Computer-assisted identification and design of subtype selective GPCR antagonist

描述（由申请人提供）：cns相关G偶联蛋白受体（GPCR）的高选择性激动剂和拮抗剂的鉴定和设计对于开发已知GPCR靶点的新药以及对各种精神障碍的新靶点的临床验证至关重要。我们项目的长期目标是利用最近在GPCR晶体结构测定方面的突破，并使用先进的分子建模来创建和验证一个高效的硅GPCR配体亲和/选择性分析面板（GLASP面板），用于计算机辅助合理设计亚型选择性候选药物。在第一阶段的第一年，我们计划开发和验证GLASP面板，用于cns相关GPCR靶点的胺结合家族，其结构和功能与最近解决的b2-肾上腺素能（b2AR）高分辨率3D模板最接近。考虑到受体的构象灵活性和配体诱导的可塑性，我们将建立肾上腺素能、血清素和多巴胺家族中所有GPCR亚型的详细全原子模型。单个模型和整个亲和/选择性面板将被优化，并通过一套全面的公开可用的配体亲和数据进行验证。此外，我们计划通过盲预测对一组选定配体的亲和力来严格验证GLASP面板，然后由第三方进行生化测试和评估面板的准确性。在第二年，我们计划在几个合理的药物设计应用中使用经过验证的硅分析面板。因此，我们将使用这些工具对制药行业合作伙伴选择的目标进行计算机辅助先导物优化。此外，简化版本的全原子模型将准备和验证快速虚拟配体筛选（VLS）的应用。VLS模型将用于搜索可用和虚拟化合物的大型文库，并在生化分析中筛选选定的重点文库，以确定新的候选先导化合物。经过验证的GLASP面板和快速VLS模型将为具有成本效益的新型高亲和力支架和亚型选择性激动剂和拮抗剂的合理设计提供强大的工具，用于临床重要的GPCR靶点。在一期项目取得成功后，我们设想将这种方法进一步扩展到其他与中枢神经系统相关的A类gpcr，并应用于制药行业和学术界的一些合作项目。公共卫生相关性：大约一半的精神障碍和神经退行性疾病的药物靶向g偶联蛋白受体（gpcr），但许多这些药物通过与中枢神经系统（CNS）或心血管系统中的其他gpcr“脱靶”活动而产生严重的副作用。现有治疗方法的改进和CNS靶点新治疗方法的开发依赖于作用于特定GPCR亚型的高选择性分子的设计。最近在GPCR原子结构测定和改进的分子建模工具方面取得的突破性进展使创建和验证新的候选药物的计算机辅助合理设计的有效应用成为可能，这是我们项目的主要目标。

DESCRIPTION (provided by applicant): The identification and design of highly selective agonists and antagonists for CNS-related G coupled protein receptors (GPCRs) is of paramount importance for the development of new drugs to well-known GPCR targets and for the clinical validation of new targets for a wide range of mental disorders. The long term goal of our project is to capitalize on the recent breakthroughs in GPCR crystal structure determination and to use advanced molecular modeling to create and validate an efficient in silico GPCR ligand affinity/selectivity profiling panel (GLASP panel) for applications to the computer assisted rational design of subtype selective drug candidates. In year one of Phase I, we plan to develop and validate the GLASP panel for amine binding families of CNS-related GPCR targets with the closest structural and functional homology to the recently solved b2-adrenergic (b2AR) high resolution 3D template. We will build detailed all-atom models of all GPCR subtypes in adrenergic, serotonin and dopamine families, taking into account conformational flexibility and ligand-induced plasticity of receptors. The individual models and the whole affinity/selectivity panel will be optimized and verified with a comprehensive set of publicly available ligand affinity data. In addition, we plan to strictly validate the GLASP panel though blind prediction of affinity for a selected set of ligands, followed by biochemical testing and assessment of the panel accuracy by a third party. In year two, we plan to employ the validated in silico profiling panel in several rational drug design applications. Thus, we will use these tools to perform computer assisted lead optimization for targets selected by our partners in pharmaceutical industry. In addition, simplified versions of the all-atom models will be prepared and validated for fast virtual ligand screening (VLS) applications. The VLS models will be used to search large libraries of available and virtual compounds, and a selected focused library will be screened in biochemical assays to identify new lead candidates. The validated GLASP panel and fast VLS models will provide powerful tools for the cost-efficient rational design of novel high-affinity scaffolds and subtype selective agonists and antagonists for clinically important GPCR targets. Upon success of the Phase I project, we envision further extension of this approach to other CNS-related Class A GPCRs, and application to a number of collaborative programs in pharmaceutical industry and academia. PUBLIC HEALTH RELEVANCE: About half of the medications for mental disorders and neurodegenerative diseases target G-coupled Protein Receptors (GPCRs), but many of these drugs have serious side effects through "off target" activity with other GPCRs in the central nervous system (CNS) or the cardiovascular system. Improvement of existing treatments and development of new treatments for CNS targets depend on the design of highly selective molecules acting on one specific GPCR subtype. Recent breakthrough developments in GPCR atomic structure determination and improved molecular modeling tools make it possible to create and validate efficient applications for computer assisted rational design of new candidate drugs, which is the prime goal of our project.