In vitro biotransformation of surfactants in fish. Part II--Alcohol ethoxylate (C16EO8) and alcohol ethoxylate sulfate (C14EO2S) to estimate bioconcentration potential.

Recent regulatory pressures (e.g., REACh, CEPA) requiring bioaccumulation assessments and the need for reduced animal use have increased the necessity for the development of in vitro-based methods to estimate bioaccumulation. Our study explored the potential use of subcellular and cellular hepatic systems to determine the biotransformation potential of two surfactants: octaethylene glycol monohexadecyl ether (C16EO8) and diethylene glycol monotetradecyl ether sulfate (C14EO2S). The subcellular systems tested were liver homogenates and microsomes from the common carp (Cyprinus carpio) and rainbow trout (Oncorhynchus mykiss). Cellular systems consisted of primary hepatocytes from the common carp (C. carpio) and PLHC-1 cells, hepatocarcinoma cells from the desert topminnow (Poeciliopsis lucida) cell line. Each in vitro system was exposed to radiolabeled test compounds and assayed for biotransformation using liquid scintillation and thin layer chromatographic methods. First-order kinetics were used to estimate rates of biotransformation. Bioconcentration of test materials in fish were predicted using an in vitro to in vivo metabolic rate extrapolation model linked to a mass-balance model commonly used to predict bioaccumulation in fish. Both subcellular and cellular tests using microsomes, liver homogenates and hepatocytes respectively showed biotransformation of the parent surfactants. Biotransformation rates were fastest for hepatocytes, followed by microsomes and homogenates. Rates were too low from homogenate tests to extrapolate to in vivo-based biotransformation rates using the extrapolation model. Trout microsomes metabolized C16EO8 faster than carp microsomes, yet rates were approximately the same for C14EO2S. Predicted BCF values incorporating in vitro biotransformation rates from hepatocytes were similar to measured in vivo or USEPA's bioconcentration model (BCFWIN) predicted values. Predicted BCF values using microsomal-based rates from trout and carp studies were only slightly less than default BCF values which assumes a linear logKow to BCF relationship with no biotransformation. However, hepatocyte-based results showed substantially decreased BCFs compared to the default BCF values. These results indicate that BCF estimates based on in vitro metabolic rates can provide reasonable estimates of in vivo BCF values, therefore, supporting the use of in vitro approaches within a tiered approach to assess bioconcentration.