Chen Jin Y, Sutaria Saurin R, Xie Zhengzhi, Kulkarni Manjiri, Keith Rachel J, Bhatnagar Aruni, Sears Clara G, Lorkiewicz Pawel, Srivastava Sanjay
Christina Lee Brown Envirome Institute, University of Louisville, Louisville, KY 40202, United States; Superfund Research Center, University of Louisville, Louisville, KY 40202, United States; American Heart Association-Tobacco Regulation and Addiction Center, University of Louisville, Louisville, KY 40202, United States; Division of Environmental Medicine, Department of Medicine, University of Louisville, Louisville, KY 40202, United States.
Christina Lee Brown Envirome Institute, University of Louisville, Louisville, KY 40202, United States; Superfund Research Center, University of Louisville, Louisville, KY 40202, United States; American Heart Association-Tobacco Regulation and Addiction Center, University of Louisville, Louisville, KY 40202, United States; Division of Environmental Medicine, Department of Medicine, University of Louisville, Louisville, KY 40202, United States; Bellarmine University, Louisville, KY 40205, United States.
Environ Int. 2025 May;199:109516. doi: 10.1016/j.envint.2025.109516. Epub 2025 May 4.
Humans are constantly exposed to both naturally-occurring and anthropogenic chemicals. Targeted mass spectrometry approaches are frequently used to measure a small panel of chemicals and their metabolites in environmental or biological matrices, but methods for comprehensive individual-level exposure assessment are limited. In this study, we applied an integrated library-guided analysis (ILGA) with ultraperformance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-QTOF/MS) to profile phase II metabolites, specifically mercapturic acids (MAs), glucuronic acids (GAs), and sulfates (SAs) in human urine samples (n = 844). We annotated 424 metabolites (146 MAs, 171 GAs, 107 SAs) by querying chromatographic features with in-house structural libraries, filtering against fragmentation patterns (common neutral loss and ion fragment), and comparing mass spectra with in-silico fragmentations and external spectral libraries. These metabolites were derived from over 200 putative parent compounds of exogenous and endogenous sources, such as dietary compounds, benzene/monocyclic substituted aromatics, pharmaceuticals, polycyclic aromatic hydrocarbons, bile acids/bile salts, and 4-hydroxyalkenals associated with lipid peroxidation process. Further, we performed statistical analyses on 214 metabolites found in more than 75% of samples to examine the association between metabolites and demographic characteristics using integrated network analysis, principal component analysis (PCA), and multivariable linear regression models. The network analysis revealed four distinct communities of 37 positively correlated metabolites, and the PCA (using the 37 metabolites) presented 4 principal components that meaningfully explained at least 80% of the variance in the data. The multivariable linear regression models showed some positive and negative associations between metabolite profiles and certain demographic variables (e.g., age, sex, race, education, income, and tobacco use).
人类不断接触天然存在的和人为产生的化学物质。靶向质谱方法经常用于测量环境或生物基质中的一小部分化学物质及其代谢物，但用于全面个体水平暴露评估的方法有限。在本研究中，我们应用了一种集成库引导分析（ILGA），结合超高效液相色谱-四极杆飞行时间质谱（UPLC-QTOF/MS）来分析人类尿液样本（n = 844）中的II相代谢物，特别是巯基尿酸（MAs）、葡萄糖醛酸（GAs）和硫酸盐（SAs）。我们通过使用内部结构库查询色谱特征、根据碎片模式（常见中性丢失和离子碎片）进行筛选以及将质谱与计算机模拟碎片和外部光谱库进行比较，注释了424种代谢物（146种MAs、171种GAs、107种SAs）。这些代谢物来源于200多种外源性和内源性推定母体化合物，如膳食化合物、苯/单环取代芳烃、药物、多环芳烃、胆汁酸/胆盐以及与脂质过氧化过程相关的4-羟基烯醛。此外，我们对在超过75%的样本中发现的214种代谢物进行了统计分析，使用集成网络分析、主成分分析（PCA）和多变量线性回归模型来检查代谢物与人口统计学特征之间的关联。网络分析揭示了37种正相关代谢物的四个不同群落，PCA（使用这37种代谢物）呈现了4个主成分，这些主成分有意义地解释了数据中至少80%的方差。多变量线性回归模型显示代谢物谱与某些人口统计学变量（如年龄、性别、种族、教育程度、收入和烟草使用）之间存在一些正相关和负相关关系。
