【作者】 杨盈赤；

【导师】 赵玉沛； 吴元德； 郭俊超；

【作者基本信息】 中国协和医科大学 ， 外科学， 2010， 博士

【摘要】 目的：胰腺癌的发病率逐年上升,其致死率已经位居消化道恶性肿瘤的第二位。胰腺癌早期诊断非常困难,治疗效果及预后极差,根本原因在于目前对其细胞起源、发生和发展等机制尚不清楚。在本实验室的前期研究,诱导建立的大鼠胰腺癌模型是接近人胰腺导管腺癌的体内模型,通过观察其胰腺癌发生过程的形态及组织学变化,为认识胰腺癌细胞起源提供了最新的实验证据。但为了进一步探讨胰腺癌发生和发展过程的细胞分子生物学调控机制,本课题拟在诱导大鼠胰腺癌模型过程中,收集不同时间点组织标本,通过基因芯片技术探讨胰腺癌形成和发展过程中相关基因表达调控的动态改变,以期能从细胞分子生物学水平阐明胰腺癌细胞发生和发展机制,并可以为胰腺癌的早期诊断和基因治疗提供新的线索。方法：1.通过化学致癌剂二甲基苯并蒽(DMBA)胰腺原位包埋诱导建立SD大鼠胰腺癌动物模型。2.利用Affymetrix大鼠全基因组表达谱芯片(Genechip Rat Expression Set 230)对胰腺癌大鼠动物建模过程中的各时间点的胰腺组织进行检测,初步筛选其差异表达基因。3.应用Genespring软件对差异表达基因进行功能分析,根据Gene Ontology (GO)对所筛选出的差异表达基因进行分子功能及生物学过程的分类,并根据其基因表达的模式运用层次聚类分析的方法进行分析。按照固定顺序对差异表达基因进行自组织映射(SOM)聚类分析。4.应用Mann-Kendall单调趋势检验分析基因芯片检测原始数据,并通过将人类全基因组和大鼠全基因组进行对比,筛查胰腺癌大鼠动物建模过程中的各时间点的人鼠同源的差异表达基因。5.分别采用荧光实时定量PCR技术及免疫组织化学染色方法对芯片检测结果中经过分析确定的有重要意义的基因进行：mRNA及蛋白表达水平的验证,以确定芯片检测结果的可靠性。结果：1.采用5mg剂量的二甲基苯并蒽(DMBA)直接置入胰腺被膜下的实质内的方法成功诱导大鼠胰腺癌发生,成癌率高,非实验性意外死亡率低。DMBA实验组术后1个月癌发生率达到80%(12／15)并伴有2例高级别]PanIN。DMBA实验组术后3个月大鼠癌发生率为100%(14／14),死亡率仅为6%(1／15)。2.基因芯片技术对比检测正常胰腺、DMBA实验组7天、2周、1月和3月组间基因总体表达水平的改变情况。在所检测的22000个转录本中,共有661条基因在5组中表达差异均有统计学意义(P<0.05),经组间两两比较显示,随着差异倍数的增加,差异基因的数目随之减少。3.根据GO分类,所筛选出的661条差异表达基因涉及不同的分子功能分类,并参与了多个生物学过程。根据差异表达基因表达谱的相似性,通过双向层次聚类分析,能从基因表达谱上完全区分开正常胰腺、急慢性胰腺炎、Pan IN和早期及进展期胰腺癌五组样本。其中正常胰腺和进展期胰腺癌的分叉最远,说明相似度最低,提示这些差异表达基因与胰腺癌发生之间有非常高的相关性。4.只保留25%样本表达量超过100个荧光单位且在所有样本范围内IQR超过0.5的2,786个探针。进行Mann-Kendall单调趋势检验,获得上调基因(探针)11个,下调基因(探针)142个(P<0.05)。成功筛查出了胰腺癌大鼠动物建模过程中的各时间点的人鼠同源的差异表达基因。5.采用荧光实时定量PCR技术及免疫组织化学染色方法对CXCR7、ATP6vlg2和UBe2c分别进行了mRNA及蛋白水平的验证,显示实时定量PCR的结果与芯片结果具有良好的一致性,蛋白表达水平也与芯片检测结果基本一致。进一步证实了本研究中基因芯片检测数据及分析的可靠性。结论：1.可在短期内获得较高发生率的大鼠胰腺癌模型从而获得胰腺癌发生过程中不同时期的大鼠胰腺癌组织标本,使动态研究胰腺癌发生发展过程中细胞分子生物学的调控机理成为可能。2.应用大鼠全基因组的表达谱芯片,高通量、动态地筛选出胰腺癌发生及发展过程中差异表达的基因。3.结合胰腺癌大鼠动物模型建模过程中不同时间点的病理状态,通过对部分已知基因表达相关功能的分析,发现胰腺癌发生中涉及到炎症反应、趋化因子、细胞应激反应的增加及抗氧化损伤能力的增强、肿瘤细胞凋亡的抑制、调控增殖与生长抑制的信号途径的失衡、细胞周期及有丝分裂的紊乱以及细胞骨架结构变化等众多遗传学改变,并且这些改变在胰腺癌发生发展过程的不同时段有不同的特点,其共同作用的结果可能与胰腺癌发生密切相关。4.筛查了大鼠胰腺癌发生和发展过程中动态的基因表达差异。并通过Blast对比成功筛查出了胰腺癌大鼠动物建模过程中的各时间点的人鼠同源的差异表达基因,对于将研究过渡到人类胰腺癌细胞株层面进一步实现对胰腺癌患者的研究有重要意义。5.通过实时定量PCR检测结果和蛋白表达的检测结果表明利用基因芯片技术可完成本实验要求的基因差异表达的筛查工作。同时验证了CXCR7和UBe2c与肿瘤的发生密切相关,并可促进细胞增殖和恶性转化,提示CXCR7和UBe2c可能与胰腺癌的发生和进展相关,对这些基因的进一步研究将可能为胰腺癌发生的细胞分子机理及治疗提供新的线索。更多还原

【Abstract】 Objective:The incidence of pancreatic cancer increases year by year, and the death rate has been ranked second in gastrointestinal cancer. Early diagnosis of pancreatic cancer is very difficult, simply because the mechanisms of its occurrence and development are not clear. In preliminary studies in our laboratory, we had successfully established the rat model of pancreatic cancer. This in vivo model is similar to the process of human pancreatic cancer. By observing its morphological and histological changes, we provided the latest experimental evidence to understand the occurrence of pancreatic cancer. In this issue, we plan to collect the tissue samples at different time points in the process of inducing rat model of pancreatic cancer. Then we investigate the gene expression profile among those samples in order to further explore the molecular biologic mechanism of the occurrence and development of pancreatic cancer cell and offer new clues for early diagnosis and gene therapy of pancreatic cancer.Method:1. Establish SD rat animal model of pancreatic cancer induced by chemical carcinogen DMBA.2. Gene expression profile was screened in all time points tested of rat model of pancreatic cancer using oligonucleotide microarray (Affymetrix Genechip Rat Expression Set 230).3. Genespring software was applied in functional analysis of differentially expressed genes, according to Gene Ontology (GO) in molecular function and biological process categories. And according to their gene expression patterns, hierarchical cluster analysis and self-organizing map clustering (SOM) were analyzed.4. Microarray raw data were analyzed by using Mann-Kendall trend monotone test of human and rat whole genome were compared in order to get homologous differentially expressed genes of rat and human.5. Fluorescence real-time quantitative PCR and immunohistochemical staining were used to exam levels of mRNA and protein validation of the significance differentially expressed gene to determine the reliability of microarray results.Results:1. Successfully induced the rat models of pancreatic cancer by using 5mg dose DMBA with low non-experimental accident death rate. After 1 month, the incidence of cancer occurrence was 80%(12/15) in DMBA experimental group, accompanied by two cases of high-level Pan IN. After 3 months, the incidence of cancer occurrence was 100%(14/14) in DMBA experimental group, mortality was only 6%(1/15).2. A comprehensive, differential gene expression profile was obtained for normal pancreas, DMBA experimental groups after 7 days,2 weeks,1 month and 3 months. Totally,661 genes group changes in the overall situation.3. According to GO classification, the selected 661 differentially expressed genes involved in different molecular function categories, and participated in many biological processes. According to the similarity of differentially expressed gene expression by using two-way hierarchical clustering analysis, we could completely discriminate the normal pancreas, acute and chronic pancreatitis, Pan IN and early and advanced pancreatic cancer samples.4. There were 11 up-regulated genes (probe) and 142 down-regulated genes (probe) (P <0.05) by Mann-Kendall trend Monotone test. We successfully screened out the homologous genes of rat and human at different time points in the process of rat models of pancreatic cancer. 5. The real-time quantitative PCR and immunohistochemical staining of CXCR7 and UBe2c revealed a similar expression pattern to microarray results.Conclusion:1. Obtaining the different stages of pancreatic cancer tissue samples in rat models made the dynamic study for the molecular biologic mechanism of occurrence and development of pancreatic cancer possible.2. Obtaining differentially expressed genes dynamically and high throughput during the development of pancreatic cancer by application of the expression of rat whole genome microarrays.3. Combined with rat models of pancreatic cancer at different time points and the different pathological state, the global transcriptome profiling shows that occurrence and development of pancreatic cancer is a complicated and multifactorial process and may involve some of all of the following changes:inflammation, chemokines, cell stress response increased and enhanced resistance to oxidative damage, inhibition of tumor cell apoptosis, regulation of proliferation and growth inhibition of the signaling pathways of balance, cell cycle and mitosis of the disorder as well as structural changes in the cytoskeleton, and many other genetic changes. And these changes in pancreatic cancer development and progression of different periods have different characteristic, the results may be closely related to occurrence and development of pancreatic cancer.4. Successfully screening the homologous genes of rat and human was significance for the further study for the molecular biologic mechanism of occurrence and development of pancreatic cancer in human pancreatic cancer cell lines and pancreatic cancer patients.5. Suggesting that CXCR7 and UBe2c may be related to the occurrence and development of pancreatic cancer, further studies of these genes may provide new clues of molecular mechanisms and treatment of pancreatic cancer.更多还原