肝细胞癌和前列腺癌的分子分型标记物研究

发布时间：2018-05-20 19:50

本文选题：肝细胞癌(Hepatocellular + Carcinoma　；参考：《浙江大学》2016年博士论文

【摘要】：近年来,下一代测序技术(Next generation sequencing, NGS)迅猛发展,加速了肿瘤基因组学研究,帮助科学家找到了许多肿瘤特有的突变模式和驱动因子。生物信息技术对转录组测序、外显子测序和全基因组测序得到海量数据的快速而精准的分析促进了NGS的发展。NGS应用于肿瘤基因组学的研究揭示了许多不同类型癌症的基因组和转录组的突变图谱；帮助了科学家认识到肿瘤的异质性,还有肿瘤发生、发展以及对药物抵抗的分子机制。此外,这些研究有助于癌症的分子亚型的鉴定和划分。本研究以肝细胞癌和前列腺癌为例,证明了下一代测序技术在癌症分子机制研究和分子分型的个性化治疗中的应用的可行性。本文的第一部分第一章,我们对三例长期感染HBV肝癌患者的癌组织、癌旁组织和其正常血液样本进行全基因组测序。通过借助Complete Genomics全基因组测序平台,我们鉴定了体细胞单核苷酸突变(Somatic single-Nucleotide Variation, SNV)包括单核苷酸多态性(Single Nucleotide Polymorphism, SNP)、小片段插入或缺失(Insertion or Deletion, Indel),染色体结构变异(Structural Variation),还有基因拷贝数变化(Copy Number Variation,CNV)。我们分析比较了癌组织、癌旁组织和血液样本的所有的突变类型并划分了三个组别进行具体分析：癌特有的突变、癌旁特有的突变以及癌和癌旁共有的突变(血液样本没有)。据我们所知,这是目前首次针对HBV感染肝癌患者的癌、癌旁和血液样本同时进行全基因组测序和比较分析。我们鉴定得到癌特有的突变、癌旁特有突变以及癌和癌旁共有(血液没有)的突变位点以及重要生物学信号通路,对肝细胞癌发生和发展的分子机制的研究具有重要意义。在针对第一章的高通量测序的结果基础上,本文的第一部分第二章在测序鉴定的突变结果中筛选了多个在以往文献报道的肿瘤相关的单核苷酸多态性位点(SNP)和新发现的SNP以及结构变异(SVs)在大量肝细胞病人的癌、癌旁和血液样本中验证。在验证的结果中,我们发现抑癌基因—组蛋白H3K4三甲基化转移酶KMT2C (Lysine-Specific Methyltransferase 2C),又称MLL3 (mixed-lineage leukemia 3, MLL3) (C1114R)突变在肝细胞癌中非常普遍(癌、癌旁和血液的突变频率分别是97.8%,96.2%和76.9%)。致癌基因VCX(L104P)是癌和癌旁共有,血液中没有的突变频率最高(癌和癌旁中突变频率分别是14.6%和11.1%)的SNP位点。抑癌基因TP53 (R249S)位点在7.7%的癌组织中发生特异性突变,且该突变和不良预后显著相关。我们还发现巨型蛋白AHNAK核蛋白2CAHNAK2)基因M1761I突变在100%的癌和癌旁中发生突变,在10%的HCC患者血液中发生突变。锌指蛋白ZNF717的L689H突变在癌、癌旁和HCC患者血液里都发生突变(突变率分别是34%、23%和10%)多聚腺苷酸结合蛋白PABPC3 Y377H位点和人类白细胞Ⅱ型抗原HLA-DQB1的S233G位点在验证的HCC的在癌、癌旁和血液中100%检测到。染色体结构变异方面,我们发现17号染色体TK1胸苷激酶1 (Tyrosine Kinase 1)基因和8号染色体的非基因片段在癌和癌旁发生高频融合(融合频率分别是29%和16%)。整合分析癌和癌旁样本外显子和剪切位点的SNV对生物学信号通路的影响,我们发现以下2条信号通路在癌中发生特异性的显著变化：胞外基质受体互作(ECM-receptor interaction)和细胞粘附分子CAMs (Cell adhesion molecules)表明这些肿瘤微环境相关通路的改变可能是HCC发展的主要驱动力。然而,嗅觉诱导(Olfactory transduction)是在癌旁中发生特异变化的信号通路,预示着它可能与HBV介导的HCC的启动相关。我们的研究表明,HBV介导的HCC的发生和发展过程共性和个性并存。我们找到的癌组织和癌旁组织发生的重要突变位点和生物学信号通路,可能是HCC发生和发展的关键驱动因子,对HCC的早期诊断和治疗具有重要意义。本文的第二部分,我们对74例(50例FFPE样本,24例新鲜组织样本)接受前列腺根治术的低级别原位前列腺癌(Gleason=7)的样本(44例未生化复发和30例生化复发)进行全外显子组测序。针对外显子测序找到的复发和未复发组的体细胞突变,我们用随机森林机器学习算法鉴定能够区分复发和未复发的标志物。同时,鉴定了PCa复发相关的显著突变位点、突变基因以及关键生物学过程。通过着重比较了复发组和未复发组的前列腺癌病人的外显子组突变图谱差异,我们找到了33个和复发特异相关的体细胞显著突变基因,排前6位的是STK31、ALMS1、PCSK5、 AHRR和NCOR2。33个显著突变基因绝大部分都是首次被报道。在复发前列腺癌样本显著突变的基因中,ALMS1 (p.E15delinsEE)和NCOR2 (p.Q78delinsQQ)各自存在高频的非移码插入(分别为46%和43%)。MAP3K9的第一个外显子的113位发生CCT碱基非移码缺失(p.38_39del),缺失频率为37%; KDM6B基因的第9个外显子的796位发生ACC缺失(p.252_253del),缺失频率为17%。此外,20%的复发样本中IDI2基因的第5个外显子402位发生终止获得突变(p.Y134X)。这些突变都很大程度上可能对蛋白质的表达产生影响,可能参与重要的肿瘤发展、转移相关的生物学通路。使用随机森林机器学习算法,鉴定到22个突变集,用于高效区分发展缓慢和发展快速的前列腺癌样本,是复发预测的潜在标志物。METS影响巨噬细胞分化(METS affect on Macrophage Differentiation)是复发组特有的显著变化信号通路。在复发的PCa中,该通路的关键基因：NCOR2、HDAC2和METS发生显著突变,可能进一步抑制TAMs的增殖,促进其分化为M2型,从而促进肿瘤细胞的侵袭和周围炎症反应。第二部分研究揭示了PCa复发组和未复发组的基因突变图谱,鉴定到复发相关的重要突变、基因和生物学通路。此外,研究还找到了能够区分发展缓慢和发展快速的突变集,作为复发预测的候选标志物。该研究对前列腺癌的分子分型和个性化治疗具有重要意义。
[Abstract]:In recent years, the rapid development of Next generation sequencing (NGS) has accelerated the study of tumor genomics, helping scientists find a number of tumor specific mutation patterns and driving factors. Bioinformation technology has been sequenced, exon sequencing and whole genome sequencing to get a rapid and accurate score of massive data. The development of NGS's development.NGS applied to tumor genomics has revealed the mutation map of many genomes and transcripts of different types of cancer; it has helped scientists to recognize the heterogeneity of tumors, and the molecular mechanisms of tumorigenesis, development and drug resistance. In addition, these studies contribute to the molecular subtypes of cancer. Identification and division. In this study, the feasibility of the application of next generation sequencing in molecular mechanism research and individualized treatment of molecular typing was demonstrated in the case of hepatocellular carcinoma and prostate cancer. Chapter 1, Chapter 1 of this article, three cases of cancer tissue, para cancer tissue and normal blood samples of three patients with long-term infection of HBV liver cancer. Complete genome sequencing. By using Complete Genomics whole genome sequencing platform, we identified somatic single nucleotide mutation (Somatic single-Nucleotide Variation, SNV) including single nucleotide polymorphisms (Single Nucleotide Polymorphism, SNP), small fragment insertion or deletion (Insertion or), chromosome structure Variation (Structural Variation) and gene copy number change (Copy Number Variation, CNV). We analyzed and compared all types of mutations in cancer tissue, para cancer tissue and blood samples and divided three groups into specific analyses: cancer specific mutations, paracancerous mutations, and cancer and common mutations near Cancer (blood samples) As far as we know, it is the first time that this is the first time to sequence and compare the whole genome of the cancer of the HBV infected patients with liver cancer, the para cancer and the blood samples. We have identified the cancer specific mutations, the paracancerous endemic mutations, the mutation sites of the cancer and the side of the cancer and the important biological signal pathways, and the liver cells. The study of the molecular mechanisms of carcinogenesis and development is of great significance. On the basis of high throughput sequencing in the first chapter, the first part of this article, in the second chapter, screened a number of tumor related single nucleotide polymorphisms (SNP) and newly discovered SNP and structural changes in the previously reported mutation results. SVs is verified in a large number of patients with hepatocellular carcinoma, para cancer and blood samples. In the results, we found that the tumor suppressor gene, the histone H3K4 tri methylation transferase KMT2C (Lysine-Specific Methyltransferase 2C), also known as MLL3 (mixed-lineage leukemia 3, MLL3) (C1114R) mutation is very common in hepatocellular carcinoma (cancer, adjacent to cancer) The frequency of mutation of the blood is 97.8%, 96.2% and 76.9% respectively. The oncogene VCX (L104P) is the common mutation of cancer and cancer, and the highest mutation frequency in the blood (cancer and the mutation frequency in the side of cancer is 14.6% and 11.1% respectively). The tumor suppressor gene TP53 (R249S) site has a specific mutation in 7.7% of the cancer tissue, and the mutation and poor prognosis Significant correlation. We also found mutations in the giant protein AHNAK nucleoprotein 2CAHNAK2 gene M1761I mutation in 100% of cancer and cancer, mutation in the blood of 10% HCC patients. The L689H mutation of the zinc finger protein ZNF717 in the cancer, the side of the cancer and the blood of the HCC patients (the mutation rate is 34%, 23% and 10%) polyadenylate binding eggs The S233G loci of the white PABPC3 Y377H site and human leucocyte type II antigen HLA-DQB1 were detected in the HCC of cancer, adjacent to the cancer and 100% in the blood. The chromosome structural variation, we found that the 17 chromosome TK1 thymidine kinase 1 (Tyrosine Kinase 1) gene and the non gene fragment of chromosome 8 have high frequency fusion between cancer and cancer. The combined analysis of the effects of SNV on the biological signal pathways of the exons and shear sites of the paracancerous and paracancerous samples, we found that the following 2 signalling pathways changed significantly in cancer: extracellular matrix receptor interaction (ECM-receptor interaction) and cell adhesion molecule CAMs (Cell adhesion molecules). The changes in the microenvironment related pathways of these tumors may be the main driving force for the development of HCC. However, the olfactory induction (Olfactory transduction) is a specific signaling pathway in the paracancerous, suggesting that it may be associated with the initiation of HBV mediated HCC initiation. Our study shows that the occurrence and development of HBV mediated HCC is common in common. The important mutation sites and biological signaling pathways that occur in the cancer and paracancerous tissues we find may be the key driving factors for the occurrence and development of HCC, and are important for the early diagnosis and treatment of HCC. In the second part of this article, we accepted the prostate in 74 cases (50 FFPE samples, 24 fresh tissue samples). A total exon group was sequenced in a sample of low grade orthotopic prostate cancer (Gleason=7) (44 cases of non biochemical recurrence and 30 cases of biochemical recurrence). The recurrent and non recurrent somatic mutations found in the recrudescent and non recurrent group were identified. We used random forest machine learning algorithms to identify the markers that could distinguish between recurrent and non recurrence. PCa recurrence related significant mutation sites, mutation genes, and key biological processes were determined. By comparing the mutations in the exons of the prostate cancer patients in the recurrent and non recurrent groups, we found 33 significant mutations in the somatic cells associated with the recurrence specificity, and the top 6 were STK31, ALMS1, PCSK5, AHRR and NC. Most of the OR2.33 significant mutations were reported for the first time. In the genes with significant mutations in the recurrent prostate cancer samples, ALMS1 (p.E15delinsEE) and NCOR2 (p.Q78delinsQQ) respectively have high frequency non graft insertion (46% and 43%).MAP3K9 of the 113 exons of the first exon, CCT base non shift code deletion (p.38_39del). The loss of frequency was 37%; 796 of the ninth exons of the KDM6B gene had ACC deletion (p.252_253del), the frequency of the deletion was 17%., and the fifth exons of the IDI2 gene in 20% of the recurrent samples were terminated by the mutation (p.Y134X). These mutations, to a large extent, may affect the expression of protein and may be involved in important tumors. Development, transfer related biological pathways. Using a random forest machine learning algorithm, 22 mutation sets have been identified to efficiently distinguish between slow development and rapid development of prostate cancer samples. The potential marker of recurrence prediction.METS affects macrophage differentiation (METS affect on Macrophage Differentiation), which is unique to the recurrent group. In the recurrent PCa, significant mutations in the key genes of the pathway, NCOR2, HDAC2 and METS, may further inhibit the proliferation of TAMs and promote its differentiation into M2 type, thus promoting the invasion of the tumor cells and the surrounding inflammatory response. The second part of the study revealed the genetic mutation map of the PCa and the non recurrent groups, and identified the identification of the gene mutations in the recurrent and non recurrent PCa groups. The major mutations, genetic and biological pathways related to recurrence. In addition, the study has also found a mutation set that distinguishes slow development and rapid development as a candidate marker for recurrence prediction. This study is of great significance for molecular typing and individualized treatment of prostate cancer.
【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：R735.7;R737.25

【相似文献】