Identification of protein-metabolite interactome.

描述（由申请人提供）：本提案的长期目标是开发有效的综合计算和实验方法，以在蛋白质组学规模上识别蛋白质和主要代谢物之间的相互作用。所提出的方法允许在蛋白质组学尺度上快速鉴定配体结合蛋白并构建蛋白质-代谢物相互作用网络，这将为理解生物系统提供重要信息。作为一个模型案例，我们将在大肠杆菌蛋白质组中鉴定NAD+和NADP+结合蛋白。特别提出了以下具体目标：(1)开发一套预测与代谢物结合的蛋白质的计算方法。我们采用了基于序列和基于结构的方法。我们将开发和采用的基于序列的方法包括我们的新功能预测方法，PFP及其变体，这些方法比传统方法具有更高的灵敏度和更高的功能分配覆盖率。基于结构的方法包括快速局部蛋白质表面形状比较方法，直接比较局部表面区域的形状和物理化学性质。(2)应用基于能量学的靶标鉴定方法高效筛选与代谢物结合的蛋白。与NAD+和NADP+结合后稳定的蛋白质将在大肠杆菌裂解物中通过结合蛋白酶和定量质谱法进行短暂孵育。通过计算或实验方法鉴定的蛋白质将通过互补方法交叉验证。该项目的成功完成将为系统鉴定与特定代谢物结合的蛋白质建立方法，从而使我们能够提供细胞中蛋白质和代谢物的相互作用网络。待开发的方法和由此产生的相互作用网络将有助于药物发现的早期阶段，因此拟议的项目可能具有重要的治疗效用。

DESCRIPTION (provided by applicant): The long-term goal of this proposal is to develop effective integrated computational and experimental approaches to identify interactions between proteins and major metabolites on a proteomics scale. The proposed approach allows rapid identification of ligand binding proteins on a proteomics scale and constructing protein-metabolite interaction network, which will provide crucial information for understanding biological systems. As a model case, we will identify NAD+ and NADP+ binding proteins in the E. coli proteome. In particular, the following specific aims are proposed: (1) To develop a set of computational methods for predicting proteins that bind to metabolites. We employ both sequence- based and structure-based methods. The sequence-based methods we will develop and employ include our novel function prediction methods, PFP and its variant, which are shown to have higher sensitivity and higher function assignment coverage than conventional methods. Structure-based methods include fast local protein surface shape comparison method, which directly compare shape and physicochemical property of local surface regions. (2) To apply energetics-based target identification approach to efficiently screen proteins that bind to metabolites. Proteins stabilized upon binding to NAD+ and NADP+ will be identified in a E. coli lysate by combining a brief incubation with a protease and quantitative mass spectrometry. Proteins identified by either computational or experimental methods will be cross-validated by the complementary approaches. Successful completion of this project will establish methodology for systematic identification of proteins that bind to specific metabolites and thus will enable us to provide interaction network of proteins and metabolites in cells. The methodology to be developed and the resulting interaction network will assist in the early stages of drug discovery, and hence the proposed project could have significant therapeutic utility. PUBLIC HEALTH RELEVANCE: The goal of the project is the development of intergrated computational and experimental approach to determine interactions between proteins and major matabolites in cells on a systems level. The proposed approach will allow rapid identification of ligand binding proteins on a proteomics scale and will enable the construction of protein-metaboolite interaction networks. The outcome of this project will provide information crucial in understanding biological systems and useful in hypothesis generation in the early stages of drug discovery.