The human glioma U87MG (established by J. Ponten and identified by STR profiling) (cat. no. TcHu138), U251 (cat. no. TcHu58) and 293T (cat. no. GNHu17) were purchased from the Chinese Academy of Science (Shanghai, China). The human astrocyte (HA) cells (cat. no. 1800) were purchased from ScienCell Research Laboratories. These cells were cultured in Dulbecco's Modified Eagle Medium (DMEM) (Corning, Inc.) medium containing 10% fetal bovine serum (Procell Life Science & Technology Co., Ltd.) in a humidified incubator at 37°C with 5% CO₂, and the medium was replaced every 2–3 days or during cell passage until when the cell confluence reached 70–90%.

In the present study, whole-cell RNA was extracted using TRIzol^® reagent (Beijing Solarbio Science & Technology Co., Ltd.), RNA concentration was detected using a Nanodrop Spectrophotometer (Thermo Fisher Scientific, Inc.) at a wavelength of 260 nm, and intracellular RNA expression levels were measured using RT-qPCR. DANCR and GAPDH expression was analyzed using HiScript III RT Supermix for qPCR and ChamQ Universal SYBR qPCR Master mix (Vazyme Biotech Co., Ltd.) according to the manufacturer's instructions, Briefly, initial denaturation at 95°C for 30 sec was followed by 40 cycles of 95°C for 10 sec and 60°C for 30 sec. MiRNA-33b and U6 expression was quantified using miRNA 1st Strand cDNA Synthesis kit (stem-loop) and miRNA Universal SYBR qPCR Master mix (Vazyme Biotech Co., Ltd.) according to the manufacturer's instructions. MiRNA-33b primers were designed by software provided by the manufacturer of the miRNA reverse transcription kit, and the universal reverse primer in the kit was used. All RT-qPCR reactions were performed using the 7500 fast RT-PCR System (Thermo Fisher Scientific, Inc.). Finally, each expression was calculated through the 2^−ΔΔCq method (24). The primers used in the present study are listed in Table SI.

Before transfection, 5×10⁵ cells were seeded into 24-well plates, and 1 µg plasmid vector and Lipofectamine 3000 reagent (Thermo Fisher Scientific, Inc.) were mixed according to the manufacturer's instructions and added to the 24-well plates containing cells. After 24 h, the culture medium was replaced, and G418 (500 µg/ml) or puromycin (2 µg/ml) was added to screen the cells containing the plasmid vector, according to the transfection reagent instructions. A full-length DANCR plasmid was constructed with pcDNA 3.1. DANCR [small hairpin (sh-DANCR)] and DLX6 (sh-DLX6) silencing plasmids were constructed with pGPU6. MiR-33b overexpression (pre-mir-33b) and miR-33b interference (anti-miR-33b) plasmids were prepared by Shanghai GenePharma Co., Ltd. All plasmids and negative controls were purchased from Shanghai GenePharma Co., Ltd. The sequences used by the plasmids in the present study are shown in Table SII.

Cell proliferation assays were performed using the Cell Counting Kit-8 (CCK-8) assay (APeXBIO Technology LLC) in accordance with the manufacturer's instructions. U87MG or U251 cells were seeded in 96-well plates and 3.0×10³ cells were added to each well. To determine the OD value, the medium was replaced with 0.1 ml of fresh medium containing 0.01 ml of CCK-8 solution 3 h prior to detection at 24, 48 and 72 h. All of the aforementioned samples were placed in an incubator that contained 5% CO₂ and maintained at 37°C. Then, the absorbance value at 450 nm was detected with microplate reader (BioTek Instruments, Inc.).

Wild-type DANCR (DANCR-wt) and DLX6 (DLX6-wt) sequences containing miR-33b binding sites or mutant DANCR (DANCR-mut) and DLX6 (DLX6-mut) sequences without miR-33b binding sites were inserted into the vector. 293T cells were co-transfected with DANCR-wt + pre-miR-33b, DANCR-wt + pre-NC, DANCR-mut + pre-miR-33b, DANCR-NC + pre-NC, DLX6-wt + pre-miR-33b, DLX6-wt + pre-NC, DLX6-mut + pre-miR-33b, or DLX6-NC + pre-NC in 96-well plates using Lipofectamine 3000 reagent (Thermo Fisher Scientific, Inc.). After 48 h, the dual-Luciferase activity was detected using the Dual-Luciferase^® Reporter Assay System (Promega Corporation) according to the manufacturer's instructions. The Renilla luciferase activity was used as internal control.

RIPA buffer (Beyotime Institute of Biotechnology) supplemented with 1% protease inhibitor (added 2 min before use) was added to the cells and placed on ice for 30 min. Nucleic acids were crushed with an ultrasonic crusher and centrifuged at 13,800 × g at 4°C for 30 min. After detection of protein concentration using the bicinchoninic acid (BCA) assay, the loading buffer (5X) was added to the supernatant and heated at 100°C for 5 min. Proteins were separated using 12 and 10% SDS-PAGE and transferred to 0.22-µm polyvinylidene difluoride (PVDF) membranes (MilliporeSigma). The PVDF membranes were blocked by placing them in Tris-buffered saline containing 0.1% Tween 20 and 5% non-fat milk at room temperature for 120 min. The membranes were incubated with primary antibodies including DLX6 (1:2,000; cat. no. 23216-1-AP), LC3 (1:5,000; cat no. 14600-1-AP; both from ProteinTech Group, Inc.), p62 (1:1,000; cat. no. 23214; Cell Signaling Technology, Inc.), beclin-1 (1:8,000; cat. no. 66665-1-Ig), ATG7 (1:10,000; cat. no. 67341-1-Ig) and GAPDH (1:20,000; cat. no. 10494-1-AP; all from ProteinTech Group, Inc.) at 4°C for 14 h. The PVDF membrane was incubated with HRP-conjugated Affinipure Goat Anti-Mouse IgG (H+L) (1:10,000; cat. no. SA00001-1; ProteinTech Group, Inc.) or HRP-conjugated Affinipure Goat Anti-Rabbit IgG (H+L) (1:10,000; cat. no. SA00001-2; ProteinTech Group, Inc.) at room temperature for 2 h. The immunoblot bands were detected using an enhanced chemiluminescence detection kit (ECL kit; Beyotime Institute of Biotechnology), and images were captured using a chemiluminescence imaging analysis system (Bio-Rad Laboratories, Inc.). Finally, the ImageJ 1.53K software (National Institutes of Health) was used to calculate the relative gray value.

The required cells were collected according to the manufacturer's instructions (Dojindo Laboratories, Inc.). Briefly, 100 µl of 1X Annexin V binding buffer containing 5 µl of Annexin V-FITC and 5 µl PI was added to the cells, and incubation followed at room temperature for 15 min. Then, 400 µl of 1X Annexin V binding buffer was added to terminate the reaction and the assay was performed on Beckman DxFLEX (Beckman Coulter, Inc.). CytExpert1.1 software (Beckman Coulter, Inc.) was used for analysis. FITC^+/PI⁻ cells indicated early apoptosis, and FITC^+/PI⁺ cells indicated late apoptosis.

Cells were treated with trypsin and centrifuged into clumps, followed by electron microscope fixative (Wuhan Servicebio Technology Co., Ltd.) at room temperature for 30 min, then at 4°C overnight. Next, they were pre-embedded with agarose, post-fixed, and dehydrated at room temperature. This was followed by resin penetration and embedding using EMBed 812, polymerization, ultrathin sectioning and staining. Briefly, 2% uranium acetate saturated alcohol solution (avoiding light) was used for staining at room temperature for 8 min, then 2.6% Lead citrate (avoiding CO₂) was used for staining at room temperature for 8 min; subsequently, they were put into the grids board and dried overnight at room temperature. Finally, observation and image capturing via transmission electron microscopy was performed (Hitachi, Ltd.).

A ChIP assay was performed using a Simple Chip Enzymatic Chromatin IP kit (Cell Signaling Technology, Inc.), according to the manufacturer's instructions. Briefly, 1% formaldehyde was added to the cells at room temperature for 10 min to cross-link the protein to the DNA, which was then worked with an ultrasonic crusher (VCX-130, Sonics & Materials, Inc.) at 25% power for 15 sec and repeated 10 times to truncated into fragments of ~600 bp in length. The samples were incubated with anti-DLX6 rabbit antibodies (1:25; cat. no. sc-517058; Santa Cruz Biotechnology, Inc.) or normal rabbit IgG antibodies (1:250; cat. no. 30000-0-AP; ProteinTech Group, Inc.). After DNA cross-links were reversed, and the desired DNA fragment was purified, 2X Tap Master Mix for PAGE (Vazyme Biotech Co., Ltd.) was used to amplify the DNA fragment. The primers used for PCR were as follows: PCR1 forward, 5′-TTCAGCAAGCCACCTAAC-3′, and reverse, 5′-AACAACAAGAGCGATGACT-3′; and PCR2 forward, 5′-TGACAGAGCGAGACTCCGT-3′ and reverse, 5′-GGCTGGTCTTGAACTACTGGA-3′. Finally, agarose electrophoresis was used to verify the PCR products using 3% agarose gels and using SYBR Green staining.

First, glioma cells (U87MG) stably expressing sh-DANCR, pre-miR-33b, sh-DLX6 and sh-DANCR + pre-miR-33b + sh-DLX6 were established (n=6). BALB/c nude mice (female; 5 weeks-old; weight, 14–16 g) (Hua FuKang, Beijing, China) were used. Mice were housed under SPF conditions (The temperature of the SPF animal house was between 24–26°C, the humidity was ~70%, the air exchange rate was 15 times/h, and the light/dark cycle was 12/12 h) and the adequacy of food and water was checked daily to ensure that the mice had access to food and water ad libitum at all times. Each mouse was injected with 100 µl PBS containing 1×10⁶ cells subcutaneously under the right armpit. Tumor volumes were recorded on day 7 after injection and then measured every three days. At day 28, the mice were sacrificed by cervical dislocation after inhalation of 3% isoflurane anesthesia and death of mice was confirmed by respiratory arrest. Tumor weights were also compared. The present experiment was approved (approval no. 2022PS021K) by the Medical Ethics Committee of Shengjing Hospital (Shenyang, China).

After the tumor was removed and fixed with 4% paraformaldehyde at 4°C for 24 h, it was embedded in paraffin and cut into 5-µm thin sections. After dewaxing with xylene, rehydration (absolute ethanol, 95% ethanol, 90% ethanol, 80% ethanol and 70% ethanol were used for 10 min each in turn), antigen repair, incubation with anti-ATG7 antibodies (1:200; cat. no. 67341-1-Ig; ProteinTech Group, Inc.) at 4°C overnight, and incubation with biotin-labeled sheep anti-mouse/rabbit IgG polymer (cat. no. KIT-9710; Maxim Biotech, Inc.) at room temperature for 30 min, the tumor was finally stained with diaminobenzidine (Maxim Biotech, Inc.) for 5 min and images were captured under a light microscope (Nikon Corporation).

GEPIA (http://gepia.cancer-pku.cn/) was used to analyze the difference in DANCR levels between glioma and normal tissues with a P-value Cutoff of 0.01. The starbase website (https://starbase.sysu.edu.cn) was used to predict potential sites of connection between DANCR and miR-33b. The potential binding sites of DLX6 to the promoter binding region of ATG7 were predicted by JASPAR (https://jaspar.genereg.net).

All experiments were repeated at least three times. GraphPad Prism 8 software (GraphPad Software, Inc.) was used for all experimental data, and the results were presented as the mean ± SD. Unpaired student's t-test was used to conduct the comparison between two groups. More than two groups of data were analyzed through a one-way analysis of variance followed by LSD post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To identify the relationship between DANCR and gliomas, GEPIA (gepia.cancer-pku.cn) was used to preliminarily verify the DANCR expression level in glioma. A higher expression of DANCR was identified in glioma tissues compared with normal tissue (Fig. 1A). Then, the relative expression of DANCR was detected in HA, U251 and U87MG cells using RT-qPCR. It was revealed that the DANCR expression level was higher in the glioma cell lines U251 and U87MG than in HA cells (Fig. 1B).

To study the function of DANCR in glioma cells, DANCR-overexpressing U251 and U87MG cell lines were constructed via transfection of the pcDNA3.1-DANCR plasmid vectors into U251 and U87MG cells. Meanwhile, DANCR-silenced U251 and U87MG cell lines were constructed via transfection of the sh-DANCR plasmid vectors in U251 and U87MG cells. First, DANCR expression levels were in DANCR-overexpressing or DANCR-silenced U251 and U87MG cell lines using RT-qPCR to verify the effectiveness of the transfected plasmid vectors. Compared with the pcDNA3.1-NC group, DANCR expression increased significantly in U251 and U87MG cells transfected with pcDNA3.1-DANCR. Meanwhile, compared with the sh-NC group, DANCR expression in U251 and U87MG cells transfected with sh-DANCR decreased significantly (Fig. 2A and B). Then, the CCK-8 assay showed that upregulated DANCR expression could promote U251 and U87MG cell proliferation, while its inhibition slowed the rate of U251 and U87MG cell proliferation (Fig. 2C and D). Flow cytometric analysis demonstrated that the increased DANCR expression in U251 and U87MG cells reduced their apoptotic levels, while decreased DANCR expression exerted the opposite effect (Fig. 2E). Furthermore, the WB assay showed that the ATG7, beclin-1 and LC3 II/I protein levels increased and p62 levels decreased in DANCR-overexpressing U251 and U87MG cells. However, decreasing DANCR expression in U251 and U87MG cells decreased the ATG7, beclin-1 and LC3 II/I protein expression and increased p62 expression (Fig. 2F). Electron microscopic observation showed that the number of autophagosomes increased in DANCR-overexpressing U251 and U87MG cells compared with that in the control group. On the contrary, DANCR inhibition significantly reduced the number of autophagosomes in U251 and U87MG cells (Fig. 2G). In conclusion, DANCR can enhance glioma cell proliferation and autophagy and inhibit their apoptosis.

To explore how DANCR functions in glioma cells, a bioinformatics database was used (starbase.sysu.edu.cn). MiR-33b was a predicted target miRNA of DANCR (Fig. 3C). Then, the targeted binding and binding site of DANCR with miR-33b were verified through a dual-luciferase reporter assay. The relative luciferase intensity of cells in the experimental group transfected with DANCR-wt + pre-miR-33b was significantly lower than that in other groups (Fig. 3D). In the present study, as revealed in Fig. 3A, miR-33b expression in U251 and U87MG cells was downregulated, compared with that in HA cells. MiR-33b expression levels were then detected in miR-33b-overexpressing or miR-33b-silenced U251 and U87MG cell lines using RT-qPCR to verify the effectiveness of the transfected plasmid vectors. Compared with the pre-NC group, miR-33b expression increased significantly in U251 and U87MG cells transfected with pre-miR-33b. Meanwhile, compared with the anti-NC group, miR-33b expression in U251 and U87MG cells transfected with anti-miR-33b decreased significantly (Fig. 3B). Intracellular miR-33b expression levels were upregulated in U251 and U87MG cells and it was found that the cell proliferation slowed down (Fig. 3E and F), intracellular autophagy levels decreased, apoptotic levels increased (Fig. 3G and H), and miR-33b expression was inhibited, showing opposite effects. Hence, miR-33b may be a regulator of cell proliferation and autophagy.

It was explored whether miR-33b regulates the function of DANCR in glioma cells. Through co-transfection of the desired plasmid (pcDNA3.1-DANCR + pre-NC, pcDNA3.1-DANCR + pre-miR-33b, sh-DANCR + anti-NC, or sh-DANCR + anti-miR-33b) in U87MG and U251 cells, CCK-8, flow cytometry, and WB assays demonstrated that pre-miR-33b and anti-miR-33b could respectively inhibit the accelerated proliferation and decreased apoptosis or decreased proliferation and increased apoptosis caused by pcDNA3.1-DANCR or sh-DANCR. Furthermore, they respectively enhanced or slowed proliferation (Fig. 4A and B) and increased or decreased autophagy (Fig. 4C and D). These findings suggested that miR-33b is involved in regulating DANCR in glioma cells.

Next, to research the function of DLX6 in glioma cells, WB was used to verify its protein expression level in U251, U87MG and HA cells. DLX6 expression level in the glioma cell lines U251 and U87MG was significantly higher than that in HA cells (Fig. 5A). Then, a stable sh-DLX6-transfected cell line was established and its effectiveness was verified (Fig. 5E). CCK-8 and flow cytometry showed that cell proliferation slowed down and apoptosis increased (Fig. 5B-D). Meanwhile, the WB assay verified that inhibiting DLX6 expression also decreased cell autophagy levels (Fig. 5E). These results suggested that DLX6 knockdown slows down the progression of glioma cell malignancy.

Firstly, by altering the DANCR or miR-33b levels in U251 and U87MG cells, it was found that DANCR upregulation or miR-33b downregulation increased the expression level of DLX6, while DANCR downregulation or miR-33b upregulation decreased the expression level of DLX6 (Fig. 6A and B). Then, through co-transfection of the desired plasmid (pcDNA3.1-DANCR + pre-NC, pcDNA3.1-DANCR + pre-miR-33b, sh-DANCR + anti-NC, or sh-DANCR + anti-miR-33b) in U87MG and U251 cells, it was revealed that the increased expression of DLX6 induced by DANCR was reversed by pre-miR-33b, and the decreased expression of dlx6 induced by DANCR was also reversed by anti-miR-33b (Fig. 6C). To examine the relationship between DLX6 and miR-33b, a dual-luciferase reporter assay was first employed to verify the targeted binding effect between them. The relative luciferase intensity of cells in the experimental group transfected with DLX6-WT and pre-miR-33b was significantly lower than that in other groups (Fig. 6D and E). Then, U87 and U251 glioma cells were co-transfected with anti-miR-33b + sh-DLX6, and cells co-transfected with anti-miR-33b + sh-NC were set as a negative control. CCK-8 and flow cytometric assays showed that DLX6 knockdown could reverse the increase in the degree of malignant glioma cells caused by the decreased miR-33b expression (Fig. 6F-H). In addition, WB also verified that DLX6 knockdown reversed the increase in autophagy caused by decreased miR-33b expression (Fig. 6I). The aforementioned results demonstrated that DLX6 is involved in the regulation of miR-33b in the progression of glioma cell malignancy.

Jaspar predicted DLX6 as an ATG7 transcription factor, the predicted binding site was located between 2,000 base pairs (bp) upstream and 1,000 bp downstream of the transcription start site (TSS), and the control sequence was between 2,000 and 3,000 bp upstream of the TSS (Fig. 7A). A ChIP assay was used to verify whether DLX6 directly binds to the promoter sequence of ATG7 and regulates its transcription, and the same conclusion was obtained (Fig. 7B). Meanwhile, a WB assay verified that ATG7 expression level could be inhibited by DLX6 knockdown in U251 and U87MG cells (Fig. 5E). Thus, DLX6 is a transcription factor of ATG7 and promotes its expression.

To further confirm the aforementioned conclusion, in vivo experiments were conducted. The results showed that the tumor volumes in the sh-DANCR, pre-miR-33b, sh-DLX6 and sh-DANCR + pre-miR-33b + sh-DLX6 groups were smaller than that in the control group. The tumor volume was smallest in the group with DANCR and DLX6 knockdown combined with miR-33b overexpression (Fig. 8A-C). When ATG7 protein levels were measured in vitro via immunohistochemistry, it was found that they were lower in the sh-DANCR, pre-miR-33b, sh-DLX6 and sh-DANCR + pre-miR-33b + sh-DLX6 groups than in the control group, and ATG7 protein levels were the lowest in the sh-DANCR + pre-miR-33b + sh-DLX6 group (Fig. 8D).