CAR and Antibodies Structure-Activity Relationships and molecular architecture

CAR 和抗体构效关系和分子结构

• 批准号: 10702800
• 负责人: Philippe Youkharibache
• 金额: $ 18.92万
• 依托单位: DIVISION OF BASIC SCIENCES - NCI
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10702800
• 关键词: Antibodies; Antigen Targeting; Architecture; Big Data; Binding; Bioinformatics; Biological; CD19 gene; CD28 gene; CD8B1 gene; CTLA4 gene; Cell Surface Proteins; Cells; Chimera organism; Clinical Trials; Complement; Complex; Computational Technique; Data; Data Science; Disease; Elements; Engineering; Extracellular Domain; Goals; Immunoglobulin Domain; In Vitro; Individual; Knowledge; Lead; Ligands; Light; Machine Learning; Measurement; Mediating; Membrane; Methods; Microtomy; Mining; Modeling; Molecular; Process; Property; Proteins; Reporting; Research; Role; Route; Specificity; Structure; Structure-Activity Relationship; Supersecondary Protein Structure; System; T-Lymphocyte; Techniques; Tertiary Protein Structure; Time; Toxic effect; X-Ray Crystallography; antibody engineering; antigen binding; base; biophysical properties; chimeric antigen receptor; combinatorial; computerized tools; data repository; design; flexibility; improved; in silico; in vivo; interest; molecular assembly/self assembly; nanobodies; novel strategies; programmed cell death ligand 1; programmed cell death protein 1; protein structure; rational design; receptor; tool

An important focus of this project is to support CARs (Chimeric Antigen Receptors) and antibodies' precise bottom-up engineering by acquiring the detailed knowledge of the interplay between their constitutive molecular elements (Antibody binding moiety, Hinge, Transmembrane, and stimulatory domains) and the characterization of their functional role. State-of-the-art CARs design strategies rely on the buildup of chimeras from these modular domains used as components. This is done through a combinatorial process using an expensive trial and error approach to select the most promising candidates. My research has focused in the last few years on the study of self-association determinants of molecular systems, especially proteins, as revealed by their structural symmetries at several levels of molecular organization: molecular assemblies, protein chains, protein domains, and protein supersecondary structures (protodomains) [Youkharibache 2019,2021]. While symmetry is high and apparent in large molecular assemblies, it is difficult to detect at the level of protein domains or receptor-ligand interactions. For example, while antibodies' heavy and light chain symmetries are well known, the individual Immunoglobulin domains consist themselves of intrinsically pseudo-symmetric protodomains [Youkharibache 2021], a property largely ignored that can open new routes to antibody engineering, especially nanobodies. At the same time, many of the cell surface protein receptors, from T-Cell to their target cells (TCRs, CD4, CD8, CD28, CTLA4, PD1, PDL1, etc.) are composed of Ig domains interacting through oligomeric pseudo-symmetric arrangements. We have developed methods for their structural analysis to inform and possibly design new Ig-based immunoreceptors. We use experimental and computational tools to explore these questions. Our modeling tools' application helped us characterize anti-CD19 and anti-BCMA CARs using variable CD8 and CD28 hinge components based on flexibility analysis [Brudno et al. 2020], suggesting that CAR toxicity can be related to its flexibility. Our exploration of CAR extracellular domains by X-ray crystallography techniques resulted in the first report on the formation of a spontaneous rearrangement of a CAR scFv mediated by quasi-symmetry, resulting in an unexpected VL-VL arrangement [PDBid: 7JO8 Cheung et al. 2020]. The use of NMR confirmed bioinformatics and structure prediction analyses on the intrinsic disorder character of the CD8 HIinge (Chen et al., 2022). These early promising results supported developing a pipeline for the rapid modeling of CARs at different scales of complexity and the exploration of their properties through computational techniques. These techniques are based on the application of a combination of data science methods, mining of big-data repositories, new approaches to in-silico structure analysis, and machine learning tools [Wang et al. 2020, Wang et al. 2022; Youkharibache et al. 2019, Youkharibache et al. 2020, Youkharibache et al. 2021]. The Immunoglobulin domain swapping observed [Cheung et al. 2020] gave support to our growing interest in exploring symmetry-related rearrangements as a data/knowledge organizing criteria and its application to antibody design, both stand-alone and as CAR antigen binding moieties. We expect to expand our experimental characterization of these domains by performing stability measurements to complement the available data. We plan to shift our experimental exploration of CARs by emphasizing ultrastructure analysis and CAR fragment analysis. Ultrastructure analysis (how CARs are organized in the membrane) by thin-section TEM studies may lead to Cryo-ET studies in favorable cases. The use of diffraction techniques to analyze CAR fragments may expedite the uncovering of some of the missing structures. The two approaches are complementary and will provide critical missing information to improve our understanding of these complex systems.

该项目的一个重要焦点是通过获得对其构成分子元件(抗体结合部分、铰链、跨膜和刺激结构域)之间相互作用的详细知识以及对其功能作用的表征来支持CARS(嵌合抗原受体)和抗体的精确自下而上工程。最先进的汽车设计策略依赖于作为组件的这些模块化领域的嵌合体的建立。这是通过组合过程完成的，使用昂贵的试错法来选择最有希望的候选人。在过去的几年里，我的研究集中在分子系统的自结合决定因素的研究上，特别是蛋白质，因为它们在分子组织的几个水平上的结构对称性：分子组装、蛋白质链、蛋白质结构域和蛋白质超二级结构(原始结构域)[Youkharibache 2019,2021]。虽然在大分子组装中对称性很高且明显，但在蛋白质结构域或受体-配体相互作用的水平上很难检测到对称性。例如，虽然抗体的重链和轻链对称性是众所周知的，但单个免疫球蛋白结构域本身由固有的伪对称原域组成[Youkharibache 2021]，这一特性被很大程度上忽略了，可以为抗体工程开辟新的途径，特别是纳米体。同时，许多细胞表面蛋白受体，从T细胞到它们的靶细胞(TCRs、CD4、CD8、CD28、CTLA4、PD1、PDL1等)。是由Ig结构域通过低聚伪对称排列相互作用组成的。我们已经开发了它们的结构分析方法，以告知并可能设计新的基于Ig的免疫受体。我们使用实验和计算工具来探索这些问题。我们的建模工具的应用程序帮助我们使用基于灵活性分析的可变CD8和CD28铰链组件来表征抗CD19和抗BCMA的汽车[Brudno等人。2020]，这表明汽车的毒性可能与其灵活性有关。我们用X射线结晶学技术对CAR胞外区的探索导致了第一个关于准对称性介导的CAR单链抗体自发重排形成的报道，导致了意想不到的VL-VL排列[PDBid：7JO8 Cheung等人]。2020年]。核磁共振的使用证实了对CD8铰链内在紊乱特征的生物信息学和结构预测分析(Chen等人，2022)。这些早期有希望的结果支持开发一条管道，用于对不同复杂程度的汽车进行快速建模，并通过计算技术探索其特性。这些技术基于数据科学方法、大数据存储库挖掘、硅内结构分析的新方法和机器学习工具的组合应用[Wang等人。2020年，Wang等人。2022年；Youkharibache等人。2019，Youkharibache等人。2020年，Youkharibache等人。2021年]。观察到免疫球蛋白结构域交换[Cheung et al.2020]支持了我们对探索对称相关重排作为数据/知识组织标准的日益增长的兴趣，以及它在抗体设计中的应用，包括独立的抗体设计和汽车抗原结合部分。我们希望通过进行稳定性测量来补充现有数据，从而扩大我们对这些区域的实验表征。我们计划通过强调超微结构分析和CAR碎片分析来转移我们对CARS的实验探索。超微结构分析(CAR在膜中是如何组织的)通过薄片的透射电子显微镜研究可能会导致在有利的情况下进行冷冻研究。使用衍射技术来分析汽车碎片可能会加速发现一些缺失的结构。这两种方法是相辅相成的，将提供关键的缺失信息，以提高我们对这些复杂系统的理解。