Synthetic Protein Families by Structure-Guided SCHEMA Recombination

描述(由申请人提供)：在努力理解基本的序列-结构-功能关系方面，合成蛋白质家族有力地补充了自然遗传多样性。我们建议使用一个独特的嵌合细胞色素P450家族，通过结构导向重组，在模拟P450的结构、稳定性和特异性方面取得进展。这些不同的序列有许多不同的氨基酸，但仍有可能折叠成高度相似的结构，这些不同的序列具有不同的稳定性，并显示出一系列的活性和特异性。我们寻求在给定一组相对容易在这些实验室生成的序列上执行的实验测量的情况下，最大化准确的结构、稳定性和特异性预测的数量。我们建议使用这个独特的P450嵌合体文库来研究酶的结构模块化，并利用这种模块化来简化结构-功能预测。我们将确定几个嵌合的P450血红素结构域的X射线晶体结构。晶体结构将有助于我们同时努力通过计算预测准确的嵌合体结构。我们将测试重组优化的酶亚结构的新算法，以采样相关的蛋白质骨架构象。我们的目标是弥合低分辨率同源模型和精确的高分辨率结构之间的差距。嵌合酶也是研究序列-结构-稳定性关系的丰富资源。嵌合蛋白具有不同的稳定性；许多比它们的双亲更稳定。我们计划测量选定嵌合体的稳定性，并构建和测试蛋白质稳定性的可加性模型，并量化非可加性耦合效应。重组在亲本蛋白质之间创建序列中间体，并允许我们协调序列块交换的影响和单个突变的作用。这项工作将有助于提供关键数据和洞察远程相互作用和耦合的影响。最后，我们计划探索嵌合细胞色素P450的功能多样性，结合它们在药物发现和药物代谢中的应用。这些可溶的、表达良好的嵌合P450是生产真实的人类药物代谢物和实现铅多样化的可行平台。初步结果表明，不同的序列在广泛的底物上显示出活性。我们将对一组嵌合P450针对药物相关支架的活性进行系统分析，使用聚类和回归来确定是否可以从选定嵌合体的数据中预测活性和特异性。我们将尝试在结构P450-底物复合体模型方面使P450特异性结果合理化。这项研究继续进行有效的跨学科合作，配合强大的实验和理论能力。公共卫生相关性：我们建议演示如何使用实验室产生的细胞色素P450家族来研究P450的结构、稳定性和功能的基础。我们将测试一种预测药物和P450酶之间相互作用的新策略，这是一项潜在的突破，将使我们能够预测药物代谢和相互作用，并有效地生产用于毒性研究和药物发现的药物代谢物。最后，这项研究代表了一种利用重组来研究蛋白质稳定性的全新方法，其结果可应用于对酶的一般理解。

DESCRIPTION (provided by applicant): Synthetic protein families powerfully complement natural genetic diversity in efforts to understand fundamental sequence-structure-function relationships. We propose to use a unique family of chimeric cytochrome P450s made by structure-guided recombination to make advances in modeling P450 structure, stability and specificity. Differing from one another by many dozens of amino acids yet still likely to fold into highly similar structures, these diverse sequences have different stabilities and exhibit an array of activities and specificities. We seek to maximize the number of accurate structure, stability, and specificity predictions given a set number of experimental measurements that are relatively easily performed on these laboratory-generated sequences. We propose to use this unique library of P450 chimeras to study enzyme structural modularity and exploit that modularity to simplify structure-function predictions. We will determine the X-ray crystal structures for several chimeric P450 heme domains. The crystal structures will assist our concurrent efforts to computationally predict accurate chimera structures. We will test new algorithms that recombine optimized enzyme substructures to sample relevant protein backbone conformations. Our goal is to bridge the gap between low resolution homology models and accurate high resolution structures. The chimeric enzymes are also a rich resource for investigating sequence-structure-stability relationships. The chimeric proteins have diverse stabilities; many are more stable than their parents. We plan to measure stability for selected chimeras and construct and test additive models for protein stability and quantify non- additive coupling effects. Recombination creates sequence intermediates between parent proteins and allows us to reconcile the effects of swapping blocks of sequence with the action of individual mutations. This work will contribute key data and insight into the effects of long-range interactions and coupling. Finally, we plan to explore the functional diversity of the chimeric cytochrome P450s, in the context of their applications in drug discovery and drug metabolism. These soluble, well-expressed chimeric P450s are a viable platform for production of authentic human metabolites of drugs and for lead diversification. Preliminary results indicate that the diverse sequences exhibit activity on a wide range of substrates. We will perform a systematic analysis of the activities of a set of chimeric P450s towards pharmaceutically relevant scaffolds, using clustering and regression to determine whether activity and specificity can be predicted from data on selected chimeras. We will attempt to rationalize the P450 specificity results in terms of structural P450- substrate complex models. This research continues an effective interdisciplinary collaboration that partners powerful experimental and theoretical capabilities. PUBLIC HEALTH RELEVANCE: We propose to demonstrate how a laboratory-generated family of cytochrome P450s can be used to study the basis of P450 structure, stability, and function. We will test a new strategy for predicting the interactions between drugs and P450 enzymes, a potential breakthrough that would allow us to predict drug metabolism and interactions and to efficiently produce drug metabolites for toxicity studies and drug discovery. Finally, this research represents a fundamentally new approach to study protein stability using recombination, with results that can be applied to understanding enzymes in general.