Transcriptomics provides mechanistic insights into chemical toxicity and serves as a hypothesis-generating tool for prioritizing potential adverse outcomes. Here, we introduced a transcriptomics-guided outcome prediction (T-GOP) framework, a hypothesis-informed approach that uses transcriptomic enrichment to prioritize end points for targeted experimental validation. As a case study, the ecotoxicological effects of the PFOS alternative, sodium p-perfluorous nonenoxybenzenesulfonate (OBS), were evaluated. After 28 days of exposure to environmentally relevant OBS concentrations (0.1, 1.0, and 10 μg/L), adult zebrafish accumulated OBS in the brain (363-2364 μg/kg), triggering extensive transcriptional reprogramming with 61, 134, and 1026 differentially expressed genes at the respective exposure levels. Transcriptomic analysis implicated disruption of neurotransmitter pathways, which was confirmed by targeted metabolomics, revealing profound alterations in dopaminergic and serotonergic systems. These neurochemical perturbations coincided with concentration-dependent downregulation of essential neuronal genes (e.g., bdnf, syn2a, and elavl3), increased acetylcholinesterase (AChE) activity, and brain histopathological changes. At the highest concentration, T-maze assays revealed increased memory latency, consistent with cognitive impairment as an apical outcome of the observed upstream perturbations. Benchmark concentration modeling indicated neurotoxic responses with BMC10 values ranging from 0.03 to 8.97 μg/L, with corresponding 95% credible intervals (BMCL10-BMCU10) of 0.01-20.04 μg/L. Overall, this proof-of-concept framework provides evidence for the neurotoxicity of OBS and highlights its potential environmental risk.
Zeng et al. (Fri,) studied this question.