作者
Vishal Midya,Elena Colicino,Dinesh Kumar Barupal,Chris Gennings,Ruth J. F. Loos,Ryan W. Walker,Douglas I. Walker,Damaskini Valvi
摘要
BACKGROUND AND AIM: Growing experimental and epidemiology evidence suggests that exposure to perfluoroalkyl substances (PFAS) increase type 2 diabetes (T2D) risk; however, effects of PFAS-mixtures and underlying mechanisms are unclear. We examined associations of exposure to PFAS-mixture with incident T2D-risk and underlying plasma metabolic-pathways using untargeted metabolomics. METHODS: We conducted a nested case–control study within BioMe, an electronic health record-linked biobank of >60,000 patients seeking primary care at Mount Sinai Hospital, New York, since 2007. A total of 180 incident T2D cases (33% African Americans, 33% Hispanics, 33% Whites) were matched to 180 controls by age, sex and ancestry. We used liquid-chromatography with high-resolution mass spectrometry, to quantify seven PFAS (PFOS, PFOA, PFHxS, PFHpA, PFDA, PFHpS and PFNA) and measured untargeted metabolomic profiles in prediagnostic plasma. Cases had an average of ~6 years between blood draw and T2D diagnosis. We used (1)Weighted Quantile Sum(WQS) regression to evaluate the PFAS-mixture association with incident T2D risk; (2)performed metabolome-wide association analysis using Hierarchical-Bayesian WQS and logistic regression to separately analyze PFAS-mixture and T2D associations with 656 annotated metabolites, adjusting for confounders. Pathway enrichment analyses were performed using Mummichog. We used inverse probability weighting to account for nested case-control design in PFAS-metabolites analysis. Multiple comparisons were accounted by calculating q-values. RESULTS: Each tertile increase in PFAS-mixture(ng/mL) was associated with increased odds of incident T2D (OR[95%CI]=1.45[1.01, 2.09]) with PFOS and PFHpA having highest contributions to PFAS-mixture. T2D risk and PFAS-mixture associated top metabolites were aminoadipic acid, cortisol, gamma-glutamylglytamine, sulfolithocholylglycine, n-acetylaspartic acid and top pathways were alanine and aspartate metabolism, C21-steroid hormone biosynthesis and metabolism, androgen and estrogen biosynthesis and metabolism(p-value<0.05;q-value >0.2). CONCLUSIONS: Exposure to PFAS mixtures may dysregulate pathways of amino acid metabolism and steroid-hormone biosynthesis and metabolism, increasing risk of T2D diagnosis in multiethnic populations. KEYWORDS: Perfluoroalkyl-substances; Endocrine disrupting chemicals; Type 2 Diabetes; Metabolomics