A 9-methylated LncRNA Signature Predicts Survival In Patients With Glioma

Objective Glioma is one of the most common primary malignancies,accounting for more than 30% of all primary brain tumors.Over the past few decades,gliomas have been characterized by necrosis,aggressive growth,and angiogenesis.The World Health Organization(WHO)divides gliomas into four types according to their morphological characteristics and prognosis.Highly gliomas,which make up the majority of all gliomas,are heterogeneous and consist of tumor cells,glioma-like stem cells,and a wide range of blood vessels and immune cells.Currently,the main treatment options for gliomas include surgical resection,oral alkylating agents,and radiation therapy.Despite great advances in the treatment of glioma,the prognosis for patients with glioma remains poor.Therefore,there is an urgent need to identify the underlying molecular mechanisms of glioma development.Lnc RNA methylation plays an important role in the biological progression of human tumors,and abnormal lncRNA methylation is significantly associated with the prognosis of glioma patients.Therefore,we speculated that the detection of lncRNA methylation is very important for the prognosis of glioma patients.TCGA,as a foreign tumor database,provides a large number of clinical samples.We extracted clinicopathological characteristics and 450 K methylation data from glioma patients,constructed a risk prognostic signature of glioma patients with lncRNA methylation,and conducted further studies on lncRNA methylation as a prognostic biomarker for glioma and the development of a more accurate treatment process.Methods 450 K methylation data and clinicopathological characteristics of glioma patients were downloaded from TCGA.All clinical follow-up data and methylation data obtained were investigated.After our comparison,we found that 649 samples were successfully matched.The total samples were randomly divided into a training cohort(n = 325)and a validation cohort(n = 324).We used a univariate Cox model to select the major CpG loci associated with prognosis and set P <0.05 as a significance threshold.We only included CpG loci in patients with survival information.Then,CpG sites associated with patient outcomes were obtained and each CpG site was annotated with lncRNA.Finally,a total of 286 methylated regulated lncRNAs were identified.In the training cohort,regularized Cox regression was performed using the lncRNA methylation level and survival data of glioma patients,and nine survival related methylation regulated lncRNAs were obtained.Next,we used R language software to construct a risk factor model by using multivariate Cox regression equation between each gene expression and clinical samples.Clinical samples were divided into low-risk groups based on median risk.Kaplan-Meier survival analysis was used to determine the difference in survival between the two groups.Univariate and multivariate analyses were used to determine whether risk factor could be used as an independent prognostic factor.AUC curves were used to analyze the specificity and sensitivity of risk factor as a prognostic factor.GSEA compared the function of related genes and predicted the function and pathway of related genes.We identified two lncRNAs PCBP1-AS1 and LINC02875 that may be involved in malignant progression of glioma cells using TCGA database.Si RNA was used to construct PCBP1-AS1 and LINC02875 silencing models.Transwell migration and invasion assay was used to detect the effects of silencing PCBP1-AS1 and LINC02875 on glioma cell migration and invasion.Cloning assay and MTT assay were used to determine cell proliferation ability after silencing.Result(1)Analysis of lncRNA methylation levels and prognostic classification in glioma samples of different grades.(2)Univariate analysis identified nine lncRNAs that are methylated and regulated for survival.(3)The risk score of 9-methylated lncRNA was closely related to the clinicopathological characteristics of glioma patients.(4)Verification of 9-methylated lncRNA signatures in the verification queue.(5)Biological functions related to the high risk group in 9-methylated lncRNA model.(6)The expressions of 9 lncRNAs in normal brain tissues and glioma tissues were significantly different,and the prognosis of glioma patients was different.(7)Knocking out PCBP1-AS1 and LINC02875 could significantly inhibit the proliferation,migration and invasion of glioma cells.Conclusion We identified the nine long non-coding RNA associated with methylation as potential predictive markers or therapeutic targets in glioma patients.A risk model based on 9-methylated lncRNA expression was developed to better predict glioma OS.Functional analysis of related genes showed that these 9 lncRNAs were closely related to tumor development.The effects of PCBP1-AS1 and LINC02875 on the malignant biological functions of glioma cells were studied,and it was found that the proliferation,migration and invasion ability of glioma cells were inhibited after silencing the two lncRNAs respectively.We further confirmed that these lncRNAs play an important role in gliomas.