Accelerating drug development and repurposing using systematic genetic interactio

描述（由申请人提供）：药物开发的主要限制是确定新药物的细胞靶点（和非靶点），以及发现有效的药物组合来对抗快速发展的疾病，如癌症。此外，药物筛选需要大量处理，这限制了吞吐量并增加了可变性。最后，遗传背景往往决定药物敏感性，使最佳治疗方案的选择复杂化。该提案描述了我的实验室开始解决这些问题的广泛努力：(1)使用生物反应器自动化细胞培养，显著提高全基因组筛选的速度、准确性和可扩展性，(2)开发工具来高通量测量药物靶点之间的成对遗传相互作用，(3)使用这些工具来识别特定于应激和癌基因状态的合成致死组合靶点。我们在初步研究中表明，我们的新型高复杂性shRNA文库（25个shRNA /基因）可用于识别具有显著特异性的抗癌药物靶标（Matheny等人，2013）。我们还开发了一种可扩展的、快速的策略来创建双shrna文库，同时测量100,000个基因对之间的遗传相互作用（Bassik等人，2013）。我们将利用这个平台创建新的、高复杂性的基因调节文库，由shRNA和CRISPR元件组成，靶向fda批准的药物靶标对，然后在一组应激和癌基因的背景下鉴定这些基因之间的合成致死相互作用。阐明这些协同作用将使我们了解应激信号传导和细胞死亡调控的潜在生物学，并阐明利用现有药物对这些应激普遍存在的疾病（如癌症）进行新的治疗组合。同时，基因相互作用图将允许我们直接调查

DESCRIPTION (provided by applicant): Major limitations in drug development are identification of cellular targets (and off-targets) for novel drugs, and discovery of effective drg combinations to combat rapidly evolving diseases such as cancer. Furthermore, drug screens require extensive handling, which limits throughput and increases variability. Finally, genetic background often determines drug susceptibility, complicating selection of optimal treatment regimens. This proposal describes a broad effort in my lab to begin to address these issues by (1) dramatically enhancing the speed, accuracy, and scalability of pooled genome-wide screens using bioreactors to automate cell culture, (2) developing tools to measure pairwise genetic interactions between drug targets in high-throughput, and (3) using these tools to identify synthetic lethal combination targets that are specific to stress and oncogene states. We have shown in pilot studies that our novel high-complexity shRNA libraries (25 shRNAs/gene) can be used to identify the target of a cancer-killing drug with remarkable specificity (Matheny et al., 2013). We have also developed a scalable, rapid strategy to create double-shRNA libraries to simultaneously measure genetic interactions between 100,000's of gene pairs (Bassik et al., 2013). We will adapt this platform to create novel, high-complexity gene modulation libraries composed of shRNA and CRISPR elements targeting pairs of FDA-approved drug targets, and then identify synthetic lethal interactions between these genes in the context of a panel of stresses and oncogenes. Elucidating these synergies will inform our understanding of the underlying biology of stress signaling and cell death regulation, and illuminate new therapeutic combinations using available drugs in diseases such as cancer where these stresses are prevalent. At the same time, the genetic interaction maps will allow us to directly investigate off target effects of drugs, which would streamline the identification of viable candidate molecules for further development. Together, we expect the tools developed here will be broadly useful for functional genomics screening and drug development efforts, and that this work will establish a new paradigm for empirical testing and identification of drug response biomarkers.