(PQ4) Novel tools for in vivo study of genetic interactions in cancer progression

项目总结： 人类癌症的进化是一个由多个分子和细胞事件驱动的复杂过程。癌细胞通常 拥有许多可以以相加、平行、拮抗、上位或协同方式起作用的像差。那些 相互作用有助于肿瘤的发生、发展、转移、耐药或其他危及生命的特征。而当 这些相互作用可以从肿瘤序列数据的分析中微弱地推断出来，阐明了体内的遗传相互作用 对于快速构建癌症发展的强大地图和加快治疗发展至关重要。然而， 目前，几乎没有有效的工具来精确地在体内对癌症进行多基因操作，这限制了我们对 准确地剖析这些相互作用。我们努力利用单效应子RNA引导的内切酶(RGNs)来 人类癌症的基因组编辑、并行筛查和体内建模。最近，我们创建了一个平台来 直接在活体内系统询问数百个基因座。为了克服目前多基因编辑和 为了更准确地控制多等位基因肿瘤模型的同时性和顺序性，我们使用了Cpf1和 RGN可以简单地使用独立于trrRNA的crRNA编辑其目标，从而允许同时编辑多个 基因与单一的crRNA阵列。我们开发了一个基于Cpf1的crRNA阵列筛选(CCAS)系统 并将其应用于小鼠的进展和转移模型。在我们的第一个目标中，我们将执行 CCAS用于癌症共同驱动因素体内双基因敲除表型的验证和优化。我们将建立其 同时编辑的技术严谨性、效率和专用性，以及开发一组计算管道 用于准确调用统计上有意义的基因对。我们将应用这种方法来研究基因的相互作用 在耶鲁癌症中心和医院的肺癌患者中发现肿瘤抑制因子，并确定潜在的共同驱动因素 转移到重要器官。在第二个目标中，我们将对Cpf1-Flip系统进行验证和优化 癌症靶点的顺序突变。我们将通过检测临床相关基因来证明其更广泛的适用性。 从转移基因组学公开研究中确定的集合以及大型多样本转移数据集 收集了耶鲁大学癌症患者的信息。然后，我们将应用这一方法作为无偏见的耗尽筛查，以确定目标 在特定致癌背景下生存所必需的基因。我们将开发新的多功能转基因小鼠品系 以及用于直接模拟小鼠多基因肿瘤发生的伴生病毒载体。我们将结合这些工具来 直接在本地器官的健康细胞中进行高通量遗传相互作用筛查，以确定病因 驱动肿瘤发生的突变对。我们预计，开发和建立这些工具将使 人类癌症的多基因肿瘤模型和临床前研究，直接解决了NCI挑衅性的问题4。 这些强大的工具包将使科学家能够同时或顺序地瞄准任何基因对或组合， 评估它们在肿瘤进展、转移、合成致命性、药物等方面的体内相互作用的表型结果 癌症进化中的敏感性或其他过程。

PROJECT SUMMARY: The evolution of human cancer is a complex process driven by multiple molecular and cellular events. Cancer cells often harbor numerous aberrations that can act in additive, parallel, antagonistic, epistatic or synergistic fashion. Those interactions contribute to tumorigenesis, progression, metastasis, drug resistance or other life-threatening features. While these interactions can be weakly inferred from analysis of tumor sequence data, elucidating genetic interactions in vivo is essential for rapidly building a robust map of cancer development and to accelerate therapeutic developments. However, there are currently few effective tools for precise multigenic manipulation of cancer in vivo, limiting our scope for accurately dissecting these interactions. We endeavored to harness single-effector RNA-guided endonucleases (RGNs) for genome editing, parallel screening and in vivo modeling of human cancer. Recently, we generated a platform to systematically interrogate several hundred loci directly in vivo. To overcome current limitations in multigene editing and achieve more accurate control of simultaneity and sequentiality of multi-allelic tumor modeling, we utilized Cpf1, an RGN that can edit its target simply with crRNAs independent of tracrRNA thus allowing simultaneous editing of multiple genes with a single crRNA array. We developed a preliminary Cpf1-based crRNA array screening (CCAS) system in mammalian cells, and applied it in mouse models of progression and metastasis. In our first aim, we will perform validation and optimization of CCAS for in vivo double-knockout phenotyping of cancer co-drivers. We will establish its technical rigor, efficiency and specificity for simultaneous editing, as well as developing a set of computational pipelines for accurate calling of statistically significant gene pairs. We will apply this approach to study the genetic interactions of tumor suppressors found in lung cancer patients at Yale Cancer Center and Hospital, and identify potential co-drivers of metastasis to vital organs. In the second aim, we will carry out validation and optimization of a Cpf1-Flip system for sequential mutagenesis of cancer targets. We will demonstrate its broader applicability by testing clinically relevant gene sets identified from public studies of the genomics of metastasis as well as a large multi-sample metastasis dataset gathered on Yale cancer patients. We will then apply this methodology as an unbiased depletion screen to identify targets that are essential for survival in specific oncogenic backgrounds. We will develop novel versatile transgenic mouse strains and companion viral vectors for direct modeling of multigenic tumorigenesis in mice. We will combine these tools to enable high-throughput genetic interaction screening in healthy cells directly in the native organ to identify causative mutation pairs that drive tumorigenesis. We anticipate that developing and establishing these tools will transform multigenic tumor modeling and pre-clinical studies of human cancer, directly addressing NCI Provocative Question 4. These powerful toolkits will enable scientists to target any gene pairs or combinations simultaneously or sequentially, assessing the phenotypic outcome of their in vivo interactions in tumor progression, metastasis, synthetic lethality, drug sensitivity or other processes in cancer evolution.