A minimally invasive synthetic bio-driven approach for natural products discovery

 描述(由申请人提供)：本提案中概述的工作旨在建立一个新的交钥匙天然产品发现平台--从开发用于捕获和鉴定新型生物合成多样性的模式菌株和简化方法，到天然产品本身的表达和表征。我们的努力最终将不仅是开发一条新的天然产物发现管道，而且还将开发一系列工具/资源(优化的菌株、改进的集群克隆和注释方法、改进的天然产物产生门菌株的基因/基因组编辑方法等)。旨在消除基因组DNA转化为天然产品的现有障碍。我们的方法致力于开发工具，将从DNA序列生产分子所需的操作降至最低，从而最大限度地提高真正高通量小分子发现管道的潜力。在我们的第一个目标中，我们将通过经验性地选择细菌来开发传统上与自然产品发现(放线菌/γ-蛋白细菌)相关的来自Phyla的改进的模型宿主 对天然产品生产表现出最大先天倾向的菌株。然后，利用现代合成生物学/基因组编辑工具和高通量经验簇表达筛选，我们将进一步优化这些菌株，以提高它们转录合成的能力，并支持从不同的生物合成基因簇集合中生产天然产物。第二个目标是建立使用现代基因组/基因簇编辑工具(CRISPR，MAGE)在模式天然产物生产属中的协议。在我们的第三个目标中，我们建议开发一种生物信息学管道来识别基因簇家族，通过 基因内容和/或基因组织并不类似于任何已知的基因簇家族(即隐蔽基因簇家族，CCF)。第三个目标还建立了改进的集群克隆和注释方法，使微生物生物合成多样性(无论来自培养或未培养的微生物)能够以一种单一的格式快速存档，这种格式可以很容易地为CCF添加注释和挖掘，只需最少的测序工作。在我们的第四个目标中，我们建议利用目标1-3中开发的新工具来建立更有效地从DNA序列中获取天然产品的方法。这一目标包括建立一个改进的功能(元)基因组学平台以及两个可扩展的基于CRISPR的、基因簇不可知的激活策略，这两种策略利用CRISPR系统的体内或体外能力来同时针对一个簇中的多个启动子区域进行转录激活。这一目标还包括开发半自动的、基于MS/核磁共振的方法，对小规模发酵的发酵液提取物中存在的异源天然产物进行结构评估。我们的发现策略提供了更具侵略性的簇诱导方法，我们相信，作为一个整体，我们代表了一种从基因组序列中获取天然产物的稳健、全面的方法。

 DESCRIPTION (provided by applicant): The work outlined in this proposal is designed to establish a new turnkey natural product discovery platform -- from the development of model strains and streamlined methods for the capture and bioinformatics identification of novel biosynthetic diversity, through to the expression and characterization of natural products themselves. Our efforts will culminate in the development of not only a new natural product discovery pipeline but also a series of tools/resources (optimized strains, improved cluster cloning and annotation methods, improved gene/genome editing methods for strains from natural product producing phyla, etc.) that are designed to remove existing roadblocks in the conversion of genomic DNA into natural products. Our approach strives to develop tools that will minimize the manipulations required for molecule production from DNA sequence and, thereby, maximize the potential for truly high throughput small molecule discovery pipelines. In our First Aim we will develop improved model hosts from Phyla that are traditionally associated with natural product discovery (actinomycetes, ß/γ-proteobacteria), by empirically selecting bacterial strains that show the greatest innate tendency for natural product production. Then, using modern synthetic biology/genome editing tools and high throughput empirical cluster expression screening, we will further optimize these strains to improve their ability to transcriptionally indce and support the production of natural products from diverse collections of biosynthetic gene clusters. The Second Aim focuses on establishing protocols for using modern genome/gene cluster editing tools (CRISPR, MAGE) in model natural product producing genera. In our Third Aim, we propose to develop a bioinformatics pipeline for identifying gene cluster families that, by gene content and/or gene organization, do not resemble any known gene cluster family (i.e. cryptic gene cluster families, CCF). The Third Aim also establishes improved cluster cloning and annotation methods that will allow for the rapid archiving of microbial biosynthetic diversity (whether from cultured or uncultured microbes) in a single format that can be easily annotated and mined for CCFs using minimal sequencing effort. In our Fourth Aim we propose to leverage the new tools developed in Aims 1-3 to establish methods for more efficiently accessing natural products from DNA sequence. This Aim includes the establishment of an improved functional (meta)genomics platform as well as two scalable CRISPR-based, gene cluster agnostic, activation strategies that take advantage of either the in vivo or in vitro ability of the CRISPR system to simultaneously target multiple promoter regions within a cluster for transcriptional activation. This aim also includes the development of methods for semi-automated, MS/NMR-based structural evaluation of heterologously produced natural products present in broth extracts from small-scale fermentations. Our discovery strategy provides successively more aggressive cluster induction methods and we believe, as a whole, represents a robust, comprehensive approach for accessing natural products from genomic sequence.