| Objective Glioma is a life-threatening brain tumor with an extremely poor prognosis.It accounts for about 70 percent of all primary brain tumors.Glioblastoma（GBM）is the most deadly brain tumor.The 5-year survival rate of GBM is less than 5%.According to the diagnosis,the therapy of gliomas includes surgery,chemotherapy,and radiation,which is rarely effective because the prognosis is highly dependent on clinical and molecular factors.Many sno RNAs are dysregulated in cancer,expressed differently in cancer type,stage,and metastasis,and can alter the progression of disease.Sno RNA plays an important role in glioma and can impact the progression of glioma.So,we speculated that the methylation of sno RNA gene is also crucial for the treatment and prognosis of glioma.We tried to construct a sno RNA gene prognostic signature,whether it can predict the prognosis of glioma and provide new biomarkers and therapeutic targets for patients.Methods Clinical information and 450 K methylation data of glioma patients were downloaded from TCGA,which included 649 glioma samples after eliminating incomplete cases.Clinical information on these data included survival status,tumor grade,clinicopathological features,and age.Then,the glioma samples obtained were randomly selected and segmented into training（n = 326）and validation cohort（n = 323）.We downloaded the gene annotation information corresponding to the GPL13534 probe and extracted the methylated promoter region（200 bp upstream of the promoter,1500 bp upstream of the promoter,5’-untranscribed region,exon 1）.Then,we downloaded the GTF file of the UCSC genome.Finally,we annotated probes and gene names through the GTF file,extracted a total of 485,778 Cp G sites,and obtained 1,244 sno RNA Cp G sites,corresponding to 208 sno RNA gene Cp G promoter regions（multiple Cp G sites correspond to one sno RNA,take the average）.Through Cox analysis,five genes of methylated sno RNA associated with survival were identified.Through the Glmnet R package,a signature that can predict prognosis was built.The survival R package was demonstrated to have a predictable prognosis in gliomas.We also used ROC curve and Cox analysis to prove the reliability and independence of the risk signature.GO and KEGG pathway enrichment analyses were conducted to estimate the correlation with genes based on risk scores.Using the Epi DISH model,the relative proportions of six different immune cells were calculated to infer the number of immune infiltrating cells.Finally,we selected one sno RNA,SNORA71 B,which has not been studied in glioma and has survival significance in glioma patients,from SNORA14 B,SNORD113-4,SNORA71 B,SNORA80B and SNORD97 for study.si RNA was used to construct the silencing model of SNORA71 B.The effect of silencing with si-SNORA71 B on glioma cells was verified by functional experiments of cells.Result（1）The preparation of datasets,the grading of glioma patients,and the analysis of sno RNA gene methylation levels.（2）Establishment of prognostic signatures and verification of their feasibility（3）A risk signature containing 5 methylated sno RNA genes can independently predict the prognosis of glioma patients.（4）Validation of the 5methylated sno RNA gene signature in the validation cohort.（5）Clustering of methylated sno RNA genes showed that gliomas were divided into two subtypes.（6）Infiltration of immune cells in different risk groups.（7）Functional annotation of glioma by the risk signature.（8）Si-SNORA71 B was used to construct a silencing model of SNORA71 B and perform functional verification.Conclusion In this study,five methylated sno RNA genes associated with prognosis were obtained,and the methylated sno RNA gene signature was constructed in glioma for the first time.It successfully predicts the prognosis of glioma patients.The silencing model of SNORA71 B was constructed by using Si-SNORA71 B,and the functional verification was performed with the cellular model.It is proved that SNORA71 B may become a new target for glioma therapy. |
PDF Full Text Request