Predicting Rates and Regioselectivity in Cytochrome P450 Mediated Reactions

摘要和摘要-本提案的主要目标是1)提供了解细胞色素P450介导的反应机制所需的基本知识，包括速率，区域选择性和结合，以及2)提供预测ADMET代谢成分（吸收，分布，代谢，排泄，毒性）的计算工具。这些都是非常重要的目标，将对几乎所有的药物开发项目产生积极影响，并减少开发新疗法所需的时间。这些目标将根据以下四个具体目标进行描述：目标1)目前，细胞色素P450使用的活性氧物种数量仍然存在争议，我们认为这可能解释了为什么代谢预测方法的准确度限制在85%左右。代谢的预测模型在药物设计中很重要，并且有可能通过减少开发新药所需的时间来挽救生命。特异性Aim 2探讨了与血红素铁协调的底物的结合（II型结合），通过将亲和力提高高达250倍，为P450酶提供特异性。这是很重要的，因为系统给药的目的是抑制单一的P450酶通常会导致广泛的抑制许多P450酶，扰乱体内平衡，并引起药物-药物相互作用。目的3)P450介导率的计算预测仍然是该领域研究人员最重要的目标之一。具体目标3将确定这些预测的重要特征。如果这个目标能够实现，我们就可以理解预测药物从体内清除的重要特征，我们就更接近虚拟药物设计的目标。随着预测ADMET计算方法和研究出版物数量的不断增加，Specific Aim 4将提供公共可访问的测试集数据库以及开放源代码预测代码，允许根据通用基准测试不同的计算ADMET方法。这一点尤其重要，因为大多数方法都是内部开发的，测试集没有发布。我们假设，通过将开放源代码发布到一个用于代谢预测的公共网站，我们可以将我们和其他人的研究结果转化为公共领域，从而显著提高这些工具的使用和质量。我们还假设，所有研究人员都可以获得一组通用的数据，这将鼓励ADMET模型的验证、比较和增强。

DESCRIPTION (provided by Applicant): Summary and Abstract- The main goals of this proposal are 1) to provide the fundamental knowledge required for understanding Cytochrome P450 mediated reaction mechanism with regards to rates, regioselectivity, and binding, and 2) to provide computational tools for predicting the metabolic component of ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity). These are highly significant goals that will positively impact almost all drug development projects and decrease the time required to develop new therapeutics. These goals will be described in terms of the following four Specific Aims: Aim 1) At present the number of active-oxygen species used by Cytochrome P450 remains controversial, and we propose that this may explain why predictive methods for metabolism are limited to around 85% accuracy. Predictive models for metabolism are important in drug design, and have the potential to save lives by decreasing the time it takes to develop new drugs. Aim 2) Specific Aim 2 explores binding afforded to substrates designed to coordinate to the iron of the heme (type II binding) providing specificity for P450 enzymes by increasing affinity up to 250-fold. This is significant because systemic administration of drugs meant to inhibit a single P450 enzyme normally leads to broad inhibition of a number of P450 enzymes, upsetting homeostasis, and causing drug-drug interactions. Aim 3) Computational prediction of P450 mediated rates remains one of the most important targets of researchers working in the field. Specific Aim 3 will establish the features important in such predictions. If this goal can be met we can understand the important features involved in predicting the clearance of a drug from the body, and we move closer to the goal of virtual drug design. Aim 4) With the ever-expanding number of computational methods and research publications in predictive ADMET, Specific Aim 4 will provide common sets of publicly accessible test set databases along with open-source predictive code allowing different computational ADMET methods to be tested against common benchmarks. This is particularly important since the majority of the methods are developed in-house and the test sets are not published. We hypothesize that by publishing open source code to a public web site for metabolic predictions that we can translate the results of ours' and others' research into the public domain resulting in a significant increase in the use and quality of these tools. We also hypothesize that having a common set of data available to all researchers will encourage validation, comparison, and enhancement of ADMET models. Public Health Relevance: Narrative- The purpose of this grant application is to understand the important features of cytochrome P450 enzymes with respect to drug metabolism, and drug design. Cytochrome P450 enzymes are the most important drug metabolizing enzymes and are responsible for most drug metabolism. While the majority of the reactions mediated by this enzymes family are benign, a number cause activation to reactive species that can cause cancer and toxicity. Furthermore, many life threatening drug-drug interactions occur from drugs slowing cytochrome P450 mediated reactions. This grant application will develop methods to design new drugs faster, and safer than we can presently through an increased understanding of the rates and binding affinities or P450 mediated reactions.