■基于LC-MS/MS的代谢组学是研究疾病相关生物标志物的重要工具。传统上,不同的策略被用于筛选生物标志物.然而,许多通过不同策略进行生物标志物筛选的研究忽视了生物标志物水平与暴露水平之间的剂量-反应关系,没有相关研究详细描述和比较不同的策略。苯巴比妥(PHB)属于巴比妥酸盐,被选为神经毒素的典型代表。酰基肉碱已成为几种神经系统疾病和神经毒性的诊断生物标志物的有希望的候选物。在这项工作中,我们旨在使用急性PHB中毒动物模型来阐明PHB中毒对血浆和脑酰基肉碱变化的影响,以及如何合理地分析LC-MS/MS的数据。
■在致动PHB中毒动物模型中利用血浆和脑区域中的酰基肉碱分布。通过方差分析趋势检验和Spearman秩相关检验评估血浆PHB与血浆和脑中肉碱和酰基肉碱(CARs)的剂量-反应关系。在不同的策略中,主成分分析(PCA)和偏最小二乘判别分析(OPLS-DA)筛选了差分CAR,可变重要性图(VIPs)用于选择PHB诱导毒性的推定生物标志物,和受试者工作特征(ROC)曲线分析,然后说明了生物标志物的可靠性。
■在第一种策略下,获得了14种潜在的毒性生物标志物,包括8种AUC>0.8的下调CAR。在第二种战略下,获得了11种潜在的毒性生物标志物,其包含5种下调的CAR,AUC>0.8。只有当剂量反应关系得到充分考虑时,不同的策略筛选相同的生物标志物(血浆乙酰肉碱(C2)和血浆癸酰肉碱(C10)),这表明血浆酰基肉碱可能作为毒性生物标志物。此外,血浆CAR水平变化与大脑CAR水平变化不同,血浆CAR与其大脑对应物之间的相关性较弱。
■我们发现血浆C2和C10可能作为PHB中毒疾病的毒性生物标志物,和PHB中毒对血浆CAR变化的影响可能不能完全代表脑CAR的变化。
UNASSIGNED: LC-MS/MS-based metabolomics is an important tool for studying disease-related biomarkers. Conventionally, different strategies have been used to screen biomarkers. However, many studies for biomarker screening by different strategies have ignored the dose-response relationship between the biomarker level and exposure level, and no relevant studies have described and compared different strategies in detail. Phenobarbital (PHB) which belongs to the barbiturates, was selected as the typical representative of neurotoxins.
Acylcarnitines have been promising candidates for diagnostic biomarkers for several neurological disorders and neurotoxicity. In this work, we aimed to use an acute PHB poisoning animal model to clarify PHB poisoning effects on plasma and brain acylcarnitine changes and how to rationally analyze data from LC-MS/MS.
UNASSIGNED: The acylcarnitine profiles in plasma and brain regions in an actuate PHB poisoning animal model were utilized. The dose-response relationship between plasma PHB and carnitine and
acylcarnitines (CARs) in plasma and brain were assessed by the variance analysis trend test and Spearman\'s rank correlation test. In different strategies, principal component analysis (PCA) and partial least squares discriminant analysis (OPLS-DA) screened the differential CARs, variable importance plots (VIPs) were utilized to select putative biomarkers for PHB-induced toxicity, and receiver operating characteristic (ROC) curve analysis then illustrated the reliability of biomarkers.
UNASSIGNED: Under the first strategy, 14 potential toxicity biomarkers were obtained including eight downregulated CARs with AUC >0.8. Under the second strategy, 11 potential toxicity biomarkers were obtained containing five downregulated CARs with AUC >0.8. Only when the dose-response relationship was fully considered, different strategies screen for the same biomarkers (plasma acetyl-carnitine (C2) and plasma decanoyl-carnitine (C10)), which indicated plasma
acylcarnitines might serve as toxicity biomarkers. In addition, the plasma CAR level changes showed differences from brain CAR level changes, and correlations between plasma CARs and their brain counterparts were weak.
UNASSIGNED: We found that plasma C2 and C10 might serve as toxicity biomarkers for PHB poisoning disorders, and PHB poisoning effects on changes in plasma CARs may not be fully representative of changes in brain CARs.