| Background: Glioma is one of the most common malignant tumors in the world,and its incidence is still increasing year by year,and the prognosis is poor,bringing a considerable burden to society.The World Health Organization(WHO)central nervous system tumor classification system increasingly emphasizes the importance of tumor molecular subtypes and prognostic biomarkers for diagnosis and treatment.So far,many glioma-related molecular markers have been developed.However,these markers must be performed better in clinical practice due to tumor heterogeneity.In addition,conventional treatment methods for glioma,including surgical resection,chemotherapy,and radiotherapy,have not achieved optimal effects.In recent years,immunotherapy,represented by immune checkpoint blockade(ICB)therapy,has achieved a good therapeutic effect in various tumors.ICB is also expected to be a promising treatment strategy for glioma.However,the current relevant clinical trials have yet to achieve satisfactory results,and the clinical benefit of ICB therapy for glioma patients is highly heterogeneous.Therefore,the key to improving the efficacy of immunotherapy is to explore the tumor immune microenvironment and identify patients who may be suitable for ICB therapy.Super-enhancer(SE)is an active enhancer cluster formed by the tandem of successively arranged enhancers,which has a stronger ability to regulate gene transcription and plays an essential regulatory function in biological processes such as cell differentiation and immune response.Long non-coding RNA(Lnc RNA)is a research hotspot in the field of epigenetics and is widely involved in various biological behaviors such as proliferation,epithelial-mesenchymal transition(EMT),tumor immune microenvironment regulation,invasion,and chemotherapy resistance.SE generally drives high expression of lnc RNAs with tumor-promoting functions.However,the role of super-enhancer-associated lnc RNAs(SE-lnc RNAs)in gliomas remains unclear.Methods: RNA sequencing(RNA-seq)and chromatin immunoprecipitation sequencing(Ch IP-seq)data from glioma patient-derived glioma stem cells(GSCs)and neural stem cells(NSCs)were downloaded from the Gene Expression Omnibus(GEO)database(GSE119776).SE was identified using the Rank Ordering of Super-Enhancers(ROSE)algorithm based on Ch IP-seq data,and SE-regulated genes were screened.The unique super-enhancer-associated genes(SE-genes)in GSCs were selected to intersect with the highly expressed genes in GSCs obtained by difference analysis of RNA-seq data and further screened for SE-lnc RNAs,then intersected with SE-genes present in more than10% of glioma patients and the top 100 SE-genes in each patient according to Ch IP-seq data of GSCs.Finally,the final SE-lnc RNAs were obtained by combining the RNA-seq data of the Chinese Glioma Genome Atlas(CGGA)database and The Cancer Genome Atlas(TCGA)database.Consensus clustering analysis was used to carry out molecular typing.Combined with the survival of glioma patients in the CGGA database,a prognostic model was constructed by Least Absolute Shrinkage and Selection Operator(LASSO)Cox regression analysis,and the risk score of each patient was calculated.The model was further verified in the TCGA database.Kaplan-Meier survival analysis was used to determine inter-group survival differences,and heatmaps show differences in inter-group clinicopathological features and gene expression levels.The receiver operating characteristic(ROC)curve was performed to evaluate the specificity and sensitivity of the prognostic model.Univariate and multivariate Cox regression analyses determined whether the risk score could be an independent prognostic factor for glioma patients.Gene Ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)performed functional analysis of genes related to the risk score,and gene set enrichment analysis(GSEA)showed functional enrichment of high-and low-risk groups.Tumor immunogenicity and tumor purity were evaluated by immunophenoscore(IPS)and R package "ESTIMATE",respectively.Then CIBERSORT,single-sample gene set enrichment analysis(ss GSEA),QUANTISEQ,MCPcounter,and EPIC algorithms were used to evaluate the difference in immune cell infiltration between patients in the highand low-risk groups.Then,IPS and subclass mapping(Sub Map)were used to predict the therapeutic response of glioma patients to immune checkpoint inhibitors(ICIs),and Genomics of Drug Sensitivity in Cancer(GDSC)was used to predict patients’ sensitivity to chemotherapy.The expression of LINC00945 at the glioma cell and tissue levels was detected by PCR experiments,and the epigenetic activation mechanism of LINC00945 was verified by Ch IP-seq and Ch IP experiments,and then the potential regulatory molecules upstream of LINC00945 were explored by PCR and western blotting experiments.The effect of LINC00945 on glioma proliferation in vitro was studied by CCK-8,cell cloning formation,and Ed U assays,and its effect on tumor growth in vivo was explored by a tumor formation experiment in nude mice.In addition,the effects of LINC00945 on EMT,migration,and invasion of glioma cells were investigated by PCR,western blotting,and Transwell assays.Results: 6 SE-lnc RNAs were obtained by combined RNA-seq and Ch IP-seq data analysis.Based on the expression of these 6 SE-lnc RNAs,3 glioma subgroups(cluster 1,cluster 2,cluster 3)with different prognostic and clinical characteristics were identified in the TCGA and CGGA databases,respectively.Compared with cluster 1 and cluster 2,the cluster 3 subgroup presented a shorter survival time and was significantly associated with glioblastoma(GBM),older age,death status,and wild-type isocitrate dehydrogenase(IDH).Furthermore,a prognostic model was constructed in the CGGA training set.Glioma patients were divided into high-risk and low-risk groups according to the median risk score.Compared with the low-risk group,patients in the high-risk group had a worse prognosis,showing more GBM subtypes,advanced age,dead status,recurrent glioma,and cluster 3 subgroup.Moreover,IDH mutated type,chromosome 1p and 19q(1p19q)codeletion,and O6-methylguanine DNA methyltransferase(MGMT)promoter methylation were associated with lower risk scores.Multivariate analysis demonstrated that risk score could be used as an independent prognostic factor for glioma patients.Similar results were obtained in the TCGA validation set.In addition,functional analysis showed that risk score-related genes were associated with mitotic biological processes,cell cycle,extracellular matrix(ECM)-receptor interactions,etc.,and the high-risk group enriched EMT and inflammatory pathways.Further analysis showed that patients in the high-risk group presented high immunogenicity and low tumor purity,mainly enriched M2 macrophages,regulatory T cells(Tregs),and cancer-associated fibroblasts(CAFs),etc.,and were more suitable for immunotherapy and chemotherapy.In addition,compared with other common tumors,the high expression of LINC00945 in glioma has certain specificity,and its contribution to the prognostic model is the highest,so the follow-up study of LINC00945 was conducted.We found that LINC00945 was highly expressed in glioma cells and tissues.Further Ch IP experiments confirmed that SE regulated the expression of LINC00945,and BRD4 may mediate the epigenetic activation of LINC00945.In addition,overexpression of LINC00945 promotes glioma cell proliferation,EMT,migration,and invasion in vitro and xenograft tumor formation in vivo.Conclusion: We performed a new molecular typing of glioma based on SE-lnc RNAs screened by multi-omics combined analysis and constructed the first prognostic SElnc RNA signature with the ability to optimize the treatment options of glioma patients receiving immunotherapy and chemotherapy,and SE-regulated LINC00945 promoting glioma progression may act as a potential glioma therapeutic target. |
PDF Full Text Request