Predictive Models for Small-Molecule Accumulation in Gram-Negative Bacteria

革兰氏阴性细菌中小分子积累的预测模型

• 批准号: 9982190
• 负责人: DEREK S TAN
• 金额: $ 123.93万
• 依托单位: SLOAN-KETTERING INST CAN RESEARCH
• 依托单位国家: 美国
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-08-10 至 2023-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/9982190
• 关键词: Acinetobacter baumannii; Address; Algorithmic Software; Anti-Bacterial Agents; Antibiotic Resistance; Antibiotics; Architecture; Bacteria; Biochemical; Biochemistry; Biological Assay; Biological Availability; Cells; Chemicals; Communities; Data; Data Set; Detection; Development; Effectiveness; Escherichia coli; Gram-Negative Bacteria; Gram-Negative Bacterial Infections; Human; Incidence; Individual; Infection; Interdisciplinary Study; Kinetics; Knock-out; Label; Lead; Libraries; Machine Learning; Mammalian Cell; Mass Spectrum Analysis; Measurement; Measures; Membrane; Microbiology; Modeling; Oral; Partner in relationship; Penetration; Pharmaceutical Chemistry; Pharmaceutical Preparations; Pharmacology; Property; Pseudomonas aeruginosa; Public Health; Quantitative Evaluations; Quantitative Structure-Activity Relationship; Role; Structure; Testing; Variant; analog; base; biophysical model; cell envelope; cheminformatics; combat; computerized tools; density; design; drug discovery; efflux pump; high throughput screening; improved; inhibitor/antagonist; interdisciplinary approach; kinetic model; lead optimization; learning network; multidisciplinary; neural network; novel; predictive modeling; programs; prospective; public health relevance; screening; small molecule; small molecule libraries; success; tool

PROJECT SUMMARY Predictive Models for Small-Molecule Accumulation in Gram-Negative Bacteria. Antibiotic-resistant Gram-negative bacterial infections are increasing in incidence and novel antibiotics are urgently needed to combat this growing threat to public health. A major roadblock to the development of novel antibiotics is our poor understanding of the structural features of small molecules that correlate with bacterial penetration and efflux. As a result, while potent biochemical inhibitors can often be identified for new targets, developing them into compounds with whole-cell antibacterial activity has proven challenging. To address this critical problem, we propose herein a comprehensive, multidisciplinary approach to develop quantitative models to predict small-molecule penetration and efflux in Gram-negative bacteria. We have pioneered a general platform for systematic, quantitative evaluation of small-molecule accumulation in bacteria, using label-free LC-MS/MS detection and multivariate cheminformatic analysis. We have also developed unique isogenic strain sets of wild-type, hyperporinated, efflux-knockout, and doubly-compromised E. coli, P. aeruginosa, and A. baumannii that allow us to dissect the individual contributions of outer/inner membrane penetration and active efflux to net accumulation, using a kinetic model that accurately recapitulates available experimental data. Moreover, we have developed machine learning and neural network approaches to QSAR (quantitative structure–activity relationship) modeling of pharmacological properties that will now be used to develop predictive cheminformatic models for Gram-negative accumulation, penetration, and efflux. This project will be carried out by a multidisciplinary SPEAR-GN Project Team (Small-molecule Penetration & Efflux in Antibiotic-Resistant Gram-Negatives, “speargun”) involving the labs of Derek Tan (MSK, PI), Helen Zgurskaya (OU, PI), Bradley Sherborne (Merck, Lead Collaborator), Valentin Rybenkov (OU, Co-I), Adam Duerfeldt (OU, Co-I), Carl Balibar (Merck, Collaborator), and David McLaren (Merck, Collaborator), comprising extensive combined expertise in organic and diversity-oriented synthesis, biochemistry, microbiology, high- throughput screening, mass spectrometry, biophysical modeling, cheminformatics, and medicinal chemistry. Herein, we will design and synthesize chemical libraries with diverse structural and physicochemical properties; analyze their accumulation in the isogenic strain sets in both high-throughput and high-density assay formats; extract kinetic parameters for penetration and efflux from the resulting experimental datasets; develop and validate robust QSAR models for accumulation, penetration, and efflux; and demonstrate the utility of these models in medicinal chemistry campaigns to develop novel Gram-negative antibiotics against three targets. This project will provide a major advance in the field of antibacterial drug discovery, providing powerful enabling tools to the scientific community to address this major threat to public health.

项目摘要 革兰氏阴性细菌中小分子积累的预测模型。 抗生素持续的革兰氏阴性细菌感染的发生率正在增加，而新型抗生素是 迫切需要打击这种日益严重的公共卫生威胁。小说发展的主要障碍 抗生素是我们对与细菌相关的小分子的结构特征的不良理解 穿透和外排。结果，尽管通常可以针对新目标确定潜在的生化抑制剂，但 将它们发展成具有全细胞抗菌活性的化合物已被证明是挑战。 为了解决这个关键问题，我们在此提出了一种全面的多学科方法来发展 定量模型，以预测革兰氏阴性细菌中小分子渗透和外排。我们有 开创了一个通用平台，用于对细菌中小分子积累进行系统的定量评估， 使用无标签的LC-MS/MS检测和多元化学法分析。我们也发展了 野生型，高孢子，外排敲击和双重促进的大肠杆菌的独特的等源性应变集， 铜绿假单胞菌和A. baumannii，使我们能够剖析外部/内膜的个体贡献 使用动力学模型准确地概括可用的动力学模型 实验数据。此外，我们已经开发了机器学习和神经网络方法 （定量结构 - 活性关系）现已用于使用的药物的建模 开发用于革兰氏阴性积累，渗透和外排的预测性化学图案模型。 该项目将由多学科的Spear-GN项目团队（小分子渗透和 抗生素耐药的革兰氏阴性阴性中的外排，涉及Derek Tan（MSK，PI）的实验室的“ Speargun”），Helen Zgurskaya（OU，PI），Bradley Sherborne（默克公司，首席合作者），Valentin Rybenkov（OU，Co-I），Adam Duerfeldt（OU，Co-I），Carl Balibar（默克，合作者）和David McLaren（Merck，合作者），完成 有机和多样性的合成，生物化学，微生物学，高 - 吞吐量筛选，质谱，生物物理建模，化学信息学和医学化学。 本文中，我们将设计和合成具有潜水员结构和物理的化学库 特性;分析它们在高通量和高密度中的等源性应变集中的积累 测定格式；从所得的实验数据集中提取动力学参数，以渗透和排出； 开发和验证可靠的QSAR模型以积累，穿透和外排；并展示实用程序 这些模型在药物化学运动中开发了针对三种新颖的革兰氏阴性抗生素 目标。该项目将在抗菌药物发现领域提供重大进步，从而提供强大 使科学界的工具能够应对公共卫生的这一主要威胁。