Metabolomic and Transcriptomic Analysis of MCF-7 Cells Exposed to 23 Chemicals at Human-Relevant Levels: Estimation of Individual Chemical Contribution to Effects.

Background: Humans are constantly being exposed to various xenobiotics at relatively low concentrations. To date, limited evidence is available to ascertain whether a complex xenobiotic mixture at human-relevant levels causes any health effect. Moreover, there is no effective method to pinpoint the contribution of each chemical toward such an effect.

Objectives: This study aims to understand the responses of cells to a mixture containing 23 xenobiotics at human-relevant levels and develop a feasible method to decipher the chemical(s) that contribute significantly to the observed effect.

Methods: We characterized the metabolome and transcriptome of breast cancer cells (MCF-7) before and after exposure to the mixture at human-relevant levels; preexposure levels were derived from existing large-scale biomonitoring data. A high-throughput metabolomics-based "leave-one-out" method was proposed to understand the relative contribution of each component by comparing the metabolome with and without the particular chemical in the mixture.

Results: The metabolomic analysis suggested that the mixture altered metabolites associated with cell proliferation and oxidative stress. For the transcriptomes, gene ontology terms and pathways including "cell cycle," "cell proliferation," and "cell division" were significantly altered after mixture exposure. The mixture altered genes associated with pathways such as "genotoxicity" and "nuclear factor erythroid 2-related factor 2 (Nrf2)." Through joint pathways analysis, metabolites and genes were observed to be well-aligned in pyrimidine and purine metabolisms. The leave-one-out results showed that many chemicals made their contributions to specific metabolic pathways. The overall metabolome pattern of the absence of 2,4-dihyroxybenzophenone (DHB) or bisphenol A (BPA) showed great resemblance to controls, suggesting their higher relative contribution to the observed effect.

Discussion: The omics results showed that exposure to the mixture at human-relevant levels can induce significant cellular changes. Also, the leave one out method offers an effective approach for deconvoluting the effects of the mixture. https://doi.org/10.1289/EHP6641.