A Novel Defined Endoplasmic Reticulum Stress-related LncRNA Signature For Prognosis Prediction And Immune Therapy In Glioma

Objective Glioma,accounting for 44.69% of intracranial tumors,is the most common intracranial tumor in humans.According to the World Health Organization(WHO)classification,the higher the grade of glioma,the worse the prognosis.The median survival of glioblastoma multiforme(GBM,grade Ⅳ glioma)is only 12-15 months.Despite great advances in the treatment of gliomas,they remain largely incurable.Long non-coding RNA(lncRNA)can regulate the life activities of organisms by participating in the mechanism of stabilizing messenger RNA,regulating protein translation,RNA processing and modification,regulating the development of organisms,tissue differentiation,organ formation and other basic life activities,and participating in the occurrence and development of a variety of diseases including tumors.Lnc RNAs regulate DNA/RNA methylation,histone modification,and chromatin remodeling through a variety of pathways,and play multiple roles in glioma progression.Endoplasmic reticulum(ER)stress is a stress response that recovers homeostasis in the ER and is manifested as an unfolded protein response(UPR).Tumor cells,including glioma cells,can usually maintain the activation of UPR for a long time and support its survival.It has been reported that lncRNA and ER stress can regulate each other and jointly determine the fate of tumor cells.Therefore,we speculate that ER stress-related lncRNAs also play an important role in the occurrence and development of gliomas.We attempted to construct an ER stress-related lncRNA gene prognostic signature to predict the prognosis of patients and provide new biomarkers and therapeutic targets for glioma.Methods RNA-seq data of 689 glioma patients and clinical data of 648 glioma patients were downloaded from the TCGA database.The clinical information of these clinical data included survival status,WHO grade,clinicopathological features and age of disease.In addition,we also downloaded RNA-seq data of 618 glioma patients and clinical data of 465 glioma patients from the CGGA database.The clinical information of these clinical data included age of disease,gender,WHO grading,radiochemotherapy information,histopathological grading and survival information.We also downloaded 1132 normal brain tissue expression data from the GTEx database.The samples in TCGA database are taken as the training set,and the samples in CGGA database are taken as the test set.According to ER stress-related genes collected from the literature,161 ER stress-related lncRNAs were obtained by co-expression analysis.Subsequently,21 prognostic related lncRNAs were screened by differential analysis and univariate Cox analysis.A prognostic risk signature containing 6 ER stress-related lncRNAs was constructed using the "glmnet" R package,and the ability of the risk signature to predict prognosis was evaluated using ROC curve.Three cluster subgroups with different molecular subtypes were identified by consensus cluster analysis.Matrix score,immune score,ESTIMATE score and tumor purity of TCGA samples were calculated using ESTIMATE algorithm.The ss GSEA algorithm was used to assess the abundance of 24 types of immune cells in the training set and three cluster subgroups.The expression of 38 immune checkpoints in the training set and three cluster subgroups was evaluated using the "ggpubr" R package.TIDE algorithm was used to evaluate the clinical response of two risk groups in the TCGA database to two immune checkpoints,PD-1 and CTLA-4.Functional analysis was performed using GSEA,GO,and KEGG.Finally,one lncRNA,LINC00519,which has not been reported in glioma and has survival significance for glioma,was selected from the 6 lncRNAs in the risk signature for further experimental study.The gene silencing cell model was constructed by si-LINC00519 to verify the effect of LINC00519 silencing on the phenotype of glioma cells.Result(1)Differential expression profile and screening of ER stress-related lncRNAs.(2)Identification of ER stress-related lncRNA and construction of risk signature.(3)Verification of prognostic signature in training set.(4)The verification of prognostic signature in the test cohort.(5)Clinicopathological features and immune microenvironment of three clusters of patients with glioma.(6)Tumor immune microenvironment of glioma in two risk groups.(7)Prediction of immunotherapy and anticancer drug response.(8)Functional analysis of the risk signature.(9)Silencing of LINC00519 inhibited the proliferation,migration and invasion of glioma cells.Conclusion In this study,6 ER stress-related lncRNAs were obtained that were significantly associated with prognosis,and a risk signature that could predict the prognosis of glioma patients was successfully constructed.In vitro cell experiments confirmed that silencing the expression of LINC00519 in glioma cells inhibited the proliferation,migration and invasion ability of glioma cells.Therefore,ER stressrelated lncRNAs may be a new target for the treatment of glioma.