Alzheimer’s disease (AD) is the most common cause of dementia, characterized by progressive cognitive decline. Protein biomarkers of AD brain pathology, including β-amyloid and Tau, are reflected in cerebrospinal fluid (CSF), yet the identification of additional biomarkers linked to other brain pathophysiologies remains elusive. We recently reported a multiplex tandem-mass tag (TMT) CSF proteomic analysis of nearly 3000 proteins, following depletion of highly abundant proteins and off-line fractionation, across control and AD cases. Of these, over 500 proteins were significantly increased or decreased in AD, including markers reflecting diverse biological functions in brain. Here, we use a targeted mass spectrometry (MS) approach, termed parallel reaction monitoring (PRM), to quantify select CSF biomarkers without pre-depletion or fractionation to assess the reproducibility of our findings and the specificity of changes for AD versus other causes of cognitive impairment. We nominated 41 proteins (94 peptides) from the TMT CSF discovery dataset, representing a variety of brain cell-types and biological functions, for label-free PRM analysis in a replication cohort of 88 individuals that included 20 normal controls, 37 clinically diagnosed AD cases and 31 cases with non-AD cognitive impairment. To control for technical variables, isotopically labeled synthetic heavy peptide standards were added into each of the 88 CSF tryptic digests. Furthermore, a peptide pool, representing an equivalent amount of peptide from all samples, was analyzed (n = 10) across each batch. Together, this approach enabled us to assess both the intra- and inter-sample differences in peptide signal response and retention time. Despite differences in sample preparation, quantitative MS approaches and patient samples, 25 proteins, including Tau, had a consistent and significant change in AD in both the discovery and replication cohorts. Validated CSF markers with low coefficient of variation included the protein products for neuronal/synaptic (GDA, GAP43, SYN1, BASP1, YWHAB, YWHAZ, UCHL1, STMN1 and MAP1B), glial/inflammation (SMOC1, ITGAM, CHI3L1, SPP1, and CHIT1) and metabolic (PKM, ALDOA and FABP3) related genes. Logistical regression analyses revealed several proteins with high sensitivity and specificity for classifying AD cases from controls and other non-AD dementias. SMOC1, YWHAZ, ALDOA and MAP1B emerged as biomarker candidates that could best discriminate between individuals with AD and non-AD cognitive impairment as well as Tau/β-amyloid ratio. Notably, SMOC1 levels in postmortem brain are highly correlated with AD pathology even in the preclinical stage of disease, indicating that CSF SMOC1 levels reflect underlying brain pathology specific for AD. Collectively these findings highlight the utility of targeted MS approaches to quantify biomarkers associated with AD that could be used for monitoring disease progression, stratifying patients for clinical trials and measuring therapeutic response.

中文翻译：

---

靶向质谱法量化阿尔茨海默病中脑源性脑脊液生物标志物。


阿尔茨海默病 (AD) 是痴呆症的最常见原因，其特征是进行性认知能力下降。 AD 脑病理学的蛋白质生物标志物，包括 β-淀粉样蛋白和 Tau，反映在脑脊液 (CSF) 中，但与其他脑病理生理学相关的其他生物标志物的鉴定仍然难以捉摸。我们最近报道了对对照和 AD 病例中近 3000 种蛋白质的多重串联质量标签 (TMT) CSF 蛋白质组学分析，在高丰度蛋白质耗尽和离线分级分离之后。其中，超过 500 种蛋白质在 AD 中显着增加或减少，包括反映大脑不同生物功能的标记物。在这里，我们使用一种称为平行反应监测 (PRM) 的靶向质谱 (MS) 方法来量化选定的 CSF 生物标志物，无需预耗尽或分馏，以评估我们的研究结果的再现性以及 AD 变化相对于其他原因的特异性。认知障碍。我们从 TMT CSF 发现数据集中提名了 41 种蛋白质（94 种肽），代表各种脑细胞类型和生物功能，在 88 名个体的复制队列中进行无标记 PRM 分析，其中包括 20 名正常对照、37 名临床诊断的 AD例和非 AD 认知障碍 31 例。为了控制技术变量，将同位素标记的合成重肽标准品添加到 88 个 CSF 胰蛋白酶消化物中的每一个中。此外，还对代表所有样品中等量肽的肽库进行了每批分析 (n = 10)。总之，这种方法使我们能够评估样品内和样品间肽信号响应和保留时间的差异。 尽管样品制备、定量 MS 方法和患者样品存在差异，但包括 Tau 在内的 25 种蛋白质在发现和复制队列中的 AD 方面均出现一致且显着的变化。经验证的变异系数较低的 CSF 标记物包括神经元/突触（GDA、GAP43、SYN1、BASP1、YWHAB、YWHAZ、UCHL1、STMN1 和 MAP1B）、神经胶质/炎症（SMOC1、ITGAM、CHI3L1、SPP1 和 CHIT1）的蛋白质产物）和代谢（PKM、ALDOA 和 FABP3）相关基因。逻辑回归分析揭示了几种具有高敏感性和特异性的蛋白质，可用于将 AD 病例与对照组和其他非 AD 痴呆症进行分类。 SMOC1、YWHAZ、ALDOA 和 MAP1B 成为候选生物标志物，可以最好地区分 AD 和非 AD 认知障碍患者以及 Tau/β-淀粉样蛋白比率。值得注意的是，即使在疾病的临床前阶段，死后大脑中的 SMOC1 水平也与 AD 病理学高度相关，这表明 CSF SMOC1 水平反映了 AD 特异性的潜在大脑病理学。总的来说，这些发现强调了靶向 MS 方法在量化与 AD 相关的生物标志物方面的实用性，这些生物标志物可用于监测疾病进展、对临床试验患者进行分层和测量治疗反应。