glioma Through Genomics and Transgenic Mouse Model

描述(申请人提供)：胶质瘤是最常见的原发脑部恶性肿瘤。胶质母细胞瘤或GBM是最高级的胶质瘤，也是最具侵袭性的，占所有胶质瘤的50%-60%。GBM患者的中位生存期不到一年。只有大约20%的患者对治疗有反应，并在首次诊断后存活超过1.5年。过去二十年的深入研究表明，癌症的进展伴随着基因和分子变化的积累，使细胞能够获得至少六种能力(Hanahan和Weinberg，Cell 100：57-70,2000)。来自体外和体内模型系统的证据已经积累，在基因表达水平上识别了其中一些事件。包括转基因和基因敲除模型在内的动物模型已经被用来测试一些已确定的分子事件在活体环境中的重要性(van Dyke和Jack，2002)。例如，EGFR(Holland等人，1998)、PDGF(Dai等人，2002)、K-ras和AKT(Holland等人，2000)的表达已被证明在转基因模型中诱导胶质瘤，这对于筛选肿瘤形成和肿瘤表型中的关键事件具有很高的价值。一些模型，如Holland和Varmus(Holland等人，1998)开发的胶质细胞特异性小鼠模型(RCAS/TV-a系统)，允许在一个实验环境中转基因表达多个基因，从而可以评估基因组合的效果。利用基因组学和信息学方法，我们分析了不同级别胶质瘤的基因表达情况，并确定了在GBM中唯一过表达的基因，如胰岛素样生长因子结合蛋白2(IGFBP2)(Fuller等人，1999)。我们的内部实验表明，过表达IGFBP2的细胞具有更强的侵袭性。IGFBP2也与GBM存活期缩短有关(Sallinen等人，2000年，以及我们未发表的数据)。这些研究使我们假设IGFBP2是一个关键分子，有助于GBM细胞的侵袭和GMB患者的短期生存。我们还假设IGFBP2的过度表达在胶质瘤进展为基底膜的过程中起积极作用。在这个项目中，我们建议使用RCAS/TV-a小鼠模型来验证假设。我们进一步假设，在小鼠中发育的基底膜与人基底膜是同源的，这可以通过比较基因表达谱以及集中的基因集来评估。目的：(1)检测IGFBP2是否促进GBM的发育，缩短GBM小鼠的存活时间，增加GBM细胞的体内侵袭能力。(2)探讨IGFBP2在低级别胶质瘤向GBM转化过程中是否起重要作用。(3)利用基因表达谱对小鼠脑胶质瘤进行鉴定。通过这个项目，我们希望丰富我们关于IGFBP2通路的知识，IGFBP2通路可能在胶质瘤的发生和发展中起重要作用；我们还希望使用基因组学方法来评估小鼠胶质瘤是否在基因表达水平上忠实地代表人类胶质瘤。

DESCRIPTION (provided by applicant): Glioma represents the single most frequent primary brain malignancy. The most advanced form of glioma, also the most invasive, is glioblastoma or GBM that comprises 50-60% of all gliomas. The median survival for GBM patients is less than a year. Only about 20% of patients respond to therapy and live for more than 1.5 years following initial diagnosis. Intensive research during the last two decades has demonstrated that cancer progresses with an accumulation of genetic and molecular alterations that enables cells to acquire at least six capabilities (Hanahan and Weinberg, Cell 100:57-70, 2000). Evidence from in vitro and in vivo model systems has been accumulated that identifies some of those events at gene expression level. Animal models including transgenic and knockout models have been used to test the importance of some of the identified molecular events in in vivo setting (van Dyke and Jacks, 2002). For example, expression of EGFR (Holland et al., 1998), PDGF (Dai et al., 2002), K-ras in combination with akt (Holland et al., 2000), have been shown to induce gliomas in transgenic models, which has been highly valuable to sort out what the key events are in tumor formation and tumor phenotypes. Some models, such as the glial-specific mouse model (the RCAS/tv-a system) developed by Holland and Varmus (Holland et al., 1998) permit transgenic expression of multiple genes in one experimental setting thus the effect of gene combination can be evaluated. Using genomics and informatics approaches, we have profiled gene expression in different grades of gliomas and identified genes, such as insulin-like growth factor binding protein 2 (IGFBP2), that are uniquely overexpressed in GBM (Fuller et al., 1999). Our in intro experiments showed that IGFBP2 overexpressing cells was more invasive. IGFBP2 was also associated with shortened survival in GBMs (Sallinen et al., 2000, and our unpublished data). These studies led us to hypothesize that IGFBP2 is a key molecule that contributes to invasion of GBM cells and short survival of GMB patients. We also hypothesize that IGFBP2 overexpression plays a positive role in glioma progression to GBMs. In this project, we propose to test the hypotheses using the RCAS/tv-a mouse model. We further hypothesize that GBMs developed in mice are homologous to human GBMs and this can be evaluated by comparative gene expression profiling as well as focused set of genes. The three specific aims are: (1) To test whether IGFBP2 enhances GBM development shortens survival of GBM mice, and increases GBM cell invasion in vivo. (2) To examine whether IGFBP2 plays an important role in progression from low-grade glioma to GBM. (3) To characterize mouse gliomas using gene expression profiling. Through this program, we want to enrich our knowledge of IGFBP2 pathways that are likely important for glioma development and progression; we also want to use genomics approach to evaluate whether mouse gliomas faithfully represent human gliomas at gene expression levels.