Research subjects
From May 2009 to May 2013, Henan Key Laboratory for Esophageal Cancer Research of the First Affiliated Hospital admitted 98 patients with ESCC. From those patients, our study selected 52 cases (30 males and 22 females, 46 to 70 years, 56.4 ± 6.6 years) to be used as the research subjects in this study. Inclusion criteria: 1) no therapies received before treatment; 2) patients willing to join follow-up study (5-year). Exclusion criteria: 1) other medical conditions were observed; 2) any treatment received before admission; 3) history of previous malignancy. According to the staging criteria proposed by AJCC, patients were classified into stage I (n = 12), II (n = 16), III (n = 14) and IV (n = 10), respectively. Different treatments, such as esophagectomy, radio therapies, chemotherapies and the combination of them, were performed. Patients signed informed consent before admission. This study was approved by the aforementioned hospital Ethics Committee.
Specimen collection and cell lines
Fine needle aspiration was performed to collect ESCC and paired non-tumor tissue from all patients prior to therapies. Storage of tissue samples was performed at − 80 °C. Prior to therapy, blood (5 ml) was extracted after patients were fasted for 12 h. Plasma samples were collected after centrifuging (1200 g) blood samples in EDTA tubes for 15 min.
EC109 and KYSE150 cell lines were used in this study. EC109 and KYSE150 cells were purchased from ATCC (USA). FBS was added into RPMI-1640 medium to reach 10%, and the mixture was used as cell culture medium. Cells culture was performed at 37 °C with 5% CO2.
Follow-up
Patients were monitored by performing a follow-up study for 5-year. Through a monthly manner, patients were visited through telephone. All patients were excluded from deaths caused by factors unrelated to ESCC.
RT-qPCR
Following RNA isolations using RNAzol reagent, cDNAs were synthesized through reverse transcriptions using SS-IV-RT (Invitrogen). SYBR ® Green Master Mix (Toyobo, Japan) was used to perform qPCRs. The internal control of NCK1-AS1 and TGF-β1 mRNA was 18S rRNA. This expression was repeated 3 times and 2−ΔΔCT method was used for data normalizations.
Cell transient transfection
NCK1-AS1 or TGF-β1 expression vector was constructed by Sangon (Shanghai, China). Nucleofector™ Technology was used to achieve transient cell transfections with 10 nM vectors. Cells without transfection (control) and empty vector transfection (negative control) were included to serve as 2 controls. Subsequent experiments were performed at 24 h after transfections (for TGF-β inhibitor treatment cells were cultivated with medium containing TGF-β inhibitor SB431542 (SB, 10 nM, Sigma-Aldrich)).
Transwell assays
EC109 and KYSE150 cells were collected and 3 × 103 cells in 0.1 ml serum-free medium were transferred to upper chamber, while in lower chamber 20% FBS was added into medium to induce the movement of cells. Membranes were coated with Matrigel at 37 °C for 12 h before invasion assays, while migration assays were carried out using uncoated membranes. At 37 °C, cells were cultivated for 12 h, followed by 0.5% crystal violet (Sigma-Aldrich) staining for 15 min in dark. An optical microscope (Olympus, Japan) was used to count cells.
Western-blotting
RIPA (Invitrogen) was used to extrat total protein from EC109 and KYSE150 cells. After denaturing, 10% SDS-PAGE gel was used to separate proteins. After that, gel transfer to PVDF membranes. Blocking at room temperature in 5% non-fat milk was performed for 2 h. Membranes were then incubated with GAPDH (ab9485, 1: 1400, Abcam) and TGF-β1 (ab9758, 1:1600, Abcam) primary antibodies and goat anti-rabbit IgG-HRPs secondary antibody (1:1000, MBS435036, MyBioSource). ECL (Sigma-Aldrich, USA) was used for signal production and signals were processed using Image J v.1.46 software.
Statistical analysis
Gene expression levels were expressed by average values of three technical replicates, and paired t test was used for data comparison. ANOVA Tukey’s test was used to compare data of three independent replicates of multiple transfection groups, and data were expressed as mean+/- SD. Linear regression was used for correlation analyses. Patients were grouped into low (n = 28) and high (n = 24) plasma NCK1-AS1 level groups based on Youden’s index (cutoff value = 4.17). Survival curves were plotted and log-rank test was performed for survival curve comparison. Differences with p < 0.05 were statistically significant