA GGG CCC TAG CTC AA-3'; miR-370-3p forward, 5'-TGT AAC CAG AGA GCG GGA TGT-3' and reverse, 5'-TTT TGG CAT AAC TAA GGC CGA A-3'; Wnt7a forward, 5'-CTC CGG ATC GGT GGC TAT G-3' and reverse, 5'-CCC ATT TGT GAG CCT TCT CCT-3'; U6 forward, 5'-CTC GCT TCG GCA GCA CA-3' and reverse 5'-AAC GCT TCA CGA ATT TGC GT-3'; and GAPDH forward, 5'-CGT CAC CAA CTG GGA CGA CA-3' and reverse, 5'-CTT CTC GCG GTT GGC CTT GG-3'. The qPCR thermocycling conditions were as follows: 94°C for 15 sec, then 55°C for 30 sec and 70°C for 30 sec for 40 cycles. mRNA levels were quantified using the 2^-Δ∆Cq method (11); circRNA_001275 and Wnt7a were normalized to GAPDH, whereas miR-370-3p was normalized to U6.

The circRNA_001275 overexpression (OE) vector (circRNA_001275 OE), pcDNA3.1 empty vector, small interfering (si)RNA targeting circRNA_001275 (si-circRNA_001275), si-negative control (NC), miR-370-3p mimic, miR-370-3p inhibitor and miR-370-3p NC labeled with green fluorescent protein were designed and synthesized by Shanghai GeneChem Co., Ltd. Sequences were as follows: si-circRNA_001275, 5'-GTT GAA GGG GGA GCT CCT GTC ATA AAA GCC AA-3'; si-NC, 5'-GAG CCC CAG CCT TCT CCA TG-3'. miR-370-3p mimic, 5'-GCC TGC TGG GGT GGA ACC TGG T-3'; miR-370-3p inhibitor, 5'-CAG GUC ACG UCU CUG CAG UUA C-3'; and miR-370-3p NC, 5-GCC TGC TGG GGT GGA ACC TCC T-3'. Cells (1x10⁵/well) were transfected with 50 nM RNA or plasmid using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). At 48 h later, the transfection efficiency was assessed by RT-qPCR. The circRNA_001275 OE vector or pcDNA3.1 empty vector were transfected into KYSE30/DDP cells, whereas si-circRNA_001275 or si-NC were transfected into ECA109/DDP cells.

Transfected cells in the logarithmic growth phase were digested with trypsin at room temperature for 5 sec and seeded into 96-well plates at a density of 1x10⁴ cells/well. A total of 20 µl MTT solution (2.5 mg/ml) was added to the cells at 0, 12, 24, 48 and 72 h. The cells were subsequently incubated for an additional 4 h at 37°C with 5% CO₂. The purple formazan crystals were dissolved by the addition of 100 µl/well DMSO (Guangzhou RiboBio Co., Ltd.) and the absorbance was measured at a wavelength of 490 nm.

Transfected cells in the logarithmic growth phase were digested with trypsin and seeded into 6-well plates at a density of 5x10⁵ cells/well. Following incubation for 24 h at 37°C, the cells were fixed with 3% paraformaldehyde for 5 min at 37°C and washed twice with PBS. The cells were stained with 10 µl Hoechst 33258 for 15 min in the dark at 37°C and subsequently sealed with neutral balsam. Apoptotic cells in five randomly selected fields imaged under a fluorescence microscope (magnification, x400).

The 3'-untranslated region (3'-UTR) of circRNA_001275 and Wnt7a were cloned into the XbaI site of the pGL3 control vector (Promega Corporation). Cells were co-transfected with the miR-370-3p mimic or miR-370-3p NC along with wild-type (wt) or mutated (mut) circRNA_001275-3'-UTR or Wnt7a-3'-UTR using Lipofectamine 2000 at a final concentration of 50 nM for each RNA/plasmid. Firefly and Renilla luciferase activity was detected using a dual-luciferase reporter assay system (Promega Corporation). Firefly luciferase activity was normalized to Renilla luciferase activity (Promega Corporation). The luciferase assay was performed by Shanghai GeneChem Co., Ltd.

Target genes were predicted using miRNA target gene prediction databases. miRanda v5 (http://www.microrna.org/microrna/home.do), TargetScan v7.1 (http://www.targetscan.org) and miRBase (http://www.mirbase.org/). Genes which appeared in >2 databases simultaneously were predicted to be target genes.

Transfected cells in the logarithmic growth phase were digested, and total protein was collected using RIPA buffer (Beijing Solarbio Science &Technology Co., Ltd.) with 1 mM phenylmethylsulfonyl fluoride and subsequently quantified using a bicinchoninic acid protein assay kit. Total protein (50 µg/lane) was separated via 10% SDS-PAGE and then transferred onto PVDF membranes. The membranes were blocked for 1 h in blocking buffer with 5% non-fat milk at room temperature. Subsequently, the PVDF membranes were incubated with primary antibodies against Wnt7a (1:1,000; cat. no. sc-365665; Santa Cruz Biotechnology, Inc.) and b-actin (1:1,000; cat. no. sc-69879; Santa Cruz Biotechnology, Inc.) overnight at 4°C. Following primary antibody incubation, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody (1:1,500; cat. no. sc-516102; Santa Cruz Biotechnology, Inc.). The membranes were washed twice with PBS and protein bands were visualized using a ECL-PLUS kit (GE Healthcare Life Sciences). Densitometry analysis was performed using ImageJ software v1.5 (National Institutes of Health) with β-actin as the loading control.

Statistical analyses were performed using SPSS software (version 17; SPSS, Inc.). Data are presented as the mean ± SD of three experimental repeats. Student's t-test was used to analyze the differences between two groups. One-way ANOVA followed by Tukey's multiple comparison test was used to analyze differences among three groups. P<0.05 was considered to indicate a statistically significant difference.

circRNA expression profiles were evaluated using a circRNA chip assay. Polyacrylamide gel electrophoresis revealed that the total RNA obtained from the three pairs of tissues was of high purity and undegraded (Fig. 1A). The circRNA microarray hybridization signal diagram of the three tissue pairs acquired by the circRNA chip scanner is presented in Fig. 1B. The box plot (Fig. 1C), volcano plot (Fig. 2A) and scatter plot (Fig. 2B) revealed variable circRNA expression.

The cluster heat map (Fig. 3A) revealed the top ten most upregulated and downregulated circRNAs. circRNA_001275 was the most highly upregulated (14.85-fold) circRNA in cisplatin-resistant tissues with highest raw intensity. Therefore, circRNA_001275 was selected as the primary focus of the study. The circRNA chip results were validated via RT-qPCR; circRNA_001275 was upregulated in cisplatin-resistant tissues (Fig. 3B), as well as KYSE30/DDP and ECA109/DDP cells (P<0.05; Fig. 3C), compared with corresponding adjacent tissues and sensitive cells.

Fluorescence micrographs were captured at 24 and 48 h post-transfection. KYSE30/DDP and ECA109/DDP cells appeared green under a fluorescence microscope (Fig. 4A). RT-qPCR validation demonstrated that circRNA_141539 expression was significantly increased in the circRNA_001275 OE group, and significantly reduced in the si-circRNA_001275 group, compared with the corresponding control groups (both P<0.05; Fig. 4B).

MTT and EdU assays were used to detect the effects of circRNA_001275 on cell proliferation. The results showed that cell viability (P<0.05; Fig. 5A) and the ratio of EdU-positive cells (P<0.05; Fig. 5B) were significantly increased in the circRNA_001275 OE group, and significantly decreased in the si-circRNA_001275 group (both P<0.05), compared with the corresponding control groups.

Transwell invasion assays and Hoechst 33258 staining were used to detect the effects of circRNA_001275 on cell invasion and apoptosis, respectively. The apoptosis rate was significantly decreased in the circRNA_001275 OE group, and significantly increased in the si-circRNA_001275 group, compared with the corresponding control group (both P<0.05; Fig. 6).

A schematic representation of the potential binding sites between miR-370-3p and the circRNA_001275-3'-UTR is presented in Fig. 7A. miR-370-3p NC, mimic or inhibitor were transfected into KYSE30/DDP and ECA109/DDP cells. RT-qPCR results confirmed that miR-370-3p were significantly increased in the miR-370-3p mimic group, and significantly reduced in miR-370-3p inhibitor group, compared with the miR-370-3p NC group (both P<0.05; Fig. S1). A luciferase assay revealed that the relative luciferase activity in the circRNA_001275-wt and miR-370-3p mimic co-transfection group was significantly decreased compared with the circRNA_001275-mut or miR-370-3p NC groups (both P<0.05; Fig. 7B). RT-qPCR revealed that miR-370-3p was downregulated in cisplatin-resistant cells compared with corresponding sensitive cells (P<0.05; Fig. 7C); additionally, miR-370-3p levels were markedly decreased in cisplatin-resistant tissues compared with adjacent tissues. Furthermore, the expression of miR-370-3p was significantly decreased in the circRNA_001275 OE group, and significantly increased in the si-circRNA_001275 group, compared with the corresponding control groups (both P<0.05; Fig. 7D). Therefore, the results demonstrated that circRNA_001275 may serve as a sponge for miR-370-3p.

The potential binding sites of miR-370-3p with Wnt7a-wt and Wnt7a-mut were predicted. The results indicated a potential binding site for miR-370-3p in the Wnt7a 3'-UTR (Fig. 8A). A luciferase assay revealed that the relative luciferase activity in the Wnt7a-3'-UTR-wt and miR-370-3p mimic co-transfection group was significantly decreased compared with the miR-370-3p NC or Wnt7a-3'-UTR-mut groups (both P<0.05; Fig. 8B). RT-qPCR revealed that Wnt7a mRNA was significantly upregulated in cisplatin-resistant cells compared with corresponding sensitive cells (both P<0.05; Fig. 8C); additionally, Wnt7a levels were markedly increased in cisplatin-resistant tissues compared with adjacent tissues. Furthermore, Wnt7a mRNA was significantly decreased in the miR-370-3p mimic group, and significantly increased in the miR-370-3p inhibitor group, compared with the miR-370-3p NC group (both P<0.05; Fig. 8D). Additionally, the Wnt7a protein level was significantly decreased in the miR-370-3p mimic group, and significantly increased in the miR-370-3p inhibitor group, compared with the miR-370-3p NC group (both P<0.05; Fig. 9A and B). Therefore, the results revealed that Wnt7a is a direct target gene of miR-370-3p.

The effects of the circRNA_001275/miR-370-3p axis on Wnt7a expression were investigated. RT-qPCR revealed that Wnt7a mRNA expression was significantly increased in the circRNA_001275 OE group and significantly decreased in the si-circRNA_001275 group; however, these effects were attenuated by miR-370-3p mimic and inhibitor, respectively (P<0.05; Fig. 10A). Furthermore, MTT results showed that overexpression of circRNA_141539 increased cell viability, and that circRNA_001275 silencing decreased cell viability; these effects were abrogated by miR-370-3p mimic and inhibitor, respectively (P<0.05; Fig. 10B).