The study concerning human tissues was authorized by the Ethics Committee of The Second Affiliated Hospital of Zhengzhou University (no. PY-2018092). Written informed consent was provided by all patients enrolled in the present study. A total of 45-paried glioma and normal specimens (collected at a distance from the tumor tissues) were collected from patients who underwent surgery at The Second Affiliated Hospital of Zhengzhou University from March 2017 to December 2018. Clinical features are presented in Table I. Tissue specimens were immediately fresh-frozen for subsequent experiments.

The Culture Collection of the Chinese Academy of Sciences (Shanghai, China) provided the human glioma cell lines (U87, A172, LN229 and U251) and normal human astrocytes (NHA). The U87 cell line we used was the U87 MG ATCC version (glioblastoma of unknown origin), and we authenticated the cell line using STR profiling. RPMI-1640 medium (Hyclone; GE Healthcare) containing 10% FBS was utilized to incubate the cells in a humidified incubator at 37°C and 5% CO₂.

Si-TTN-AS1, miR-27b-3p inhibitor, miR-27b-3p mimics and the corresponding negative controls were purchased from Sangon Biotech Co., Ltd. and the sequences are presented in Table II. 3′UTR (untranslated region) of RUNX1 was subcloned into the pGL3-control vector (Promega which contained the luciferase reporter. U251 and LN229 cells were plated in 96-well plates. When cells reached a confluence of 80%, they were transfected with the fragments or plasmids (0.2 µg/well) by Lipo3000 reagent (Invitrogen; Thermo Fisher Scientific, Inc.), according to the protocol. After incubation at 37°C for 24 h, the transfected cells were harvested for subsequent experiments.

Trizol (Invitrogen; Thermo Fisher Scientific, Inc.) was used to extract total RNA from the cells or tissues following the manufacturer's instructions. SuperScript VILO cDNA Kit (Thermo Fisher Scientific, Inc.) was applied to reversely transcribe RNA into cDNA. qPCR thermocycling conditions were as follows: 95°C for 1 min and 45 cycles of 94°C for 15 sec, 55°C for 20 sec, and 72°C for 30 sec. SYBR Green qPCR Master Mix (MedChenExpress) was used to carry out the quantitative PCR with specific primers (Table III) according to the 2^−ΔΔCq method (16).

A 96-well plate was taken to seed glioma cells (2×10⁵ cells/well) (U251 and LN229). After incubation for 24, 48, 72 and 96 h at 37°C, CCK-8 reagent (Tiangen, Hangzhou, China) was added 10 µl/well and cells were incubated for 2 h at 37°C. The absorbance at 450 nm was read by a microplate reader (Thermo Fisher Scientific, Inc.).

The Annexin V-FITC kit (Sigma-Aldrich; Merck KGaA) was used to estimate the apoptotic rate of the cells. In brief, cells (2×10⁵ cells/well) plated in 6-well plates were treated with Binding Buffer containing Annexin-V-FITC and propidium iodide (PI). After incubation in the dark for 15 min, a FACSCalibur flow cytometer (BD Biosciences) was applied to detect the apoptotic state of the cells.

Migration was tested by a wound healing assay. Transfected cells were plated in 12-well dishes (5×10⁴ cells/well), and incubated in RPMI-1640 medium (Hyclone; GE Healthcare) without FBS at 37°C, reaching a confluence of 80%. Then the cells were scratched across the surface of the well by a 10-µl pipette. After an incubation at 37°C of 24 h, the scratches were observed.

Transfected cells (2.5×10⁴ cells) were plated in the upper chamber of Transwell inserts (Corning, Inc.) which was coated with Matrigel (BD Biosciences). After an incubation of 24 h at 37°C, the cells invaded into the bottom chamber which was filled with medium containing 10% FBS. The invaded cells in the lower chamber were treated by methanol and 0.1% crystal violet at room temperature. A light microscope (X7, Nikon) was used to take photos of membrane. Five fields (magnification, ×100) of each sample were photographed randomly from each sample.

Plasmids comprising the predicted binding sites identified by TargetScan software (http://www.targetscan.org/vert_71/). Plasmids containing 3′UTR with wild-type sites or mutant sites were purchased from Promega. Plasmids containing the sequences of miR-27b-3p were purchased from Sangon Biotech Co., Ltd. miR-27b-3p mimics were co-transfected with TTN-AS1 wt, TTN-AS1 mt, RUNX1 wt or RUNX1 mut using Lipo3000. After transfection, the cells were incubated at 37°C for 48 h. Dual-Luciferase Reporter Assay System (Promega) was used to measure the luciferase activity. Renilla activity was used as a normalization control.

EZ-Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (Labbiotech) was used to carry out the RIP assay. Cells were incubated with RIP buffer containing beads coated with Ago2 antibodies or IgG antibodies (negative control) overnight. Immunoprecipitated complexes were collected for real-time PCR.

RIPA reagent buffer (Tiangen) was used to isolate the total protein from cells or tissues, followed by protein concentration determination with BCA protein assay kit (Beijing Solabio Life Sciences Co., Ltd.). SDS-PAGE (10%) was prepared and used to separate the different proteins (30 µg/lane). The separated protein blots were transferred onto PDFV membranes (Millipore). Silk milk (5%) was used to block the membranes at 37°C and then primary antibodies (anti-RUNX1; cat. Ab3692, dilution 1:500; Sigma-Aldrich; Merck KGaA; anti-GAPDH, dilution 1:2,000; KeyGen Biotech Co., Ltd.) were applied for an incubation of 12 h. Membranes were then treated with secondary antibodies (dilution 1:2,000; Keygen, Nanjing). Finally, protein signals were visualized by ECL detection kit (Beyotime Institute of Biotechnology).

Xenograft tumor tissues were fixed with 10% formaldehyde and sectioned into 5-µm-thick slides. Primary antibody against RUNX1 (cat. no. 2883, dilution 1:200; Cell Signaling Technology, Inc.) was used for incubation at 4°C overnight. Thereafter the slides were incubated with HRP-conjugated streptavidin for 1 h at room temperature. DAB chromogen (Promega) was used for visualization. Images were captured by a microscope (X7, Olympus) at ×100 magnification.

A total of 10 female BALB/c nude mice (6–8 weeks, ~20 g) were purchased from the Shanghai Laboratory Animal Center (Shanghai, China). Mice were housed and maintained under specific pathogen-free conditions at ~20°C, with 20% humidity, a 12 h light:12 h dark cycle, and with commercial rat food and water ad libitum. Mice were divided into 2 groups randomly (5 mice/group) and injected with U251 cells (5×10⁶) which were stably transfected with sh-TTN-AS1 or sh-NC. Tumor volumes were detected every week, according to the formula: V (mm³)=length (mm) × width² (mm²). Five week later, the mice were anesthetized by intraperitoneal injection of 10% chloral hydrate (400 mg/kg), and then euthanized by cervical dislocation. After death confirmation by cardiac arrest and pupil enlargement, the tumors were removed for tumor weight detection and tumor tissues were collected for subsequent experimentation. The protocol involved in the animal experiments was authorized by The Second Affiliated Hospital of Zhengzhou University.

Data analysis was performed by GraphPad Prism 6 (GraphPad Software, Inc). Results were presented as mean ± standard deviation (SD). One-way ANOVA was used to compare differences among multiple groups followed by Bonferroni post hoc test. The paired Student's t- test was used for assessing the TTN-AS1 level in 45 paired glioma and control tissues. The unpaired Student's t-test was applied for two-group comparison of the other assays. Survival analysis was performed using the Kaplan-Meier method with the log-rank test. Correlations were calculated using Spearman's correlation coefficient.

In order to investigate the role of TTN-AS1 in glioma, real-time PCR analysis was used and revealed the significantly high level of TTN-AS1 in human glioma specimens in comparison with the normal tissues (Fig. 1A). High levels of TTN-AS1 were positively correlated with advanced tumor stage of the glioma patients (Fig. 1B and C). RT-qPCR detection of TTN-AS1 revealed significantly higher levels of TTN-AS1 in the glioma cell lines (U87, A172, LN229 and U251), compared with that in normal astrocytes (NHA) (Fig. 1D). Relative TTN-AS1 expressions in glioma specimens were determined and normalized by adjacent normal tissue samples. If the relative level of TTN-AS1was ≥1, it was defined as high expression. If not, it was defined as low expression. Kaplan-Meier analysis demonstrated the survival curve, showing the poorer overall survival of the patients with high TTN-AS1 expression (Fig. 1E).

To investigate whether TTN-AS1 participates in tumor genesis of glioma, loss of function assays were applied to assess the cell behavior upon TTN-AS1 depletion. Fig. 2A shows the significant transfection efficiency. CCK-8 assay revealed that TTN-AS1 silencing reduced the proliferation of U251 and LN229 cells (Fig. 2B). Flow cytometry experiments demonstrated that the percentage of apoptotic cells was significantly elevated by TTN-AS1 downregulation (Fig. 2C). Wound healing and Transwell assays demonstrated the significantly decreased abilities of migration (Fig. 2D) and invasion (Fig. 2E) of the U251 and LN229 cell lines following silencing of TTN-AS1. These results led us to propose the oncogenic role of TTN-AS1 in glioma.

Research into the relevant mechanism was further performed. As lncRNAs function by acting as ceRNAs, we first explored the potential target miR-27b-3p by Starbase 3.0 (http://starbase.sysu.edu.cn/) (Fig. 3A) (17). Luciferase reporter activity confirmed the interaction between TTN-AS1 and miR-27b-3p (Fig. 3B). RT-qPCR was performed following si-TTN-AS1 or miR-27b-3p mimic transfection, showing that TTN-AS1 depletion significantly enhanced miR-27b-3p expression when compared to the si-ctrl group (Fig. 3C), whereas miR-27b-3p overexpression significantly reduced TTN-AS1 expression compared with the ctrl mimic group (Fig. 3D). RIP assay revealed that TTN-AS1 and miR-27b-3p were immunoprecipitated in the Ago2 complex (Fig. 3E). In the glioma tissues, miR-27b-3p was downregulated (Fig. 3F), and also had a negative correlation with TTN-AS1 (Fig. 3G).

To ascertain whether miR-27b-3p is involved in the TTN-AS1-regulated glioma cell proliferation, miR-27b-3p inhibitor or control (ctrl inhibitor) was co-transfected with si-TTN-AS1. Fig. 4A shows the transfection efficiency (Fig. 4A) in the U251 and LN229 cell lines. miR-27b-3p inhibitor reversed the reduced viability (Fig. 4B) and increased apoptosis (Fig. 4C) induced by si-TTN-AS1. In addition, the reduced abilities of migration (Fig. 4D) and invasion (Fig. 4E) were reversed by the miR-27b-3p inhibitor. The quantified data are presented in Fig. 4F.

As miRNAs are well known to exert functions by targeting downstream mRNAs, we investigated the potential target RUNX1 of miR-27b-3p by TargetScan (Fig. 5A). Luciferase reporter assay verified the potential interacting sites (Fig. 5B). Moreover, following transfection of the miR-27b-3p mimic in the U251 and LN229 cell lines, miR-27b-3p overexpression significantly suppressed the expression of RUNX1 both at the mRNA (Fig. 5C) and protein (Fig. 5D) levels. In the glioma tissues, RUNX1 was upregulated when compared with that in the normal tissues (Fig. 5E), which also had a negative correlation with miR-27b-3p (Fig. 5F).

Thereafter, we observed that silencing of TTN-AS1 could inhibit RUNX1 expression both at the mRNA and protein levels, however these alterations were attenuated by miR-27b-3p inhibitor (Fig. 6A-C). Moreover, in glioma tissues, RUNX1 had a positive correlation with TTN-AS1 (Fig. 6D).

Nude mice were injected with U251 cells which were stably transfected with sh-TTN-AS1 or sh-ctrl. Thereafter, tumor volumes and weight were analyzed, showing that the tumors of mice with TTN-AS1 knockdown were much smaller and lighter than those derived from the sh-ctrl transfected cells (Fig. 7A and B). Five weeks later, tissues were collected from the sacrificed mice, and miR-27b-3p and RUNX1 expression was estimated. As shown in Fig. 7C, miR-27b-3p was upregulated in the TTN-AS1-silenced cell derived mouse tumor tissues, compared with the sh-ctrl-transfected cell derived mouse tumors. On the contrary, RUNX1 expression was detected by western blot analysis (Fig. 7D) and ICH (Fig. 7E), demonstrating that RUNX1 was decreased in the glioma tissue derived from the TTN-AS1-knockdown cells than that from the sh-ctrl-derived tumor tissue.