基于网络方法探讨miRNA在脑肿瘤中的作用机制及应用价值

发布时间：2018-05-29 20:58

本文选题：微小RNA + 生物信息学　；参考：《兰州大学》2017年硕士论文

【摘要】：背景:颅内肿瘤又被称为脑肿瘤或者颅脑肿瘤,指发生于颅腔内的神经系统肿瘤,分为恶性肿瘤和良性肿瘤。根据2016年WHO中枢神经系统分类概述将其分为:胶质瘤;室管膜肿瘤;脉络从肿瘤;神经元和混合性神经元-胶质瘤;松果体区肿瘤;胚胎类肿瘤;如髓母细胞瘤;颅神经和椎旁神经肿瘤;脑膜瘤;间叶,非脑膜皮型脑膜瘤;黑色素细胞肿瘤;淋巴瘤;生殖细胞肿瘤;鞍区肿瘤;转移瘤等。其中胶质瘤又可以根据组织细胞来源分为:弥漫性星形细胞和少突胶质细胞瘤;其他星形细胞肿瘤;室管膜肿瘤;其他胶质瘤;神经元和混合性神经元-胶质瘤等不同的类型。根据2016年最新的中枢神经系统肿瘤分类建立了分子时代CNS肿瘤诊断的新概念。结合基因特征,调整部分胶质瘤和髓母细胞瘤的的诊断。将IDH-突变型、IDH-野生型、H3 K27M-突变、IDH-突变和1p19q联合缺失等基因特征加入胶质瘤分级标准中。髓母细胞瘤在传统的组织学分类如典型的髓母细胞瘤、多纤维性/结节增生性髓母细胞瘤、广泛小结节性髓母细胞瘤、大细胞性髓母细胞瘤、非特指性髓母细胞瘤的基础上引入分子病理特征,将WNT活化型、SHH活化型和TP53突变型、SHH活化型和TP53野生型、非WNT/非SHH活化型等分子病理特征应用于髓母细胞瘤的分类和诊断中。III级胶质瘤作为中枢神经系统恶性脑肿瘤,主要分为间变型星形细胞瘤,IDH-突变型、间变少突神经胶质瘤,IDH-突变和1p19q联合缺失型、间变型多形性黄色星形细胞瘤、间变型节细胞胶质瘤。微小RNA(microRNA,miRNA)是存在于植物、动物和一些病毒中小的非编码RNA分子(含有约22个核苷酸),其在RNA沉默和基因表达的转录后调节中起作用。虽然大多数miRNA位于细胞内,但也已经在细胞外环境中发现了一些通常称为循环mi RNA或细胞外miRNA的miRNA分子,包括各种生物体液和细胞培养基。miRNA是RNA干扰的小内源性介质,并且是有机体发育,细胞专一化和体内平衡所需的许多生物过程的关键调节组分。miRNA的表达与许多严重疾病的生理以及病理过程相关,其中包括脑肿瘤。目前尚不清楚miRNA是否与肿瘤的级别、分子亚型的特异性以及肿瘤的预后有关。到目前为止,已经在几乎所有种类的生物体液中发现了微小RNA,包括脑脊液(CSF)。许多研究已经显示在中枢神经系统(CNS)肿瘤患者的组织以及体液中有异常表达的mi RNA,并且有学者提出miRNA可以作为中枢神经系统肿瘤生物标志物。目的:(1)通过加权基因通表达网络(WGCNA)以及cytoscape软件构建III级胶质瘤的无尺度共表达网络,探讨其发生机制,寻找关键基因;(2)结合样本的临床数据,找到与预后相关的miRNA和基因;(3)通过分析髓母细胞瘤差异表达的mi RNA,基于Ingenuity Pathway Analysis(IPA)软件识别髓母细胞瘤潜在的调控网络以及髓母细胞瘤发生发展过程中的关键信号分子;(4)基于生物信息学算法,筛选出颅内常见肿瘤脑脊液中差异表达的miRNA。通过聚类分析,找到可以鉴别脑膜瘤、低级别胶质瘤、胶质母细胞瘤、转移瘤、髓母细胞瘤以及淋巴瘤的的miRNA,选取部分miRNA分子通过荧光定量PCR技术验证其可靠性,从而挖掘mi RNA在脑肿瘤临床术前诊断中的应用价值。方法:(1)使用TCGA-Assembler下载TCGA数据库3级RNASeqV2基因表达数据,miRNA-seq数据以及样本的临床信息。去除表达量接近零的数据,找到正常组与III级胶质瘤差异表达的基因(different expression genes,DEG)和miRNA,使用R软件(3.3.0)中的“DESeq”包来鉴定foldChange2.0且调整的P值0.05的差异表达基因(DEG)和miRNA。通过加权基因通表达网络(WGCNA)以及cytoscape软件构建III级胶质瘤的无尺度共表达网络,并通过基因功能GO富集以及KEGG通路分析探讨其发生机制以及关键基因;(2)通过整合样本的临床数据,找到与预后相关的临床标志物;(3)从NCBI GEO数据库下载髓母细胞瘤mi RNA矩阵文件并执行log2变换。使用线性模型和Bayes检验计算差异表达基因(DEG)。利用热图,可视化差异表达基因。将差异表达的miRNA导入到IPA(Ingenuity Pathway Analysis)软件,使用microRNA Target Filter找到靶向信息。然后使用IPA核心分析模块分析差异表达microRNA相关的网络功能;(4)从GEO数据库下载中枢神经系统肿瘤脑脊液的miRNA表达谱数据,并通过R语言的limma包进行归一化并且寻找差异表达基因(DEG)。通过聚类分析,找到可以正确聚类样本的mi RNA分子。最后,留取兰州大学第二医院神经外科临床中心肿瘤患者的脑脊液标本,通过设计选定的miRNA分子,设计特异性引物,通过荧光定量PCR验证标志物的可靠性。结果:(1)数据结果显示在III级胶质瘤中,有2036个差异表达的mRNA和50个mi RNA,并且,2036个差异基因可以聚类为5个模块,这些模块主要与细胞周期、有丝分裂、微管蛋白的结合、递质传导运输以及p-53信号通路等有关。如在网络中所示,BUB1B,KIFC1,TOP2A,BUB1,SLC12A5,ESCO2,ESPL1,EPR1,KIF15,CASC5,SGOL1,NUSAP1,CCNB2,NUF2,TTK,KIF2C在共表达网络的中心。并且,BUB1B,KIFC1,TOP2A,BUB1,ESPL1,EPR1是过表达基因网络的中心;SLC12A5,VSNL1,SULT4A1,TMEM130,SNAP25位于低表达基因网络的中心。利用差异表达的mRNAseq和miRNAseq数据建立共表达网络,SLC12A5,MAL2,VSNL1,A2BP1,EPB49,SULT4A1,TMEM130,ADAM11,SNAP25,C1orf115,DNM1,SYT1位于网络的中心,并且mir-128,mir-129也参与其中。我们可以假设网络中心的基因就是关键基因并参与III级胶质瘤的重要病理过程;(2)根据样本的临床数据以及标准化后的基因表达数据,我们发现KIF4A,NCAPG,SGOL1,KLK7,SULT4A1和TSHR等基因与预后密切相关。Mir-10b,mir-27a,mir-329-1和mir-138-2等miRNA分子也与III级胶质瘤的预后密切相关;(3)根据GEO数据库的miRNA矩阵数据,我们发现髓母细胞瘤中48个显着差异的表达mi RNA。IPA核心分析显示髓母细胞瘤中28个差异表达的mi RNA与“有组织损伤和异常,生殖系统疾病,癌症”调控网络相关,并且TP53,AGO2,ERK1/2,SIRT1和YBX1基因在髓母细胞瘤的发生发展过程中起到了重要作用;(4)根据脑脊液miRNA芯片分析,mir-205,mir-664和mir-148可以作为生物标志物来区分转移与其他颅内肿瘤。MiR-21,miR-16,mi R-125b,miR-223和miR-142-3p在胶质母细胞瘤患者脑脊液中上调。Mi R-451,mi R-142-3p,miR-25,miR-15a,mi R-16和miR-144在低级别胶质瘤患者脑脊液中高度上调。我们认为mir-21,mir-16在许多恶性脑肿瘤中上调,与低级胶质瘤相比,mir-125b可以是胶质母细胞瘤或差的预后胶质瘤的特异性生物标志物。相对于正常和其他癌症组,miR-711和mir-886-3p在原发性中枢神经系统淋巴瘤中下调。与正常组相比,淋巴瘤组具有更低表达水平的miR-711。此外,与胶质母细胞瘤和髓母细胞瘤相比,mir-886-3p在淋巴瘤组中下调。结论:(1)生物信息学分析为高级别胶质瘤机制的研究提供了一种新的研究方法;(2)临床预后研究应该是多维度的,尤其是恶性肿瘤的预后不仅与样本的病理分级有关,更重要的是与特殊的分子或者基因标志物密切相关,并且这些信号分子可能在恶性肿瘤的发生和发展过程中起了重要作用;(3)髓母细胞瘤中,miRNA可能通过作用于TP53、AGO2、ERK1/2、SIRT1和YBX1在髓母细胞瘤的发生发展中起了重要作用;(4)MiRNA具有作为中枢神经系统肿瘤非侵入性检测标志物的巨大潜力。但是,目前的microRNA标志物还不能做出准确的诊断。mi RNA测定似乎在淋巴瘤和转移性脑癌的诊断中比在胶质瘤和其他颅脑肿瘤中更敏感。一方面,仍需要基于更大样本量的实验进行更进一步的验证。另一方面,随着二代测序技术的发展,将出现新mi RNA分子以提高CNS疾病诊断的准确性和可靠性。
[Abstract]:Background: intracranial tumors, also known as brain tumors or craniocerebral tumors, refer to neurologic tumors occurring in the cranial cavity, divided into malignant tumors and benign tumors. According to the classification of WHO central nervous system in 2016, it is divided into glioma, ependymal tumor, choroid from tumor; neurotransmitter and mixed neuron glioma; pineal region tumor; Embryoid tumors; such as medulloblastoma; cranial nerve and paravertebral nerve tumor; meningioma; interleaf, non meningocutaneous meningioma; melanocytic tumor; lymphoma; germ cell tumor; sellar tumor; metastatic tumor; among which gliomas can be divided into diffuse astrocytes and oligodendrocytes; other astrocytomas; other astrocytomas; Cell tumors; ependyma tumors; other gliomas; neurons and mixed neurons - gliomas. A new concept for the diagnosis of CNS tumors in the molecular age was established based on the latest central nervous system tumor classification in 2016. Combined with gene characteristics, the diagnosis of partial glioma and medulloblastoma was adjusted. IDH- mutant, IDH- Gene features such as wild type, H3 K27M- mutation, IDH- mutation and 1p19q joint deletion are added to the glioma grading standard. Medulloblastoma is classified as a typical medulloblastoma, fibroblastoma, nodular medulloblastoma, large cell medulloblastoma, and non special medulloblastoma in traditional histologic classification. On the basis of the cytomellus, the molecular pathological features are introduced, and the molecular pathological features such as WNT activation, SHH activation and TP53 mutation, SHH activation and TP53 wild type and non WNT/ non SHH activation type are applied to the classification and diagnosis of medulloblastoma as malignant brain tumors of the central nervous system, which are mainly divided into inter variant astrocytoma, I DH- mutant, oligodendroglioma, IDH- mutation and 1p19q joint deletion, cross variant polymorphic yellowish astrocytoma, alternating type of glioma. Small RNA (microRNA, miRNA) is a non coded RNA molecule (containing approximately 22 nucleotides) in plants, animals, and some viruses (containing about 22 nucleotides), which is silenced in RNA and gene expression. Although most of the miRNA is located in the cell, it has also found some miRNA molecules commonly known as circulating mi RNA or extracellular miRNA in the extracellular environment, including a variety of biological fluids and cell culture medium.MiRNA as a small endogenous medium for RNA interference, and is an organism development, cell specificity and body level. The expression of.MiRNA, a key regulatory component of many biological processes required, is related to the physiological and pathological processes of many serious diseases, including brain tumors. It is not clear whether miRNA is related to the tumor level, the specificity of the molecular subtype, and the prognosis of the tumor. Small RNA, including cerebrospinal fluid (CSF), has been found in the fluid. Many studies have shown the abnormal expression of MI RNA in the tissues and body fluids of the central nervous system (CNS) tumor patients, and some scholars have suggested that miRNA can be used as a biomarker for the tumor in the central nervous system. (1) through the weighted gene expression network (WGCNA) and cytoscap E software constructs a scale-free co expression network of III glioma to explore its mechanism and find key genes; (2) to find the miRNA and gene related to prognosis in combination with the clinical data of the samples; (3) the potential regulation of medulloblastoma based on Ingenuity Pathway Analysis (IPA) software is identified by analyzing the differential expression of MI RNA in medulloblastoma. Network and the key signal molecules in the development of medulloblastoma; (4) based on the bioinformatics algorithm, the differential expression of miRNA. in the cerebrospinal fluid of the common intracranial tumors was screened by clustering analysis to find miRNA for the identification of meningioma, low grade glioma, glioblastoma, metastatic tumor, medulloblastoma, and lymphoma. Some miRNA molecules are selected to verify their reliability by fluorescence quantitative PCR technology, so as to discover the value of MI RNA in the clinical diagnosis of brain tumors. (1) use TCGA-Assembler to download the 3 level RNASeqV2 gene expression data of the TCGA database, miRNA-seq data and the clinical information of the samples. The differentially expressed genes (different expression genes, DEG) and miRNA in normal group and III glioma were identified using the "DESeq" package in R software (3.3.0) to identify the differentially expressed genes of foldChange2.0 and the adjusted P value 0.05 (DEG) and the free ruler to construct a glioma. Degree co expression network, and through gene function GO enrichment and KEGG pathway analysis to explore the pathogenesis and key genes; (2) the clinical data of the integration of samples to find clinical markers related to the prognosis; (3) Download medulloblastoma mi RNA matrix files from the NCBI GEO database and perform log2 transformation. Linear model and Bayes test. Calculate the differential expression gene (DEG). Use the heat map to visualize the differentially expressed genes. Introduce the differentially expressed miRNA into the IPA (Ingenuity Pathway Analysis) software, use the microRNA Target Filter to find the target information. Then use the IPA core analysis module to analyze the differential expression microRNA related network functions; (4) download from the GEO database. The miRNA expression data of the cerebrospinal fluid of the armature nervous system were normalized and the differential expression gene (DEG) was searched through the limma package of the R language. By cluster analysis, the MI RNA molecules that could be correctly clustered were found. Finally, the cerebrospinal fluid specimens of the patients in the clinical center of the Department of Neurosurgery, Second Hospital Affiliated to Lanzhou University, were set up by the establishment of the cerebrospinal fluid specimens. The selected miRNA molecules designed specific primers and verified the reliability of the markers by fluorescence quantitative PCR. Results: (1) data showed that there were 2036 differentially expressed mRNA and 50 mi RNA in III grade gliomas, and 2036 differentially different genes could be clustered into 5 blocks. These modules were mainly divided into cell cycle, mitosis, microtubule eggs BUB1B, KIFC1, TOP2A, BUB1, SLC12A5, ESCO2, ESPL1, EPR1, KIF15, CASC5, as the center of the network. A1, TMEM130, and SNAP25 are located at the center of the low expression gene network. Using the differentially expressed mRNAseq and miRNAseq data to establish a co expression network, SLC12A5, MAL2, VSNL1, A2BP1, EPB49, SULT4A1, TMEM130, and also participate in it. We can assume the gene in the network center. (2) we found that the genes of KIF4A, NCAPG, SGOL1, KLK7, SULT4A1 and TSHR are closely related to the prognosis, and the miRNA molecules such as mir-27a, mir-329-1 and mir-138-2 are also closely related to the prognosis of gliomas based on the clinical data of the samples and the standardized gene expression data. (3) according to the miRNA matrix data of the GEO database, we found that 48 significant differences in the expression of MI RNA.IPA core analysis in medulloblastoma showed that 28 differentially expressed mi RNA in medulloblastoma were associated with "organized injury and abnormality, reproductive system disease, cancer" modulation network, and TP53, AGO2, ERK1/2, SIRT1, and YBX1 genes. It plays an important role in the development of medulloblastoma; (4) mir-205, mir-664, and mir-148 can be used as biomarkers to distinguish metastatic and other intracranial tumors from.MiR-21, miR-16, MI R-125b, miR-223 and miR-142-3p in the cerebrospinal fluid of the patients with gelatinous blastoma, according to the miRNA chip analysis of cerebrospinal fluid. MiR-25, miR-15a, MI R-16 and miR-144 are highly up-regulated in the cerebrospinal fluid of patients with low grade glioma. We think that miR-21, mir-16 is up-regulated in many malignant brain tumors. Compared with low grade gliomas, miR-125b can be a specific biomarker for glioblastoma or poor prognosis glioma. MiR-711 relative to normal and other cancer groups. And mir-886-3p was downregulated in primary CNS lymphoma. Compared with the normal group, the lymphoma group had a lower expression of miR-711.. Compared with glioblastoma and medulloblastoma, mir-886-3p was downregulated in the lymphoma group. Conclusion: (1) bioinformatics analysis provides a kind of high grade glioma mechanism. New methods of research; (2) the study of clinical prognosis should be multidimensional, especially the prognosis of malignant tumors not only related to the pathological classification of samples, but also closely related to special molecules or gene markers, and these signal molecules may play an important role in the development and development of malignant tumors; (3) medullary medullary In cytomas, miRNA may play an important role in the development of medulloblastoma by acting on TP53, AGO2, ERK1/2, SIRT1 and YBX1; (4) MiRNA has great potential as a noninvasive marker of the central nervous system tumor. However, the current microRNA markers can not make a accurate diagnosis of.Mi RNA as if in lymph nodes. The diagnosis of tumor and metastatic brain cancer is more sensitive than in gliomas and other craniocerebral tumors. On the one hand, further tests based on larger samples are still needed. On the other hand, with the development of the two generation sequencing technology, new mi RNA molecules will appear to improve the accuracy and reliability of the diagnosis of CNS disease.
【学位授予单位】：兰州大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R739.4

【相似文献】