Time-resolved FRET-based allostery sensors for any protein kinase drug target

摘要 蛋白激酶是开发新的癌症治疗药物的顶级药物靶标。现有 靶向激酶的高度保守活性位点的激酶抑制剂具有主要的局限性， 选择性和导致治疗失败的临床耐药的高发生率。新的变构抑制剂， 结合在激酶活性位点外的其他口袋中，并引发阻断激酶活性的结构变化， 选择性更强，并且在克服对常规激酶抑制剂的临床耐药性方面非常有效。 然而，变构激酶抑制剂已被证明非常具有挑战性，以确定与现有的药物筛选 技术，并且仅可用于少数激酶。 现有药物筛选技术失败的一个主要原因是它们不能检测原子- 定义变构激酶抑制剂作用模式的尺度结构变化。我们已经开发出一种 改变游戏规则的高通量筛选技术，基于纳秒时间分辨荧光， 可以通过原子分辨率跟踪它们在激酶中触发的结构变化来识别变构抑制剂 药物靶点将这项技术应用于有丝分裂蛋白激酶Aurora A，我们已经证明它可以 同时跟踪抑制剂结合亲和力和对激酶的变构效应，可以将抑制剂分为 不同的变构亚型，并且足够准确、快速和可扩展以处理高通量筛选 项目为了最大限度地发挥该技术对药物发现管道的影响， 需要克服，以扩大技术的范围，超越目前单一的药物靶标Aurora A。 我们目前的技术是基于一种化学标记程序，用于掺入荧光探针， 半胱氨酸标记，由于存在，它不容易适用于许多重要的激酶药物靶点 半胱氨酸残基对于结构完整性和催化功能很重要。在本提案中，我们扩大了范围 通过开发一系列新的激酶， 用于位点特异性探针掺入的工具，以及通过扩大可以被 在筛选中识别。最后，我们对该技术在现实世界药物发现工作中的适用性进行了基准测试 通过进行高通量筛选项目以鉴定至少一种蛋白质的新型变构抑制剂， 目前还没有变构抑制剂。 该项目的成功将带来一种全新的变构药物发现技术， 现有技术无法提供的独特功能。采用这种方法可以启动 发现了大量重要的癌症药物靶点的变构激酶抑制剂，拓宽了 为癌症患者提供治疗选择，并提供急需的新方法， 一线激酶抑制剂治疗的临床耐药率。

ABSTRACT The protein kinases are the top class of drug targets for the development of new cancer therapeutics. Existing kinase inhibitors, which target the highly-conserved active sites of kinases, have major limitations including poor selectivity and a high incidence of clinical resistance leading to treatment failure. New allosteric inhibitors, which bind in other pockets outside the kinase active site and trigger structural changes that block kinase activity, are far more selective and are highly effective at overriding clinical resistance to conventional kinase inhibitors. However, allosteric kinase inhibitors have proven extremely challenging to identify with existing drug screening technologies, and are only available for a small handful of kinases. A major reason for this failure of existing drug screening technologies is that they cannot detect the atomic- scale structural changes that define the mode of action of allosteric kinase inhibitors. We have developed a game-changing high-throughput screening technology, based on nanosecond time-resolved fluorescence, that can identify allosteric inhibitors by tracking with atomic resolution the structural changes they trigger in the kinase drug target. Applying this technology to the mitotic protein kinase Aurora A, we have shown that it can simultaneously track inhibitor binding affinity and allosteric effects on the kinase, can classify inhibitors into different allosteric subtypes, and is sufficiently accurate, rapid and scalable to handle high-throughput screening projects. To maximize the impact of the technology on the drug discovery pipeline, several technical barriers need to be surmounted to expand the scope of the technology beyond the current single drug target Aurora A. Our current technology is based on a chemical labeling procedure for incorporating fluorescent probes, cysteine labeling, that is not readily applicable to many important kinase drug targets due to the presence of cysteine residues important for structural integrity and catalytic function. In this proposal, we broaden the scope of the technology to make it applicable to the majority of the ~500 human kinases by developing a series of new tools for site-specific probe incorporation and by expanding the range and type of small molecules that can be identified in screening. Finally, we benchmark the suitability of the technology for real-world drug discovery efforts by performing a high-throughput screening project to identify novel allosteric inhibitors of at least one protein kinase for which no allosteric inhibitors are currently available. The success of this project will bring an entirely-new allosteric drug discovery technology into being, with unique capabilities that no existing technology can provide. Employment of this approach could jumpstart the discovery of allosteric kinase inhibitors for a large number of important cancer drug targets, broadening the range of therapeutic options for cancer patients and providing a much-needed new approach for combating the high prevalence of clinical resistance to first-line kinase inhibitor therapies.