All animal experiments were approved and conducted by the Animal Care and Ethics Committee of Lanzhou University Second Hospital.

Cell lines and cell culture

The U251 and SKMG were two kinds of human GBM cell lines which were purchased from the Cell Bank of Chinese Academy of Sciences. The corresponding TMZ-resistant cell lines, which were named as U251/TMZ and SKMG-1/TMZ, were established by continuous selection with increasing concentrations of TMZ for 6 months. The cells were routinely cultured in DMEM medium (HyClone, USA) containing 10% fetal bovine serum (GIBCO, USA) and 100 U/ml streptomycin-penicillin (HyClone, USA) at a 37 °C humidified environment with 5% CO2. Furthermore, U251/TMZ and SKMG-1/TMZ cells were maintained in the medium with 50 µg/ml TMZ (Sigma, USA) to obtain a stable phenotype of TMZ resistance.

Plasmid construction and cell transfection

The cDNA fragments with LINC00520 and/or LIN28B were synthesized and cloned into pcDNA3.1 vector (Invitrogen, USA). The short interfering RNAs (siRNA) for LINC00520 (si-LINC00520) and LIN28B (si-LIN28B) were designed to silence their expression in GBM cells (Shanghai, China), respectively. The transfection procedure was carried out following the manufacturer’s proposal. Briefly, GBM cells were inoculated into a six-well plate with a density of 5 × 105 cells/well. Cells were then transfected by Lipofectamine 3000 (Invitrogen, USA) with pcDNA3.1 control vector, pcDNA3.1-LINC00520 vector, pcDNA3.1-LIN28B vector, siRNA control vector, si-LINC00520 vector and si-LIN28B vector, respectively. The efficiency of cell transfection was detected by qRT-PCR assay.

RNA extraction and qRT-PCR

The extraction of total RNAs from TMZ-sensitive and TMZ-resistant GBM cells was performed using TRIzol Reagent (Invitrogen, USA). Then, 1 µg of extracted RNAs were used as a template to synthesize cDNA by means of a specific reverse transcription kit (Takara, China). The value of cycle threshold (Ct) was used to quantify the relative expression level of LINC00520, and the data was persented using.

the 2−ΔΔ CT method.

Cell counting kit (CCK-8)

The viability of GBM cells was measured through CCK-8 assay to investigate the effects of LINC00520 expression on TMZ chemoresistance. Briefly, the transfected TMZ-sensitive or TMZ-resistant cells were plated with 3000 cells per well. Each well was added 10 µL CCK-8 regent diluted with culture medium and was incubated for 2 h under 37 °C atmosphere with 5% CO2. The value of optical density under 490 nm was measured to detect the cell viability of GBM.

Colony formation assay

For colony formation assay, TMZ-sensitive or TMZ-resistant GBM cells were planked with 5000 cells per well. After adherence growth, GBM cells were treated with 50 µg/ml TMZ for 24 h, and then were incubated with TMZ-free medium for 14 days. 4% formaldehyde and 0.1% crystal violet was used to fix and stain cell colonies, respectively.

Flow cytometry analysis of cell apoptosis and cell cycle

After transfections, GBM cells and their TMZ-resistant cells were treated with 50 µg/ml TMZ for 48 h, and then were collected and digested with 0.25% trypsin. Next, these cells were resuspended by binding buffer at 4 °C and treated with 5 µL of FITC-labeled Annexin-V and propidium iodide (PI) reagent for 15 min under the dark. The apoptotic cells were immediately analyzed using flow cytometry (FACScan, USA). For the detection of cell cycle, GBM cells and their TMZ-resistant cells were additionally treated with serum starvation for 12 h to synchronize cell cycle. The cells were fixed with 70% ethanol at 4 °C overnight and stained with PI solution before flow cytometry of cell cycle.

TUNEL assay

The apoptosis of GBM cells treated with TMZ was further measured by TUNEL assay. In briefly, GBM cells were harvested and incubated with 0.1% Triton X-100 (Solarbio, China). Then, GBM cells were fixed with 4% paraformaldehyde and apoptotic cells were labeled by the TUNEL reagent containing the rTdT enzyme (Roche, Germany). In addition, Nuclei were stained with DAPI. The amount of staining-positive cells was counted under six random fields using a fluorescent microscope (Nikon, Japan).

Western blot

Radio-immunoprecipitation assay (RIPA) buffer with protease inhibitor was used to lyse GBM cells and extract total proteins, and their concentrations were quantified by bicinchonininc acid protein kit (Thermo, USA). The protein samples were electrophoretically separated by 10-12% SDS-PAGE gel and then were electrotransfered onto PVDF membranes (Millipore, USA). After sealing with 5% milk for 1 h, the membranes were incubated with primary antibodies against LIN28B (abcam, UK; ab229628, rabbit, polyclonal, 1:2500), LC3B (abcam, UK; ab48394, rabbit, polyclonal, 1:2000), Beclin-1 (abcam, UK; ab207612, rabbit, polyclonal, 1:2000) and β-actin (abcam, UK; ab8226, mouse monoclonal, 1:2000) under 4 °C overnight, respectively. Next, the membranes were incubated with the anti-rabbit or anti-mouse IgG HRP-labeled second antibody (1:2000) for 1 h at room temperature. The protein bands were generated by enhanced chemiluminescence (ECL) detection kit, and ECL detection system (GE Healthcare, Chicago, USA) was used for image observation.

Chromatin immunoprecipitation (ChIP)

ChIP assay was carried out by a ChIP Assay Kit (Beyotime, China). In brief, U251, SKMG-1 and their TMZ-resistant cells were collected and fixed with 1% formaldehyde for 20 min under room temperature. DNA fragments with 200–500 bp were harvested using sonication lysis. The antibody against STAT3 (anti-STAT3, abcam, UK; ab119352, rabbit, monoclonal) was used to precipitate the chromatin and nonspecific antibody against IgG (anti-IgG, Thermo Fisher Scientific, USA) was used as a negative control. After that, the immunoprecipitated DNA were detected by qRT-PCR assays.

Dual-luciferase reporter assay

The fragment sequence containing binding sites of STAT3 for the promoter regions of LINC00520 (Wild-type) were synthesized and cloned into luciferase reporter. The luciferase reporter with corresponding sites of mutant-type was used as a control. Then, U251, SKMG-1 and their TMZ-resistant cells were co-transfected with the luciferase reporters and STAT3-overexpressed vector or negative control vector. After transfection for 48 h, the luciferase activity was measured by Dual Luciferase Reporter Assay System (Promega, USA).

RNA immunoprecipitation (RIP) assay

The RNA-Binding Protein Immunoprecipitation Kit (Millipore, USA) was used in RIP assay, and all procedures were performed in line with the manufacturer’s proposals. The antibody against LIN28B (abcam, UK; ab229628, rabbit, polyclonal) and anti-IgG antibody (Thermo Fisher Scientific, USA) were incubated with cell lysates at 4 °C overnight. The RNA-protein complexes were collected and the relative expression level of immunoprecipitated RNAs was analyzed by qPCR assay.

Fluorescence in situ hybridization (FISH)

A total of five primary GBM and five recurrent GBM tissues were collected to detect the expression of LINC00520. All protocols were approved by the Ethics Review Board of our institution, and written informed consent for use of clinical sample was obtained from each subject. The FISH probe of LINC00520 was designed and synthesized by RiboBio (Guangzhou, China). In brief, the sections of fresh GBM tissues were fixed with 4% paraformaldehyde, washed by PBS for three times, and then were pretreated with prehybridization buffer at room temperature for 1 h. Next, the sections were hybridized using lncRNA FISH Probe Mix at 37 °C overnight. After hybridization, the expression of LINC00520 in primary and recurrent GBM tissues was detected using a fluorescent microscope (Olympus, Japan).

Immunofluorescence and immunohistochemistry

Briefly, GBM cells were treated with TSA and then were fixed with 4% paraformaldehyde for 15 min, permeabilized by 0.5% Triton X-100 and blocked with 5% Bovine serum for 1 h at room temperature. Subsequently, cells were incubated with primary antibody against LIN28B (abcam, UK; ab229628, rabbit, polyclonal, 1:500) at 4 °C overnight. After washing with PBS for three times, the cells were incubated with fluorescent-labeled secondary antibody in the dark for 1 h. The nuclei were stained with DAPI, and the images were photographed under a fluorescent microscope (Olympus, Japan). Furthermore, the expression of LIN28B in primary and recurrent GBM tissues was detected by immunohistochemistry staining.

Comet assay

GBM cells and their corresponding TMZ-resistant cells were treated with TMZ and then were subjected to comet assay to detect the DNA damage of these cells. All experiments were carried out according to the method described by previous studies [27, 28].

Tumor xenograft model

Four-to-six-week-old male BALB/c nude mice were used to build orthotopic xenograft model in vivo. A total of 1 × 106 U251/TMZ cells with stable LINC00520-silencing were injected into the right striatum of athymic BABL/c nude mice (n = 5). After 1 week, all nude mice in study group were treated with TMZ at the dose of 50 mg/kg for 5 days. The volume of tumor xenograft was measured each 7 days using Bioluminescence imaging system (IVIS Spectrum, USA).

Statistical analysis

At least three replicates were done for all experiments, and the data were shown as mean ± standard deviation (SD). For continuous variables, Student’s t-test or one-way ANOVA test was used to compare the statistical difference between the groups if appropriate. The data processing and statistical analysis were carried out using SPSS 23.0 version software (IBM Corp, Chicago, USA). A p-value < 0.05 was regarded statistically significant.