Genetically encodable epitopes to overcome size and resolution limits in cryo-EM

摘要 低温电子显微镜(Cryo-EM)正在给结构生物学领域带来革命性的变化，因为它提供了比 长期和更频繁使用的技术，包括X射线结晶学和核磁学 共鸣。最近的技术进步已经开始扩大蛋白质的数量和类型 可以使用冷冻-EM来表征；然而，广泛采用该技术的主要障碍仍然是 是存在的。也就是说，在目标蛋白质的相对分子质量和分辨率之间存在负相关 这可以通过电子显微镜来实现，从而限制了该技术对非常大的蛋白质或 蛋白质复合体。目前，只有大于~100,000道尔顿的蛋白质才会产生与 分辨率可与使用x射线结晶学获得的分辨率相媲美。当前规避这一问题的方法 通常依赖于增加目标蛋白质的物理体积，通常通过识别特定的蛋白质 与研究中的蛋白质相互作用。实现这一点的一种常用方法是进化出高度特定的 针对目标蛋白的抗体，然后以Fabs的形式与目标蛋白结合。而当 一般来说，这种方法的缺点是必须为每个目标蛋白开发新的抗体， 这通常需要使用动物，既耗时又昂贵。此外，不能控制该网站 使用该方法提供了靶标上的Fab结合的结果。 在这里，我们建议通过开发一个单一残基表位来应对这一挑战，该表位的形式为 非正则氨基酸(NCAA)，可被现有抗体特异性识别。利用水井- 成熟的琥珀终止密码子抑制技术，NCAA可以在 基本上是目标蛋白质中的任何位置。然后，针对NCAA产生的抗体将有望 特异性地结合目标蛋白，其中表面暴露的残基已被NCAA取代。因为 这种方法将抗体结合的表位与目标蛋白的特征分离，它避免了 需要为研究中的每种蛋白质进化出一种新的抗体，并且还可以直接控制 FAB靶向的蛋白质。我们将在两个重点目标中开始探索这一可能性。 我们将首先使用之前报道的针对药物尼古丁的抗体来探测蛋白质铁蛋白的变体。 其中包含了含有尼古丁的NCAA。我们将使用这个模型系统来识别理想 含有NCAA的尼古丁的化学参数优化Fab结合并产生刚性蛋白质-蛋白质 界面。在第二个目标中，我们将探索我们的方法在铁蛋白和其他蛋白质中的普遍性。 尝试通过将我们的技术应用于非常小的蛋白质来突破冷冻-EM的大小限制。我们最终希望 为使用低温EM生成用于高分辨率结构确定的新工具包，该工具包既去除了现有的 关于使用光纤的限制，也允许资源有限的研究人员使用这项技术。

ABSTRACT Cryo-electron microscopy (cryo-EM) is revolutionizing the field of structural biology by providing advantages over long-standing and more frequently used techniques including x-ray crystallography and nuclear magnetic resonance. Recent technological advancements have begun to expand the number and types of proteins that can be characterized using cryo-EM; however, a major barrier to the widespread adoption of the technique still exists. Namely, an inverse correlation exists between the molecular weight of the target protein and the resolution that can be achieved by electron microscopy, thereby limiting the utility of the technique to very large proteins or protein complexes. At present, only proteins larger than ~100,000 Daltons routinely give rise to data with resolutions that rival those obtained using x-ray crystallography. Current approaches to circumvent this problem generally rely on increasing the physical bulk of the target protein, often by identifying proteins that specifically interact with the protein under study. A frequently employed method of achieving this is to evolve highly specific antibodies against the target protein, which are then bound to the target protein in the form of Fabs. While general, this method suffers from the drawbacks that new antibodies must be developed for each target protein, which often requires the use of animals and is time consuming and costly. Furthermore, no control over the site of Fab binding on the target is afforded using this method. Here, we propose to address this challenge by developing a single residue “epitope” in the form of a non-canonical amino acid (NCAA) that is specifically recognized by an existing antibody. Using the well- established amber stop codon suppression technology, NCAAs can be site-specifically incorporated at essentially any position in a target protein. Antibodies raised against the NCAA would then be expected to specifically bind a target protein in which a surface-exposed residue had been replaced with the NCAA. Because this approach decouples the epitope bound by the antibody from features of the target protein, it obviates the need to evolve a new antibody for each protein under study and also affords direct control over the region of the protein targeted by the Fab. We will begin to explore this possibility in two focused aims. We will first use a previously reported antibody against the drug nicotine to probe variants of the protein ferritin in which nicotine-containing NCAAs have been incorporated. We will use this model system to identify ideal chemical parameters of the nicotine containing NCAA that optimize Fab binding and create a rigid protein-protein interface. In a second aim, we will explore the generality of our approach in proteins other than ferritin and attempt to push the size limits of cryo-EM by applying our technique to very small proteins. We ultimately hope to generate a new toolkit for high resolution structure determination using cryo-EM that both removes existing limitations regarding the use of Fabs and also allows the use of this technique by resource-limited researchers.