Berezhnoy Georgy, Bae Gyuntae, Wüst Leonie, Schulte Claudia, Cannet Claire, Wurster Isabel, Zimmermann Milan, Jäck Alexander, Spruth Eike Jakob, Hellmann-Regen Julian, Roeske Sandra, Pürner Dominik, Glanz Wenzel, Maass Fabian, Hufschmidt Felix, Kilimann Ingo, Dinter Elisabeth, Kimmich Okka, Gamez Anna, Levin Johannes, Priller Josef, Peters Oliver, Wagner Michael, Storch Alexander, Lingor Paul, Düzel Emrah, van Riesen Christoph, Wüllner Ullrich, Teipel Stefan, Falkenburger Björn, Bähr Mathias, Zerr Inga, Petzold Gabor C, Spottke Annika, Rizzu Patricia, Brosseron Frederic, Schäfer Hartmut, Gasser Thomas, Trautwein Christoph
Werner Siemens Imaging Center, Department of Preclinical Imaging and Radiopharmacy, University of Tübingen, Tübingen, Germany.
Hertie Institute for Clinical Brain Research, Department of Neurodegenerative Diseases, University of Tübingen, Tübingen, Germany.
Sci Rep. 2025 May 22;15(1):17738. doi: 10.1038/s41598-025-01352-0.
The challenge of early detection and stratification in Parkinson's disease (PD) is urgent due to the current emergence of mechanism-based disease-modifying treatments. In here, metabolomic and lipidomic parameters obtained by a standardized and targeted in vitro diagnostic research (IVDr) platform have a significant potential to address therapy-related questions and generate improved biomarker panels. Our study aimed to use IVDr nuclear magnetic resonance (NMR) spectroscopy to quantify metabolites and lipoproteins in PD blood serum from different cohorts to stratify metabolically driven subtypes of idiopathic and genetic PD. Serum aliquots from three neurodegeneration biobank cohorts (287 samples in total, including 62 PD patient samples with GBA mutation, 98/43 PD patient samples of early/late stages of disease duration, 20 PD samples from patients with mutations in recessive PD genes and some smaller subgroups of mitochondrial and double mutation cases) were prepared and analyzed with IVDr NMR spectroscopy, covering 39 blood serum metabolites and 112 lipoprotein parameters. Uni- and multivariate statistics were used to identify metabolism-driven changes under consideration of typical confounders such as age, sex and disease duration and set into context with clinical biomarkers such as CSF concentrations of alpha-synuclein, neurofilament light chain, and tau protein. Based on the different PD subgroups we performed a total of eight different comparisons. Highlights from these comparisons include increased citrate and dimethylglycine with a decrease of creatinine and methionine in healthy controls and early PD group compared to GBA, PD late and recessive PD. We furthermore identified decreased HDL-3 free cholesterol in genetic PD cases compared to sporadic subject samples (sum of the PD early and PD late groups). Considering medication, we found that the levodopa equivalent daily dose (LEDD) is mostly positively correlated with tyrosine and citrate in sporadic PD compared to pyruvate and phenylalanine in genetic PD. Cerebrospinal fluid levels of alpha-synuclein were negatively correlated with alanine. Further metabolites and lipoproteins with discriminatory power for double mutation PD cases involved ornithine, 2-aminobutyrate and 2-hydroxybutyrate as well as for mitochondrial phenotypes via LDL phospholipid, apolipoprotein and cholesterol subfractions. Quantitative IVDr NMR serum spectroscopy is able to stratify PD patient samples of different etiology and can contribute to a wider understanding of the underlying metabolism-driven alterations e.g. in energy, amino acid, and lipoprotein metabolism. Though our overall cohort was large, major confounders such as age, sex and medication have a strong impact. That is why absolute quantification and detailed patient knowledge about metabolic confounders, is a premise for future translation of NMR serum spectroscopy to routine PD diagnostics.
由于目前基于机制的疾病修饰疗法的出现，帕金森病（PD）早期检测和分层的挑战迫在眉睫。在此，通过标准化的靶向体外诊断研究（IVDr）平台获得的代谢组学和脂质组学参数具有解决治疗相关问题并生成改进的生物标志物组的巨大潜力。我们的研究旨在使用IVDr核磁共振（NMR）光谱对来自不同队列的PD血清中的代谢物和脂蛋白进行定量，以对特发性和遗传性PD的代谢驱动亚型进行分层。制备了来自三个神经退行性疾病生物样本库队列的血清等分试样（总共287个样本，包括62个具有GBA突变的PD患者样本、98/43个疾病持续时间早期/晚期的PD患者样本、20个来自隐性PD基因突变患者的PD样本以及一些线粒体和双重突变病例的较小亚组），并使用IVDr NMR光谱进行分析，涵盖39种血清代谢物和112种脂蛋白参数。使用单变量和多变量统计来识别在考虑年龄、性别和疾病持续时间等典型混杂因素下由代谢驱动的变化，并与临床生物标志物如脑脊液中α-突触核蛋白、神经丝轻链和tau蛋白的浓度相结合。基于不同的PD亚组，我们总共进行了八种不同的比较。这些比较的亮点包括，与GBA、晚期PD和隐性PD相比，健康对照和早期PD组中柠檬酸和二甲基甘氨酸增加，而肌酐和蛋氨酸减少。此外，我们发现与散发性受试者样本（PD早期和晚期组的总和）相比，遗传性PD病例中HDL-3游离胆固醇降低。考虑到药物治疗，我们发现与遗传性PD中的丙酮酸和苯丙氨酸相比，散发性PD中左旋多巴等效日剂量（LEDD）大多与酪氨酸和柠檬酸呈正相关。脑脊液中α-突触核蛋白水平与丙氨酸呈负相关。对双重突变PD病例具有鉴别能力的其他代谢物和脂蛋白包括鸟氨酸、2-氨基丁酸和2-羟基丁酸，以及通过低密度脂蛋白磷脂、载脂蛋白和胆固醇亚组分对线粒体表型具有鉴别能力的物质。定量IVDr NMR血清光谱能够对不同病因的PD患者样本进行分层，并有助于更广泛地理解潜在的代谢驱动变化，例如能量、氨基酸和脂蛋白代谢方面的变化。尽管我们的总体队列规模较大，但年龄、性别和药物治疗等主要混杂因素有很大影响。这就是为什么绝对定量以及患者对代谢混杂因素的详细了解，是未来将NMR血清光谱转化为常规PD诊断的前提。
