Co-exposure to multiple endocrine-disrupting chemicals and oxidative stress: Epidemiological evidence of nonmonotonic dose response curves

Abstract

This study aimed to investigate the effect of multiple exposure to eight classes (parabens, bisphenols, glycidyl ethers, antimicrobials, benzophenones, phthalates, tri and dichlorophenols) of endocrine disrupting chemicals (EDCs) on oxidative stress levels of 8-hydroxy-2'-deoxyguanosine (8-OHdG). A cross-sectional study was conducted with 300 healthy Brazilian children and adolescents. Urinary levels of 65 EDCs, creatinine and 8-OHdG were analyzed by Liquid Chromatography-Tandem Mass Spectrometry. Elastic net was used to estimate the associations between the levels of EDCs and 8-OHdG. The optimal hyperparameters were estimated using ten-fold cross-validation. Bayesian Kernel machine regression (BKMR) was used to investigate potential interactions and 8-OHdG level response as a function of the co-exposure to EDCs. The elastic net analysis showed that 2,4-DCP (0.149; CI 95 %:-0.033, 0.335, p = 0.02) and BPA (0.21; CI 95 %: 0.08; 0.356, p < 0.005) were associated with urinary levels of 8-OHdG. The BKMR model indicated a positive nonlinear and nonmonotonic relationship between EDCs mixture and 8-OHdG with an inverted U-shaped dose-response curve. This study suggests the first epidemiological evidence of a complex, nonmonotonic relationship between urinary levels of EDCs and 8-OHdG. However, the lack of established reference ranges for 8-OHdG limited a deeper discussion of our findings' clinical significance. Therefore, further studies should focus on validating our results across diverse populations, particularly those affected by oxidative stress-related diseases, and investigate potential mechanisms for supporting this association.

Keywords: 8-Hydroxy-2’-Deoxyguanosine; Bayesian kernel machine regression; DNA damage; Elastic net; Endocrine disruptors; Multi-pollutants mixture; Multiple environmental exposures.