Novel toxicity-associated metabolic pathways of benzotriazole UV stabilizers by cytochrome P450: mechanism-directed analysis
Shiyang Cheng, Lijing Su, Ye Han, Lingmin Jin, Juchen Ma, Chuiyuan Kong, Shubin Zhang, Jiawei Huang, Chunsheng Liu, Piotr Paneth, Li Ji · Environment International · 2025
Benzotriazole UV stabilizers (BZT-UVs) are emerging pollutants with environmental persistence and bioaccumulative risks, yet their metabolic pathways and associated toxicity remain poorly understood. We demonstrate that human cytochrome P450 metabolizes five structurally-diverse BZT-UVs (UV-P, UV-PS, UV-328, BZT, UV-329) into bioactive products: 1H-benzotriazole (persistent), o-benzoquinones and epoxides (reactive), and anti-estrogenic C–C dimers (toxic). These cryptic metabolic pathways were systematically uncovered using a mechanism-directed analysis (MDA) strategy. This hypothesis-driven approach prioritizes the elucidation of reaction mechanisms to predict and subsequently target novel metabolites by integrating density functional theory (DFT) predictions, in vitro assays in human liver microsomes, and mass spectrometry technologies. DFT predicted 24 novel metabolites via distinct pathways: Type-I ipso -decomposition (yielding o-benzoquinones and 1H-benzotriazoles/anions), Type-II ipso -decomposition (yielding hydroquinones and carbocations), homo-coupling (yielding C–C dimers), and C=C epoxidation (yielding epoxides). In vitro assays confirmed 18 metabolites of human P450 across all predicted categories. Critically, a C–C dimer (UV-P-C ortho –C′ ortho -UV-P), also detected in mice, exhibited potent anti-estrogenic activity (71.2 % inhibition at 1 μM), which was ∼ 60-fold more potent than the parent UV-P. Furthermore, the BZT epoxide formed DNA adducts in trapping assays, indicating a potential genotoxic mechanism. By anchoring computational models to physicochemical parameters (pKa, singly occupied molecular orbital, hydride ion affinity), this physical organic chemistry framework translates experimental and computational data into mechanistically interpretable metabolic networks. Therefore, our work underscores the necessity of incorporating metabolite profiles into risk assessments of BZT-UVs, and advocates MDA as a broadly applicable strategy to uncover toxicologically relevant metabolic pathways of emerging pollutants.