Patients and samples
The transcriptome data and clinical information of 321 patients from CGGA database (http://www.cgga.org.cn/) and 609 patients from TCGA database (https://tcga-data.nci.nih.gov/tcga/) was used. CGGA RNA sequencing database and microarray were used as validation set. 45 samples from Beijing Tiantan Hospital from August 2013 to December 2017 were used for further verification. The study was approved by the institutional review board of Beijing Tiantan Hospital affiliated to Capital Medical University (KY2019-143-02), and all patients or legal representatives signed an informed consent form.
45 formalin-fixed paraffin-embedded tissue blocks were sectioned. The slide was cut into 4-μm sections and incubated with anti-ARL4C (rabbit polyclonal, 1:300, ab122025, Abcam) primary antibodies. H-score was obtained by multiplying the staining intensity by a constant to adjust the mean to the strongest intensity [H-score = 3 × (percentage of strong staining)] (1.0%, weak; 2.0%, moderate; 3.0%, strong) to give a score ranging from 0 to 300.
Cell proliferation and colony formation experiment
U87-MG cells were adjusted to a density of 1×105 cells/ml. A total of 100 μl of the cell suspension was plated into each well of a 96-well plate and cultured overnight. After transient infection with 3μg vector or ARL4C or miRNA-654-3p plasmid on 1×106 cells using Lipo3000 for 24 h, 48 h, and 72 h, 20μl CCK8 solution was added to each well, and the cultures were further incubated for 4 h. Absorbance was measured at 450 nm using an ELISA plate reader (Thermo, USA).
1×103 U87-MG cells were seeded into a 6-well culture plate. After adherence, cells were transfected with ARL4C or miRNA-654-3p plasmid. Cells were maintained in a 37°C, 5% CO2 incubator for 2 weeks. To maintain transfection efficiency in the whole experiment, transfection was repeated at the 7th day after the first transfection. Cell colonies were stained with crystal violet (Zhongshan Company, Beijing, China) and photographed at the end of after 2 weeks of incubation.
Reverse transcription and quantitative PCR
Microarray hybridization and qRT-PCR were performed as previously described. Total RNA of 30 samples was extracted and purified using the Rneasy®Mini Kit (QiaGen, Hilden, Germany) following manufacturer’s instructions. RT-qPCR was performed on a QuantStudio5 (Applied biosystems, Singapore). The fold-change in differential expression for each gene was calculated using the comparative CT method (2−∆∆CT method) in R package with “PCR” functions (https://github.com/MahShaaban/pcr), a GAPDH reference gene, and the “1/2a” stage reference group. The primers of genes were listed as following: ARL4C F: atgcatttcccagaaccgtg, R：tctgatgacccaaaccacca; GAPDH F: aggtcggagtcaacggattt, R: tgacaagcttcccgttctca. miRNA-654-3p F: AGGTGACCAAGTTCGCCGAG, R: GACGTGATAGGTGGTGGCCG. U6 F: CTCGCTTCGGCAGCACA, R: AACGCTTCACGAATTTGCGT.
SDS-PAGE and western blot analyses
Samples (up to 10 mg) were lysed in lysis buffer containing 1% Nonidet P-40 (Calbiochem, Merk, Darmstadt, Germany) and protease and phosphatase inhibitor cocktails (Roche, IL, USA) overnight at 4 °C. Total extracts were centrifuged at 12, 000 g for 30 min at 4 °C, and protein concentration was determined with the BCA method (Pierce Biotechnology, IL, USA). A total of 40 μg of protein per lane was loaded onto 10% Bis-Tris SDS-PAGE gels, separated electrophoretically, and blotted onto polyvinylidene fluoride membranes (Merk). The blots were incubated with antibodies against anti-ARL4C (1:2,000) and anti-GAPDH (1:8000, ab181602) followed by a secondary antibody (1:10,000) tagged with horseradish peroxidase (Santa Cruz Biotechnology). Blots were visualized by enhanced chemiluminescence, and densitometry was performed using a fluorescence image analyzer (Amersham Imager 600, GE, MA, USA). GAPDH was used as a loading control.
Dual-luciferase reporter gene assay
The miRTarBase (http://mirtarbase.mbc.nctu.edu.tw/php/index.php) was used to predict the target mRNAs of miRNAs. ARL4C sequences were acquired from Ensembl (http://asia.ensembl.org/ index.html). Interacting miRNA and mRNA sequences were analyzed by Global Align in Blast (https://blast.ncbi.nlm.nih.gov). JAG1 sequences harboring either the wild-type (WT) or mutated binding sites (MT) for miR-654-3p were amplified by PCR and cloned into the pmirGLO luciferase vector (Promega, Madison, WI, USA) to create wild-type or mutant pmirGLO‐ARL4C vectors, respectively, using the Lipofectamine 3000 transfection kit (Promega, Madison, WI, USA). U87-MG and C6 cells harboring the wild-type or mutant pmirGLO‐ARL4C were co‐transfected with the mimics or miR‐NC, respectively, for 48 h. After transfection, luciferase reporter gene activity was determined following the instructions of the dual-luciferase detection kit (Promega, Madison, WI, USA).
Receiver operating characteristic (ROC) analysis
ROC curve analysis was performed using the ROCR package of the free software R (https://www.r-project.org/). The 95% confidence interval for area under the curve (AUC) was calculated according to Hanley and McNeil.
All statistical analyses were conducted using SPSS Statistics Version22 (IBM Corporation, Armonk, New York, USA). An unpaired Student’s test and a chi-square (Fisher’s exact) test were used to compare quantitative and qualitative data. P-value of less than 0.05 was considered significant.