| Glioma was deemed to be the most aggressive tumor in central nervous system(CNS).The annual incidence of glioma was reported to be about 30-80/1 million in the world,and increased by 1%-2%annually,the 5-year survival rate was only 10%-20%.According to the WHO standard,glioma was divided into 4 grades by its'pathological characteristics:Grade ?,pilocytic astrocytoma,which manifested as a benign tumor,patients may have full clinical recovery after total tumor resection.Grade ?,which had a poor prognosis compared with Grade ?.,but were still considered to be low-grade glioma(LGG).Grade ?,such as anaplastic astrocytoma,as well as Grade ?,GBM,were types of high-grade glioma lined to high degree of malignancy,strong invasive abilities,poor prognosis and multiple differentiation potentials,the median survival time was only about one year.Compared with pathological classification,molecular pathological characteristics were much more accurate.Commonly used molecular pathological classification were as follows:Four-grade classification of glioma by The Cancer Genome Atlas(TCGA),USA.Three subtypes(G1,G2 and G3)were divided by mutation rate of IDH and co-deletion of 1p19q gene by Chinese Glioma Genome Atlas(CGGA).Besides that,several molecular pathological markers already used in practice,such as Cytosine phosphoric acid guanine(CpG)island methylation,methylation of MGMT in promoter region,EGFRv? amplification,PTEN mutation,TP53 mutation,BRAF fusion and point mutation and so on.With the development of high-throughput sequencing technology,Human Genome Project was launched in the late of 20th century,combined with genetics,molecular biology,mathematics,statistics and computer programming,quantity of achievements was obtained,including the development of bioinformatics.For serving scholars worldwide,several bioinformatics databases had been established to collect relevant bioinformatics data,including The Cancer Genome Atlas(TCGA),Gene Expression Omnibus(GEO),Kyoto Encyclopedia of Genes&Genomes(KEGG)and etc.TCGA was funded by the National Institutes of Health(NIH).Through large-scale genome sequencing and multidimensional analysis,more than 30 human tumor group were classified,and public access to cancer genome datasets for all the scholars in the world were provided by this program.KEGG was established in 1995 by the Kanehisa Laboratory in Kyoto University,Japan.This program mainly focused on genomic,chemical,systemic functional information and genome deciphering.Based on TCGA and KEGG,bioinformatics analysis,such as Gene Set Enrichment Analysis(GSEA)and Weighted Gene Co-expression Network Analysis(WGCNA),can be performed on deciphering the interrelationship between gene expression and tumor,even more,detecting biomarkers for tumors.Those biomarkers were used not only for early diagnosis,target treatment and prognosis prediction,but also for understanding the mechanism of potential anticancer/oncogenic genes affecting the biological function of glioma.Objective:1.Detecting biomarkers for glioma through large-scale genome sequencing.2.To investigate the differential expression of GALNT7 in glioma and its effect on the biological function of glioma cells.3.To investigate the correlation between DNA methylation and MRC2 expression in glioma,meanwhile,the affecting on biological function of glioma cells.Methods:1.mRNA and Clinical data from TCGA database were processed by R X64 3.4.1.According gene differential expression analyzation and survival analysis,polypeptide N-acetyl galactosaminyl transferase 7(GALNT7)and mannose receptor C-type 2(MRC2)were considered to be the prognosis factors of glioma.2.One-way ANOVA and multiple comparisons were used to verify whether there was a different expression of GALNT7&MRC2 between glioma and normal brain tissue.The expression of GALNT7&MRC2 in glioma was divided into high and low groups by the median expression of GALNT7&MRC2 in glioma from TCGA database.The correlation between the expression of GALNT7&MRC2 and clinical features(including gender,age,tumor grade,recurrence time,overall survival time,etc.)was analyzed by T-test(P<0.05).Meanwhile,Kaplan-Meier survival was used to value the effect of the expression on the Disease-Free Survival(DFS)and Overall Survival(OS)(P<0.05).3.Download the reference set(2.cp.kegg.v5.1.symbols.gmt)from the Molecular Signatures Database(MsigDB).Random combinations were set at 1,000 times,for nominal P-values(NOM P-val)<0.05,and false discovery rates(FDR)<0.25.Gene-Set Enrichment Analysis(GSEA)was performed by GSEA3.0.As a result,signaling pathways which may be regulated by high expression of GALNT7&MRC2 were identified.4.Download Illumina Human Methylation 450 data from TCGA database,combine with mRNA expression data,relationship between gene methylation and expression were analyzed(P<0.05 and log2(foldchange)>1).5.The expression differences of GALNT7&MRC2 in glioma and normal brain tissues were verified by immunohistochemistry and Western Blot.Results:1.Correlation analyzation between GALNT7&MRC2 expression and clinical characteristics showed that GALNT7&MRC2 were highly expressed in glioma,and the expression level was proportional to malignancy and tumor grade(P<0.05),and negatively associated with age and survival rate(P<0.05).By Kaplan-Meier survival analysis,patients with high expression of GALNT7&MRC2 have relatively shorter DFS and OS than those exhibiting low expression.2.GSEA(NOM P-val<0.05;FDR<0.25)results of GALNT7 indicated 5 pathways which were activated by high expression of GALNT7,including regulation of the actin cytoskeleton signaling pathway,NK cell mediated cytotoxicity signaling pathway,JAK-STAT signaling pathway,CAMs signaling pathway and ECM-receptor interaction signaling pathway.Pearson's correlation analysis was used to predict genes that are co-expressed with GALNT7 in those pathways.As a result,Crk,RAC1,STAT3,PVR and TNC were identified as potential target genes.Crk participated in regulating the actin cytoskeleton pathway,RAC1 was involved in the NK cell mediated cytotoxicity pathway and STAT3 was associated with the JAK-STAT signaling pathway,PVR activated CAMs pathway and TNC involved in the ECM-receptor interaction pathway.Differential analyses demonstrated that Crk expression was positively associated with the malignancy of glioma;positive association was also observed with RAC1,STAT3,and TNC.PVR highly expressed in GBM,but expressed lower than normal brain tissue in LGG(P<0.05).Survival analyses demonstrated that DFS was negatively correlated with the expression of RAC1,STAT3,PVR and TNC,but not Crk.Meanwhile,the expression levels of Crk,RAC1,STAT3,PVR and TNC were negatively correlated with OS(P<0.05).3.GSEA results indicated 7 signaling pathways were activated due to high expression of MRC2,including NOTCH signaling pathway,TLRs signaling pathway,regulation of actin cytoskeleton signaling pathway,NK cell mediated cytotoxicity signaling pathway,JAK-STAT signaling,CAMs signaling pathw-ay and ECM-receptor interaction signaling pathway.In order to predict genes that are co-expressed with MRC2,Pearson's correlation analysis was used to determine the correlation between MRC2 expression and all genes were associated with these signaling pathways.As a result,NOTCH3,MYD88,MSN,HLA-A,STAT3,CD276 and TNC were identified as potential target genes.NOTCH3 regulated NOTCH signaling pathway,MYD88 activated TLRs signaling pathway,MSN participated in regulating the actin cytoskeleton pathway,HLA-A was involved in the NK cell mediated cytotoxicity pathway,STAT3 was associated with the JAK-STAT signaling pathway.Also,CD276 participated in the CAMs pathway and TNC was involved in the ECM-receptor interaction pathway.The expression of NOTCH3,MSN,HLA-A,STAT3,CD276 and TNC were positively correlated with the malignancy(P<0.05).Survival analysis showed that the expression levels of NOTCH3,MSN,HLA-A,STAT3,CD276 and TNC were negatively correlated with DFS and OS(P<0.05).4.MRC2 methylation was considered to be lower in glioma,compared with normal brain tissues,and negatively correlated with MRC2 expression.1 CpG site(cgl4764661)in promoter region and 8 CpG sites(cg00679546,cg05651778,cg11628739,cg24368031,cg13659446,cg14237903,cgl6200799,cg17226280)in intron 1 region were negatively correlated with MRC2 expression.Survival analysis also showed that the high expression of MRC2 in inverse proportion to hypomethylation and led a relatively shorter DFS and OS.5.Immunohistochemical results proved that GALNT7 and MRC2 were differentially expressed in GBM,LGG and normal brain tissues(P<0.05),with the highest expression in GBM,followed by LGG,and the lowest in normal brain tissues.This indicated that the expression of GALNT7&MRC2 was positively correlated with malignancy.In addition,Western Blot was used to detect the expression of MRC2 in normal brain tissue,WHO ? glioma and glioma cell lines(U373,A172).The results showed that the expression of MRC2 in glioma,U373 and A172 was significantly up-regulated compared with that in normal brain tissue.Conclusion:Biomarkers which were affected biological functions of tumor cell could be identified on a large scale by bioinformatics analysis.GALNT7 and MRC2 were conformed aberrantly high expressed in glioma.GALNT7 and MRC2 affect the prognosis of glioma by enhancing the proliferation and invasion of glioma cells through multiple signaling pathways,which could be used as biomarkers for early diagnosis and target treatment in glioma. |
PDF Full Text Request