Predicting the Bioconcentration of Fragrance Ingredients by Rainbow Trout Using Measured Rates of<i>in Vitro</i>Intrinsic Clearance

Bioaccumulation in aquatic species is a critical end point in the regulatory assessment of chemicals. Few measured fish bioconcentration factors (BCFs) are available for fragrance ingredients. Thus, predictive models are often used to estimate their BCFs. Because biotransformation can reduce chemical accumulation in fish, models using QSAR-estimated biotransformation rates have been developed. Alternatively, biotransformation can be measured by in vitro methods. In this study, biotransformation rates for nine fragrance ingredients were measured using trout liver S9 fractions and used as inputs to a recently refined in vitro-in vivo extrapolation (IVIVE) model. BCFs predicted by the model were then compared to (i) in vivo BCFs, (ii) BCFs predicted using QSAR-derived biotransformation rates, (iii) BCFs predicted without biotransformation, and (iv) BCFs predicted by a well-known regression model. For fragrance ingredients with relatively low (<4.7) log K(OW) values, all models predicted BCFs below a bioaccumulation threshold of 1000. For chemicals with higher (4.7-5.8) log K(OW) values, the model incorporating measured in vitro biotransformation rates and assuming no correction for potential binding effects on hepatic clearance provided the most accurate predictions of measured BCFs. This study demonstrates the value of integrating measured biotransformation rates for prediction of chemical bioaccumulation in fish.