Project 1: Discovery of proteins with altered abundance and stability

摘要 尽管 CSF Aβ42、tau、磷酸化 tau 水平与潜在 AD 之间存在关联 病理学、临床诊断生物标志物准确性的测量在研究之间差异很大。鉴于此 其他神经退行性疾病也可能出现类似 AD 的临床症状，并且 AD 患者 经常有合并症，需要额外的标记物来帮助鉴别诊断和 识别混合病理。在过去的十年中，许多候选生物标志物已被确定，反映了 一系列病理生理过程，包括胆固醇代谢、神经炎症和淀粉样蛋白 处理。然而，很少有人在临床实践中采用或在大型独立机构中得到验证。 队列1. MacCoss 实验室和其他实验室一直致力于下一代蛋白质组学方法的开发 作为经典的基于随机质谱的方法的替代方法。这些新方法提供了 目标蛋白质组学策略和全球蛋白质组学策略的混合体。虽然质谱数据是在一个 以公正的方式，以有针对性的策略分析数据，其中分析特定的肽（尽管是 1000 个） 使用先验信息。因此，可重复的靶向、通量和基于 MS/MS 的置信度 平行反应监测（PRM）的量化可以与经典发现方法的能力相结合 定性检测数千种蛋白质。这些新方法基于系统收集的质量 光谱数据可以提供类似的定量品质因数，并且可以以类似的方式进行验证 临床分析。 尽管蛋白质组学技术取得了进步，但大多数发现新 CSF 标记的尝试要么 使用 1) 定量特征较差的随机抽样方法（例如数据相关采集） 性能，2) 小样本队列，3) 完全关注总脑脊液，4) 忽略蛋白质加工，以及 5) 没有考虑蛋白质错误折叠或稳定性。该项目在合作研究中的目的 计划是将我们的方法提升到另一个水平——将真正的定量方法应用于大型油井 表征队列并将其扩展到功能相关的亚群。 我们的脑脊液检测可以从完全未分级的材料中测量 >1050 种蛋白质，并附有数字 日内/日间精度、线性、LOD/LOQ 等优点...此外，我们可以将此测定用于 CSF 蛋白质组的子集，用于评估神经生物学的功能相关方面，包括数量 溶液或脑脊液颗粒的一部分、完整蛋白质的分子量和蛋白质稳定性。最后，我们有足够的吞吐量 在足以消除观察结果由异常引起的可能性的范围内应用这些分析 亚人群。

Abstract Despite the association between the levels of CSF Aβ42, tau, phosphorylated tau and underlying AD pathology, measures of biomarker accuracy for clinical diagnosis vary widely between studies. Given that other neurodegenerative conditions can present with AD-like clinical symptoms, and individuals with AD frequently have comorbid pathologies, additional markers are needed that can aid in differential diagnosis and identify mixed pathologies. Over the last decade, many candidate biomarkers have been identified, reflecting a range of pathophysiological processes including cholesterol metabolism, neuroinflammation and amyloid processing. However, few, have been adopted in clinical practice or been validated in large independent cohorts1. The MacCoss lab and others have been pioneering the development of next generation proteomics methods as an alternative to the classic stochastic mass spectrometry-based methods. These new methods offer a hybrid between a targeted and global proteomics strategy. While mass spectrometry data is collected in an unbiased way, the data is analyzed in a targeted strategy where specific peptides, albeit 1000s are analyzed using prior information. Thus, the reproducible targeting, throughput, and confident MS/MS-based quantification of parallel reaction monitoring (PRM) can be combined with classical discovery methods' ability to qualitatively detect thousands of proteins. These new methods based on systematically collected mass spectrometry data can offer similar quantitative figures of merit and can be validated in analogous fashion to clinical assays. Despite advances in proteomics technologies, most attempts at discovering new CSF markers have either used 1) stochastic sampling methods (e.g. data dependent acquisition) with poorly characterized quantitative performance, 2) small sample cohorts, 3) focused entirely on total CSF, 4) ignored protein processing, and 5) did not consider protein misfolding or stability. The purpose of this project within the cooperative research program is to take our methods to another level – apply true quantitative methods to large well characterized cohorts and extend them to functionally relevant subpopulations. We have a CSF assay that can measure >1050 proteins from completely unfractionated material with figures of merit of within/between day precision, linearity, LOD/LOQ, etc... Furthermore, we can use this assay on subsets of the CSF proteome to assess functionally relevant aspects of the neurobiology including quantity as part of solution or CSF particles, intact protein MW, and protein stability. Finally, we have enough throughput to apply these analyses on a scale sufficient to eliminate the chance that an observation is due to an aberrant subpopulation.