Glioma, classically classified in four WHO grades, is the most common malignant neoplasm among primary central system tumors with high mortality (1). A higher glioma grade is associated with a poorer patient prognosis. Recent years, based on molecular markers, such as IDH mutation, 1p/19q codeletion, Histone H3 K27M mutation, Histone H3.3 G34R/V mutation, MGMT et.al, diffuse gliomas were divided into oligodendroglioma IDH-mutation and 1p/19q codeletion, WHO grade II-III; Astrocytoma, IDH-mutation, WHO grade II-III; Astrocytoma, IDH-mutation, WHO grade IV; Glioblastomas, IDH wild type, MGMT promoter methylation, WHO grade IV; Diffuse hemispheric gliomas, H3.3 G34-mutant, MGMT promoter methylation, WHO grade IV; Diffuse midline gliomas, H3 K27M-mutant, WHO grade IV (2). Different diffuse glioma types lead to different prognosis and treatment methods. Researchers showed that greater grading was associated with poorer prognosis and was an independent prognostic factor in IDH-mutant grade II and grade III gliomas (3). Gliomas grading and molecular characteristics all should be considered into patients’ management for improving prognosis (4). Glioblastoma (GBM), which is a World Health Organization grade IV tumor account for 53.1% of all intracranial neuroepithelial tumors with a median survival of 8 months and 5-year survival rate of 7.2% (1), is newly defined as a diffuse astrocytoma without IDH nor histone H3 genes mutation. It is characterized by microvascular hyperplasia, necrosis and/or specific molecular characteristics, including TERT promoter mutations, EGFR gene amplification, and/or a + 7/- 10 cytogenetic feature (2). Despite continuous progress in glioma diagnostic and therapeutic strategies, the mortality rate of glioma remains high. This high morbidity is associated with high tumor heterogeneity, early and widespread diffuse malignant cell infiltration, difficulty in achieving complete surgical removal, and high intrinsic chemotherapy and radiotherapy resistance (5). However, the molecular mechanisms underlying gliomas remain unclear. Therefore, studies with regard to the molecular mechanism of gliomas are urgently needed to identify potential prognostic molecular markers and effective therapeutic targets to improve the survival of patients with glioma.

Long noncoding RNAs (lncRNAs) are the largest class of noncoding RNAs; these sequences are composed of more than 200 nucleotides and have rich biological functions. LncRNAs play crucial regulatory roles in modulating the cell cycle, cell proliferation, cell invasion, and migration by regulating gene expression at the transcriptional, post-transcriptional, and epigenetic levels (6, 7). Studies have demonstrated that dysregulation of lncRNA expression is related to gliomas tumorigenesis and cancer progression and that it is a potentially effective early diagnostic, prognostic biomarker, and therapeutic target (8). However, the specific and detailed molecular mechanisms underlying the involvement of lncRNAs in glioma carcinogenesis and progression remain largely unknown.

MicroRNAs (miRNAs) are endogenous noncoding RNAs with a length of 20–22 nucleotides and regulate gene and transcription factor expression at the transcriptional and post-transcriptional levels. MiRNAs are closely related to the tumor microenvironment, which comprises various cell types and molecular patterns and, in turn, affects cancer cells and cancer-related miRNA expression. Many studies have shown that miRNAs also play important roles in tumorigenic processes, such as cell proliferation, metastasis, invasion, and angiogenesis (9).

LncRNAs act as microRNA sponges via miRNA-response elements and reverse the inhibition of miRNAs and target mRNAs to function as competing endogenous RNAs (ceRNAs) (10). An increasing number of reports have shown that the ceRNA module is involved in tumor initiation and progression. However, limited ceRNA research has been conducted on gliomas.

This study was conducted to explore the molecular mechanisms underlying the coding and noncoding RNAs of gliomas. We comprehensively analyzed the association of mRNA, lncRNA, and miRNA expression profiles and clinical outcomes in a large cohort of patients with glioma by combining The Cancer Genome Atlas (TCGA) RNA-seq data with Genotype-Tissue Expression (GTEx) normal brain RNA-seq data, as TCGA glioma data lack normal control samples. We first constructed a lncRNA-miRNA-mRNA ceRNA network with prognostic value for patients with glioma. Significant differential expression of representative lncRNAs, miRNAs, and mRNAs was further confirmed at the cellular level. To further explore the molecular mechanisms of a glioma, the functions of differentially expressed genes (DEGs) were evaluated. Gene co-expression and transcription factor (TF) prediction was conducted to clarify the regulatory relationships between DEGs.

Currently, the tumor immune environment and immune-related treatments have received increased attention. Immune cell and pathway deregulation has been studied in glioma using single-sample Gene Set Enrichment Analysis (ssGSEA). We analyzed the expression of immune checkpoint (IC) molecules (IDO, TIM-3, LAG-3, CTLA-4 (CD152), PD-1, PD-L1, PD-L2, B7-H3, B7-H4, B7-H5, and TIGIT) to provide a theoretical basis for tumor immunotherapy. Furthermore, we filtered small molecular drug targets on DEGs to identify potential treatments using the Drug Gene Interaction Database (DGIDB). These emerging strategies indicate the potential of tumor immunotherapy to overcome the limitations of existing tumor immunotherapy. It is hoped that our data will reveal the mechanisms underlying the regulation of glioma development and prognosis and provide novel prognostic biomarkers and therapeutic targets for glioma.

Glioma is a common malignant neuroepithelial neoplasm that contributes to the largest number of deaths caused by central nervous system tumors (13). WHO 2021 diffuse glioma classifications are relatively complex which based on histology and molecular markers, but it makes treatment more personalized (2). However, these precise treatments are after the accurate diagnosis. In order to improve the prognosis of glioma patients, early detection and treatment are urged needed. At present, the lack of early specific diagnostic and prognostic biomarkers and therapy resistance may contribute to the worse survival rate of patients with glioma compared with that of patients with other central nervous system tumors (5). Therefore, to improve clinical outcomes, it is essential to explore the exact intrinsic regulatory mechanisms of glioma initiation and progression and identify potential glioma-related prognostic signatures and new potential therapeutic targets. Research showed that circulating free DNA, circulating tumor cell, circulating tumor DNA and microRNA in blood or cerebrospinal fluid were used for GBM early diagnosis, early tumor progression detection, monitoring of tumor treatment response, personalized treatment identification, such as EGFR, TGF-β, IDH, miR-21and so on. Those gene play vital role in GBM early diagnosis and specific therapy and could be detected in liquid biopsy for overcoming the solid cancer biopsy limits (14).Recently, an increasing number of studies have demonstrated that miRNA-mediated ceRNA crosstalk plays a vital role in cancers, including glioma (9, 15).

As TCGA glioma data lack normal brain control data, comprehensive analyses of the DEmRNA and noncoding RNA profiles in glioma, based on large-scale RNA sequencing platforms, are lacking. Using RNA-seq data from TCGA and GTEx databases, we identified 7017 mRNAs, 437 lncRNAs, and 9 miRNAs that were aberrantly expressed in patients with glioma compared with normal brain tissues. Among these genes, 5684 mRNAs, 61 lncRNAs, and 7 miRNAs showed prognostic significance.

After constructing a ceRNA network and key sub-network, key genes were selected, and the results showed that the PART1-hsa-mir-25-SLC12A5/TACC2/BSN/TLN2/ZDHHC8 regulation axis plays an important role in the ceRNA network. LncRNA PART1 was downregulated in gliomas, and its overexpression in U87 cells induced cell growth arrest and apoptosis, indicating that it plays an anti-oncogenic role in glioma (16). Additionally, hsa-mir-25 was upregulated in gliomas and showed poor OS, which was consistent with the results of another study showing that miR-25 overexpression in U251 cells promotes cell proliferation and invasion (17). SLC12A5/TACC2/BSN/TLN2/ZDHHC8 was downregulated in patients with glioma with poor OS. SLC12A5 is a type of Kcl cotransporter that maintains neuronal chloride homeostasis, and its aberrant expression is involved in tumorigenesis in various cancer types, such as bladder urothelial carcinoma (18) and colon cancer (19). TACC2, an oncogene in breast cancer, is a member of the transforming acidic coiled-coil protein family (20). BSN is mainly expressed in the brain and is involved in organizing the presynaptic cytoskeleton. BSN mutations are associated with endometrial cancer prognosis (21). TLN2 is an intracellular cytoskeletal protein, a major component of the focal adhesion complex, involved in breast cancer and hepatocellular carcinoma (22). ZDHHC8 is a member of the zinc finger DHHC domain-containing protein family that functions as a palmitoyltransferase. The role of ZDHHC8 in human cancers remains unknown. The interaction between PART1-hsa-mir-25-SLC12A5/TACC2/BSN/TLN2/ZDHHC8 is largely unclear in glioma and requires further analysis.

Our gene co-expression network analysis showed the interactions between gene molecules. We filtered the core genes and important hubs in the network module. The HLA and MT families are pivotal in the network. The HLA family plays a key role in the mechanism of anti-tumor immune effects. Cellular immunity plays a leading role in tumor cell killing, which is facilitated by humoral immunity. The HLA family includes HLA class I (HLA-A, HLA-B, HLA-C, and HLA-E) and HLA class II (HLA-DRA, HLA-DRB1, and HLA-DG). HLA-E is overexpressed in GBM (23), and HLA-DR is upregulated in gliomas compared with normal brains and is associated with poor prognosis (24). However, further in-depth studies are required to reveal the underlying mechanisms. MT family expression alterations, such as those caused by mutations in MT-ND4 or MT-ND6, involving molecular interactions and pathways that influence apoptosis are related to GBM (25). Based on GBM scRNA-seq data, Garofano et al. classified cells into four distinct clusters, including MTC (mitochondrial cluster). They specifically clarified the role of mitochondrial in GBM. General mitochondrial functions, mitochondrial metabolism, oxidative phosphorylation and fatty acid oxidation were the hallmarks of the MTC cluster. Mitochondrial complex I were upregulated in MTC cluster which related to the Warburg effect. GBM patients benefit a significantly longer survival time in the MTC group which was the only independent prognostic factor. As for tumorous immune microenvironment, microglia-like cells were enriched in MTC. MiR-30 family activated in MTC GBM patients which inhibit glycolysis and promote mitochondrial respiration. Meanwhile, MTC were sensitivity to radiotherapy and higher intracellular ROS via inhibiting oxidative phosphorylation (26). All these results demonstrated that mitochondrion plays a vital role in GBM, whereas, the specific role of mitochondrion also should be further examined.

Chea3 was used to predict the TFs of DEGs, and KCNIP3, PIN1, and NACC2 were found to play crucial roles in regulating DEG expression. The TF overlapping genes are presented in Supplementary Table 6 . Zhang et al. reported that Pin1 promotes SUMO1 protein sumoylation to improve glioma malignancy (27) and clarified the role of Pin1 in glioma. Furthermore, KCNIP3 (28) and NACC2 (29) show a close relationship with cancer but whether they play related roles in glioma has not been previously reported.

GO and KEGG enrichment analyses of the biological functions enriched among the DEGs revealed GO terms involved in biological processes, including RNA catabolic processes, translational initiation, mitochondrial inner membranes, and neuron-to-neuron synapses. Some of the results of gene co-expression analysis were confirmed by the GO terms, such as mitochondrial gene and mitochondrial-related function. The results of KEGG analysis were all related to central nervous system diseases. GO and KEGG enrichment results also demonstrated the accuracy of our data analysis results.

Recently, tumor immunity has received increased attention. Tumor immunology is used to study the relationship among tumor antigens; immune function; and tumor occurrence, development, and prognosis, as well as the mechanism of the tumor immune response and tumor cell immune escape and immune-based tumor diagnosis and prevention. Our data showed that the HLA family plays a pivotal role in glioma, suggesting that tumor immunology is related to the occurrence and development of gliomas. ssGSEA to evaluate glioma microenvironment immune infiltration showed that many immune cells and immune functions exhibited aberrant infiltration and activation, respectively, in glioma. B cells and CD8+T cells showed higher and lower glioma infiltration, respectively. Han et al. observed abnormal proliferation of B cells in glioma, whereas CD8+T cell activity was suppressed (30). CD8+T cells with Pdia3, Mgat5, Emp1, or Lag3 editing enhanced the survival of GBM-bearing mice (31). Lee et al. reported that local delivery of B cell-depleting anti-CD20 immunotherapy improves the OS of GBM mouse models, and that PD-L1 promotes tumoral B cell function to suppress CD8+T cell activation (32). This evidence suggests a potential role for B and CD8+T cells in glioma progression and indicates that macrophages show higher infiltration in glioma. Macrophages are recruited to the glioma environment to create a supportive stroma for glioma cell proliferation, survival, and migration through immune functions, such as the release of transforming growth factor-β, epidermal growth factor, interleukin (IL)-6, and IL-1β (33). Our results showed that neutrophils infiltrated glioma cells. IL-8, chemokine ligand 8, and macrophage migration inhibitory factor in the GBM microenvironment led to neutrophil infiltration. Neutrophils promote glioma malignancy and regulate S1004A expression to mediate glioma angiogenesis. In contrast, neutrophil depletion prolonged GBM mouse survival, suggesting that neutrophils play a significant role in glioma progression. However, the mechanism underlying the role of neutrophils in gliomas requires further study (34). Tregs are central agents responsible for modulating immunity by inhibiting CD4+T helper, CD8+T, and B cells in cancer (35). We confirmed that Treg scores were increased in glioma tissue, whereas El Andaloussi et al. reported that circulating Treg scores are increased in patients with GMB (36). The role of Tregs in glioma is currently under exploration. Stimulated cytolytic activity was observed in glioma, as demonstrated by the results of Wang et al. (37). Study showed the safe and feasible of CAR-T therapy in GBM, but CAR-T therapy also showed uncertain curative effect among GBM patients (38). Considering the limitations of CAR-T therapy, Gatto et al. firstly proposed the idea that the possibility of novel strategy CAR-M therapy in GBM (39). At present, there have been no research on the use of CAR-M in GBM patients, CAR-M therapy treat GBM need further research. Immune cytolytic activity is highly associated with glioma malignancy. The cytolytic activity-related PDGFA and EGFR genes have been found to be overexpressed in glioma, and a positive association between cytolytic activity and HLA expression has been observed; however, the underlying mechanisms are still being evaluated (37, 40). HLA also showed higher expression in glioma. The type II IFN response is a significant immune pathway found to be inhibited in the present study. IFN‐γ, the main molecule involved in the type II IFN response, plays a dual role in the immune system. IFN‐γ can inhibit tumor angiogenesis (41) and induce PD‐L1 expression to promote immune escape (42). The exact role of the type II IFN response is currently unclear. Antigen-presenting cell co-stimulation and co-inhibition, CCR, inflammation-promotion, parainflammation, T cell co-inhibition, and co-stimulation are significant immune pathways; however, their specific roles in glioma have not been investigated. ssGSEA showed that ICs expressed on immune cells were upregulated in gliomas. ICs negatively regulate T-cell activation involved in anti-tumor immune responses. Our data confirmed that the expression of ICs (B7-H3, B7-H4, B7-H5, IDO, LAG-3, PD-1, PD-L1, PD-L2, and TIM-3) was significantly higher expressed in gliomas than in normal brain tissues. Many PD-1-, PD-L1-, and IDO-targeting drugs have shown some efficacy in clinical trials, and other related ICs show potential as therapeutic targets. Researchers summarized that PD-L1 and PD-1 overexpression were associated to poorer prognosis, however, other article showed no statistical significance between PD-L1 expression and GBM prognosis. Meanwhile, the results of clinical trials demonstrated that only 7.8% of nivolumab (PD-1 inhibitor)-treated recurrent GBM patients experienced objective response compared with 23.1% in bevacizumab group, and there is no statistical significance of de novo MGMT-unmethylated GBM between nivolumab and standard radiotherapy compared with temozolomide and radiation therapy group. But, nivolumab could bring survival benefit for MGMT-methylated GBM patients (43). Therefore, the role of PD-L1, PD-1 and their relationship with MGMT, IDH, VISTA, B7-H3 in glioma need specific deep more research.

Using a series of bioinformatic analyses, we identified DEGs and constructed a ceRNA network of prognosis-related miRNAs. We screened out the core regulatory axis to understand the occurrence and development of glioma from the perspective of the ceRNA hypothesis. The transcriptional level of regulatory axis genes was verified at the cell level, establishing a foundation for further research.

From the perspective of coding genes, co-expression network analysis of glioma DEG profiles revealed interactions between genes, reflected glioma gene expression regulation relationships, and screened out the core genes. GO and KEGG enrichment analyses were performed to determine the biological functions of different genes. To understand tumorigenesis in terms of TF-regulated coding genes, TF prediction was conducted.

Co-expression network analysis suggests that tumor immune-related genes are associated with gliomas. Further ssGSEA analysis and the expression of IC genes suggested that tumor immune disorders are closely related to glioma, providing new targets for tumor immunotherapy. Further investigation of the mechanisms of glioma pathogenesis and progression is required to yield novel glioma treatment strategies.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by Ethics Committee of Jiangyin people’s Hospital. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

ZY and WG designed and performed the study and wrote and revised the manuscript. TZ performed cell biological experiments and analyzed related data. HG conducted the total study process. All authors contributed to the article and approved the submitted version.

This work was supported by a grant from the Wuxi Commission of Health (Q201938) and Wuxi overseas students science and technology activity project (2019-06JY).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.