Abstract Gliomas are the most fatal malignant cerebral tumors. Temozolomide (TMZ), as the primary chemotherapy drug, has been widely used in clinics. However, resistance of TMZ still remains to poor defined. LncRNAs have been reported to play crucial roles in progression of various cancers and resistance of multiple drugs. However, the biological function and underlying mechanisms of most lncRNAs in glioma still remains unclear. Based on the TCGA database, a total of 94 differentially expressed lncRNAs, including 16 up-regulated genes and 78 downregulated genes were identified between gliomas and normal brain tissues. Subsequently, lncRNA DLEU1, HOTAIR, and LOC00132111 were tested to be significantly related to overall survival (OS) between high- and low-expression groups. Additionally, we verified that lncRNA DLEU1 was high expressed in 108 gliomas, compared with 19 normal brain tissues. And high expression of lncRNA DLEU1 predicted a poor prognosis (HR = 1.703, 95%CI: 1.133–2.917, p-value = 0.0159). Moreover, functional assays revealed that knockdown of lncRNA DLEU1 could suppress the proliferation by inducing cell cycle arrest at G1 phase and reducing the S phase by down-regulating the CyclinD1 and p-AKT, as the well as migration and invasion by inhibiting the epithelial–mesenchymal transition (EMT) markers, such as ZEB1, N-cadherin, β-catenin and snail in glioma cells. Furthermore, silencing lncRNA DLEU1 suppressed TMZ-activated autophagy via regulating the expression of P62 and LC3, and promoted sensitivity of glioma cells to TMZ by triggering apoptosis. Conclusively, our study indicated that lncRNA DLEU1 might perform as a prognostic potential target and underlying therapeutic target for sensitivity of glioma to TMZ. Keywords: LncRNA DLEU1, Glioma, Temozolomide, Autophagy, Epithelial-Mesenchymal Transition Introduction Glioma, accounting for approximately 80% of malignant tumors, is one of the most primary and fatal intracranial tumors ([44]Jin et al., 2019; [45]Wu et al., 2019). Despite advances in diagnosis and therapy such as surgical resection followed by adjuvant radiotherapy and chemotherapy, patients with glioblastoma have a median survival time of merely 12–15 months [46]Parsons et al., 2008; [47]Mostafa et al., 2016). Temozolomide (TMZ), the primary chemotherapeutic drug for gliomas, was reported to prolong the survival time of glioblastoma patients ([48]Nanegrungsunk et al., 2015; [49]Wait et al., 2015). Nevertheless, TMZ resistance mechanisms were still ubiquitous. Although several biomarkers such as MGMT, STAT3, and APNG have been explored to be associated with the sensitivity of TMZ ([50]Jacinto and Esteller, 2007; [51]Agnihotri et al., 2012; [52]Lee et al., 2011), TMZ resistance in gliomas is still incompletely elaborated. Thus, the research for novel chemotherapeutic targets is crucial for glioma therapy. Long non-coding RNAs (lncRNAs) were defined as long RNA transcripts (>200 nucleotides) with no protein-coding capability. However, a few articles discovered the peptide-coding ability of certain lncRNAs ([53]Necsulea et al., 2014; [54]Quinn and Chang, 2016). Recently, increasing studies have revealed that lncRNA plays an important and various role in progression of glioma ([55]Lv et al., 2016a; [56]Chen H. et al., 2019; [57]Li C. et al., 2019). For example, lncRNA ZEB1-AS1, which serves as an oncogene, could promote tumorigenesis of gliomas by activating epithelial-to-mesenchymal transition (EMT), such as ZEB1, N-cadherin, and MMP2 protein markers ([58]Lv et al., 2016a). On the contrary, lncRNA CPS1-IT1 was discovered to be an anti-oncogene gene in glioma ([59]Chen H. et al., 2019). Moreover, linc00467 and LncRNA HANR were found to aggravate the malignant progression and promote invasion and proliferation of glioma cells by targeting miRNA-485-5p and miRNA-335, respectively ([60]Jiang and Liu, 2020; [61]Wang et al., 2020). These findings implied that more detailed mechanisms of lncRNA still remained unclear. It is important to illuminate the potential molecular mechanisms and explore prognostic biomarkers to help progress of therapeutic targets and strategies for glioma. The Cancer Genome Atlas (TCGA), a landmark cancer genomics program, is widely used in the genetic research of cancer. Based on RNA-seq data from TCGA, we demonstrated that the differential expression gene in glioma, named lncRNA deleted in lymphocytic leukemia 1 (lncRNA DLEU1), was dramatically associated with a poor prognosis. Emerging research has reported that lncRNA DLEU1 performed an indispensable role in the genesis and progression of various tumors such as gastric cancer, osteosarcoma, pancreatic ductal adenocarcinoma, non-small cell lung cancer, and breast cancer, as well as resistance to chemotherapy in tumors ([62]Li X. et al., 2018; [63]Chen X. et al., 2019; [64]Gao et al., 2019; [65]Zhang et al., 2019; [66]Wang C. et al., 2019; [67]Li Y. et al., 2019). However, the progression and TMZ chemosensitivity of lncRNA DLEU1 in glioma are still vague. In the current study, we investigated the differential expression of lncRNA DLEU1 between 108 gliomas and 19 normal cerebral trauma tissues as well as the prognostic value. In addition, we examined the potential molecular mechanisms of lncRNA DLEU1 in the regulation of proliferation, migration, and invasion in U87 and U251 cells. Furthermore, we revealed the underlying mechanism of lncRN ADLEU1 involved in sensitivity of glioma cells to TMZ, which provided a potential target for the individualized therapy of temozolomide in glioma. Materials and Methods Patients Samples and Follow-Up A total of 108 human glioma tissues and clinical information were acquired from patients who underwent surgical resection before treatment with radiation and chemotherapy between November 2010 and June 2013 in the Department of Neurosurgery, Xiangya Hospital (Hunan, China). Nineteen normal brain specimens obtained from patients with cerebral trauma surgery were normalized as controls. Informed consents were provided to all patients or their family members. All participants signed the written informed contents. This study was approved by the Hospital Ethics Committee (No. 201803806). Microarray Analysis Raw RNA-seq data (count files) and clinical information were obtained from the TCGA database ([68]https://tcga-data.nci.nih.gov/tcga/). Then the expression value of lncRNA was flited from the raw files. The differential expressed genes were identified by DEseq with a threshold: adjusted-p < 0.05, log2FoldChange > 1. All p-values were adjusted by the Benjamini–Hochberg’s method. Then the association between the expression of differentially expressed lncRNAs and overall survival was evaluated by univariate Cox proportional hazards regression analysis using the survival R package of R 3.6.1 ([69]https://www.r-project.org/). LncRNAs with p-value<0.05 were considered as candidate variables. The log-rank test was performed to evaluate the prognostic value of lncRNAs. Functional Enrichment Analysis To identify potential biological processes and pathways that the filtrated lncRNAs were involved in, functional enrichment analysis was performed. First, we integrated differential lncRNAs’ neighboring (10 Kb) mRNAs as underlying target genes (TGs). Differentially expressed lncRNAs and mRNA sequences were extracted and primed by blast (e < 1E-5), and then screened again by the software [RNAplex (G < −20)] to identify the possible target genes of lncRNA. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was carried out for those target genes using the database ([70]http://www.genome.jp/). Gene ontology (GO) was confirmed by hypergeometric distribution. The p-value <0.05 was set as the cutoff criterion for both GO and KEGG functional analysis. Cell Lines and Materials The human glioma cell lines (U87 and U251) were acquired from the American Type Culture Collection (ATCC) and kept in Dulbecco’s modified Eagle’s medium (DMEM, HyClone, United States) supplemented with 10% fetal bovine serum (FBS, GIBCO, Carlsbad, California, United States) and in a 5% CO2 humidified incubator at 37°C. TMZ was purchased from Sigma- Aldrich (St. Louis, Missouri, United States). Cell Transfection Two different small interfering RNA against lncRNA DLEU1 (siRNA-DLEU1) and negative control siRNA (NC) were synthesized by RiboBio (Guangzhou, China). Cells that were growing exponentially were seeded in six- well plates and cultured overnight. When the plated cells arrived at approximately 50–60% confluence, the siRNAs were transiently transfected into U87 and U251 cells using Lipofectamine® RNAiMAX Transfection Reagent (Invitrogen, Carlsbad, California, United States) according to the manufacturer’s directions. RNA Extraction and Real-Time Polymerase Chain Reaction Total RNA was isolated from cultured cells or tissues samples using TRIzol reagent (Invitrogen, Carlsbad, California, United States). The cDNA was generated by reverse transcription of 1 µg of total RNA using Thermo Scientific Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Wilmington, Delaware, United States). According to the manufacturer’s instructions, real-time PCR was conducted by using the SYBR Green Real-Time PCR Kit (Takara, Dalian, China) on a LightCycler480 (Roche, San Francisco, California, United States) to detect the expression of lncRNA DLEU1, with GAPDH as a normalizing control. PCR was performed at the following conditions: 95.0°C for 3 min, and 40 circles of 95.0°C for 10 s and 60°C for 30 s. The relative quantitative value was evaluated by the 2^−ΔΔCT method. The primers for lncRNA DLEU1 were F: 5′-GCG​GAG​GTG​AAG​TGA​ACT​TAG​A-3′; and R: 5′-CTC​CTA​AGC​AGG​ACC​CGT​ATT-3′. The primers for GAPDH were F: 5′-CCC​ATC​ACC​ATC​TTC​CAG​GAG-3′; and R: 5′-GTT​GTC​ATG​GAT​GAC​CTT​GGC-3′. Western Blotting Analysis The proteins were extracted and measured from cultured cells as previously described ([71]Lv et al., 2016b). Anti-AKT, anti-p-AKT, anti-N-cadherin, anti-Snail, anti-P62, anti-LC3, anti-GAPDH and anti-β-actin antibodies were purchased from CST Biotechnology (Boston, Massachusetts, United States). The relative protein expression was analyzed based on gray value. Each test was repeated three times. The β-actin and GAPDH were used as the internal references.