| [Background]Glioma is the most common intracranial malignant tumor, and astrocytoma is the most common one, followed by glioblastoma. Many researches have shown that the tumorigenesis of gliomas is often correlated with the inactivation of several tumor suppressor genes such as p53, p16, RB, PTEN, LRRC4 and activation of c-Myc, HRAS, EGFR, MDM2. We found that these genes have their specific expression profile in different pathological grades of gliomas. The mutation and overexpression of p53 and MDM2 are common in gliomas and showed no apparent correlation with the tumor grades, however there has also been reported that the mutation and overexpression of p53 is even obvious in high-grade gliomas. p16 and LRRC4 are down-expressed in gliomas and their expression levels are correlated with the glioma grades. The mutation of PTEN, loss of RB and overexpression of c-Myc, HRAS and EGFR are involved in high-grade gliomas.Gene expression is regulated at the genomic, transcriptomic, post-transcriptomic and proteomic levels. Recently the function of miRNAs in post-transcriptomic regulation are highly been concerned. Many researches have shown that miRNAs are playing important roles in cell proliferation, differentiation, apoptosis, development and metabolism. And it is noticed that the alteration of miRNA expression is common in most type of tumors including glioma. Studies showed that mir-21, mir-221 and mir-lOb are overexpressed, whereas a set of brain-enriched miRNAs such as mir-128 and mir-181 are down-expressed in glioblastomas. Further researches have found that these miRNAs could act as oncogene or tumor suppressors during tumorigenesis. However the miRNA expression profile in different pathological grades of gliomas was little known.According to the knowledge above, we analyzed simultaneously the expression profiles of miRNA and mRNA in gliomas with different pathological grades by using miRNA and cDNA microarray. Then the dynamics expression profiles of miRNAs and mRNAs were obtained, and docking study was done in gliomas with different pathological grades.[Construction of miRNA dynamics expression profiles in gliomas with different pathological grades]The expression levels of 637 miRNAs were detected in 3 normal brain tissue and 13 gliomas (3 grade I,5 grade II and 5 grade III) by miRNA microarray. Multi-methods were used to analyze the differentially expressed miRNAs in different pathological grades. Then the expression profiles of significant miRNAs were analyzed, and further classified into 8 categories. miR-26a is overexpressed in all gliomas especially in grade III whereas miR-21 and miR-23a are overexpressed mainly in grade III, which indicate that miR-26a may contribute to the initiation and progression of gliomas while miR-21 and miR-23a are mainly involved in the progression of gliomas. Additionally, miR-107, miR-124, miR-128a/128b, miR-138, miR-149, miR-181a, miR-181b, miR-302b are significantly down-regulated in all gliomas, and we infer that these miRNAs are correlated with the tumorigenesis of gliomas. These results implied that differentially expressed miRNAs have their specific expression profiles in different pathological grades of glioma.[Validation of miRNA dynamics expression profiles for six miRNAs in different pathological grades]The expression profiles of miR-107, miR-124, miR-138, miR-149, miR-302b and miR-23a in 6 normal brain tissue and 50 gliomas (6 grade I,20 grade II,9 grade III and 15 grade IV) were analyzed by Real-time PCR. It has been validated that miR-107, miR-124, miR-138, miR-149 are down-regulated in all gliomas and their expression levels are negatively correlated with pathological grades, which are consistent with the results of microarray. The expression level of miR-23a is increased gradually in gliomas and reaches the highest level in grade III, whereas no expression is detected in grade IV. Then we detected the expression level of miR-23a by in situ hybridization in same samples. The results showed that miR-23a is mainly expressed in nucleus of normal neurons and cytoplasma of glials in glioma gradeⅡandⅢ. We speculated that the expression of miR-23a in nucleus of neurons is possibly due to the inability of miR-23a probe to discriminate the pre-miRNA and mature miRNA. We don't exclude the possibility that there has an expression transition of miR-23a during the progression of gliomas. So, further experimental validation is very necessary.[The tumor suppressor function of miR-124 and miR-149 in glioma cells lines SF126 and U251]miR-107, miR-124, miR-138 and miR-149 were transfected into glioma cell lines to observe their cellular biological functions by multiple methods. MTT assay showed that miR-107, miR-124, miR-138 and miR-149 can inhibit the growth of SF126 and U251, and the inhibitory effects of miR-124 and miR-149 are the most significant. At the fourth day after transfection with miR-124 and miR-149, the proliferation rate of SF126 has decreased by 40% and 33% respectively, the similar result was observed in U251. Would healing assay indicated that miR-124 and miR-149 can inhibit the migration ability of SF126 and U251 significantly.48 hours after wounding, all cells have repaired their wounds except those transfected with miR-124 and miR-149 mimics. These results implied a possible tumor suppressor role of miR-124 and miR-149 during the tumorigenesis of gliomas.[Construction of mRNA dynamics expression profiles in gliomas with different pathological grades] By cDNA microarray, the expression levels of 30968 genes were detected in the same samples that ever used in miRNA microarray, and the same statistical approach used in miRNA data analysis was also applied. We found the expression profiles of significant genes in gliomas with different pathological grades could be classified into 14 categories. BMP1, CDK4, DDX5, DEK, HARS, HNRNPA2B1, IKBKB, LAMA4, p53, PCNA, PLAGL2, TLR3, VCAM1 are up-regulated in all gliomas. The expression levels of HARS, HNRNPA2B1, p53, PLAGL2 showed no apparent correlation with pathological grades while the expression levels of BMP1, DDX5, DEK, IKBKB, LAMA4, PCNA, TLR3 are positively correlated with pathological grades. ANXA1, JAK2, JUN, MMP14, RAP1B, STAT1, VCL, VIM are down-regulated mainly in glioma grade II and III. Additionally, the expression levels of ANK1 and APC are negatively correlated with glioma grades.[Bioinformatic analysis of differentially expressed mRNAs]A comprehensive functional annotation software DAVID was used to analyze the Gene Ontology_BP (biological process) of differentially expressed genes. Analysis showed that the up-regulated genes in glioma grade I are mainly involved in transcription, RNA splicing, cell proliferation and so on. While the up-regulated genes in glioma grade III are mainly involved in apoptosis, biological adhesion, RNA splicing, cell cycle, signal transduction, and so on. The down-regulated genes of glioma grade I, II and III are mainly involved in phosphorus metabolic process, intracellular signaling cascade, protein localization, intracellular transport, neuron differentiation, and so on. These results implied that the biological function of down-regulated genes has no significant difference among the glioma grades, whereas the biological function of up-regulated genes has made an obvious transition from low-grade glioma to high-grade glioma.Then the signaling pathways were analyzed and found several genes that involved in the regulation of MAPK, JAK/STAT, WNT, NF-κB and Toll-like receptor signaling pathways have altered expression in gliomas. The positive factors in regulation of the signaling pathways mentioned above such as RAP1B, JUN, FOS, JAK2, STAT1, TCF7L1, IKBKB, TRL3, MYD88 are up-regulated in gliomas while the negative factors such as DUSP8, APC, CYLD are down-regulated in gliomas. These results implied that MAPK, JAK/STAT, WNT, NF-κB and Toll-like receptor signaling pathways are activated during the tumorigenesis of gliomas.[Docking study of differentially expressed miRNAs and mRNAs]Pearson correlation analysis was used for estimating the correlation between differentially expressed miRNAs and their predicted target mRNAs, and the mRNAs negatively correlated in expression level with their corresponding miRNAs are considered as putative target mRNAs. The negative correlation analysis indicates that miR-23a, miR-26a, miR-338-5p, miR-491-3p, miR-107, miR-124, miR-128a/128b, miR-138, miR-149, miR-181a, miR-181b and miR-302b have more putative target mRNAs than others and PLAGL2, HNRNPA2B1, RAP1B, DEK, etc have more corresponding miRNAs. The negative correlation between RAP IB and miR-149 in expression level has been validated by Real-time PCR. These results implied that one miRNA can regulate several mRNAs while one mRNAs can be regulated by several miRNAs. Taken together, the miRNAs and genes mentioned may play an important role in the initiation and progression of gliomas.[Conclusion]We have analyzed simultaneously the expression profiles of differentially expressed miRNAs and mRNAs in gliomas with different pathological grades and obtained some preliminary conclusions, which contribute to the further clarification of the molecular mechanisms involved in the tumorigenesis of gliomas and the finding of diagnostic and deteriorated markers in gliomas.
|
PDF Full Text Request