Background: Endosulfan, an organochlorine compound, and fipronil, a phenylpyrazole, are insecticides with a common mechanism of toxicity. They interfere with Cl- influx by binding to the gamma-aminobutyric acid receptor (GABAAR) and blocking the inhibitory actions of GABAA. In vivo they cause neurotoxicity, hepatotoxicity, developmental toxicity, and can alter endocrine and immune systems. The thyroid is a target of fipronil toxicity. Human exposure occurs via food residues, skin contact and/or air dispersion. They are environmentally persistent and bioaccumulate in food chains.
Method: Compare in vivo data with in vitro results from the U.S. Environmental Protection Agency (USEPA) Toxicology Forecaster (ToxCast) high-throughput screening assays and zebrafish models to assess their usefulness in predicting toxicity.
Results: Fipronil’s in vivo toxicity occurred at lower doses than endosulfan for similar effects. ToxCast was a weak predictor of liver toxicity and estrogen receptor interaction. Missing is evidence of “true actives” for fipronil ToxCast assays with the thyroid receptor and for either compound with GABAAR or androgen receptors. Zebrafish models were good predictors of endosulfan and fipronil neurotoxicity in mammalian in vivo studies. Conclusion: ToxCast assays do not provide support for in vivo neurotoxicity or endocrine disruption where zebrafishs are good predictors of both parameters.
Marilyn H Silva and Svetlana Koshlukova