Understanding mixed environmental exposures using metabolomics via a hierarchical community network model in a cohort of California women in 1960's

• Type of Material: Article
• Authors: Shuzhao Li, Emory University; Piera Cirillo, Center for Research on Women and Children’s Health; Xin Hu, Emory University; Tran ViLinh, Emory University; Nickilou Krigbaum, Center for Research on Women and Children’s Health; Shaojun Yu, Emory University; Dean Jones, Emory University; Barbara Cohn, Center for Research on Women and Children’s Health
• Language: English
• Date: 2020-03-01
• Publisher: PERGAMON-ELSEVIER SCIENCE LTD
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2019 Elsevier Inc. All rights reserved.
• License: Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1016/j.reprotox.2019.06.013
• Title of Journal or Parent Work: REPRODUCTIVE TOXICOLOGY
• Volume: 92
• Start Page: 57
• End Page: 65
• Grant/Funding Information: This project has been funded in part by the California Breast Cancer Research Program 21UB-8002 (Cohn), the National Institutes of Health S10OD18006 (Jones), UH2AI132345 (Li) and U01OD026489 (Li).
• Keywords: BREAST-CANCER; Variance analysis; Hierarchical community network; HEALTH; Gene environment interaction; Toxicology; EXPOSOME; WIDE ASSOCIATION; Reproductive Biology; Metabolomics; DDT; Breast cancer; RISK; Life Sciences & Biomedicine; RESPONSES; AGE; Exposome; PFAS; Science & Technology; Multi-omics integration; Metabolic phenotype; PCB; MWAS; SERUM; Mixed exposures; GENOTYPE

Abstract

Even though the majority of population studies in environmental health focus on a single factor, environmental exposure in the real world is a mixture of many chemicals. The concept of “exposome” leads to an intellectual framework of measuring many exposures in humans, and the emerging metabolomics technology offers a means to read out both the biological activity and environmental impact in the same dataset. How to integrate exposome and metabolome in data analysis is still challenging. Here, we employ a hierarchical community network to investigate the global associations between the metabolome and mixed exposures including DDTs, PFASs and PCBs, in a women cohort with sera collected in California in the 1960s. Strikingly, this analysis revealed that the metabolite communities associated with the exposures were non-specific and shared among exposures. This suggests that a small number of metabolic phenotypes may account for the response to a large class of environmental chemicals.