To improve current bioaccumulation assessment methods, a methodology is developed, applied, and investigated for measuring in vivo biotransformation rates of hydrophobic organic substances in the body (soma) and gastrointestinal tract of the fish. The method resembles the Organisation for Economic Co-operation and Development (OECD) 305 dietary bioaccumulation test but includes reference chemicals to determine both somatic and gastrointestinal biotransformation rates of test chemicals. Somatic biotransformation rate constants for the test chemicals ranged between 0 d(-1) and 0.38 (standard error [SE] 0.03)/d(-1) . Gastrointestinal biotransformation rate constants varied from 0 d(-1) to 46 (SE 7) d(-1) . Gastrointestinal biotransformation contributed more to the overall biotransformation in fish than somatic biotransformation for all test substances but 1. Results suggest that biomagnification tests can reveal the full extent of biotransformation in fish. The common presumption that the liver is the main site of biotransformation may not apply to many substances exposed through the diet. The results suggest that the application of quantitative structure-activity relationships (QSARs) for somatic biotransformation rates and hepatic in vitro models to assess the effect of biotransformation on bioaccumulation can underestimate biotransformation rates and overestimate the biomagnification potential of chemicals that are biotransformed in the gastrointestinal tract. With some modifications, the OECD 305 test can generate somatic and gastrointestinal biotransformation data to develop biotransformation QSARs and test in vitro-in vivo biotransformation extrapolation methods.

In vivo dietary bioaccumulation experiments for 85 hydrophobic organic substances were conducted to derive the in vivo gastrointestinal biotransformation rates, somatic biotransformation rates, bioconcentration factors (BCF), and biomagnification factors (BMF) for improving methods for bioaccumulation assessment and to develop an in vivo biotransformation rate database for QSAR development and in vitro to in vivo biotransformation rate extrapolation. The capacity of chemicals to be biotransformed in fish was found to be highly dependent on the route of exposure. Somatic biotransformation was the dominant pathway for most chemicals absorbed via the respiratory route. Intestinal biotransformation was the dominant metabolic pathway for most chemicals absorbed via the diet. For substances not biotransformed or transformed exclusively in the body of the fish, the BCF and BMF appeared to be closely correlated. For substances subject to intestinal biotransformation, the same correlation did not apply. We conclude that intestinal biotransformation and bioavailability in water can modulate the relationship between the BCF and BMF. This study also supports a fairly simple rule of thumb that may be useful in the interpretation of dietary bioaccumulation tests; i.e., chemicals with a BMF of <1 tend to exhibit BCFs based on either the freely dissolved (BCF) or the total concentration (BCF) of the chemical in the water that is less than 5000.

The Organisation for Economic Co-operation and Development guideline 305 for bioaccumulation testing in fish includes the option to conduct a dietary test for assessing a chemical's bioaccumulation behavior. However, the onecompartment toxicokinetic model that is used in the guidelines to analyze the results from dietary bioaccumulation tests is not consistent with the current state of the science, experimental practices, and information needs for bioaccumulation and risk assessment. The present study presents 1) a 2-compartment toxicokinetic modeling framework for describing the bioaccumulation of neutral hydrophobic organic chemicals in fish and 2) an associated toxicokinetic analysis tool (absorption, distribution, metabolism, and excretion [ADME] B calculator) for the analysis and interpretation of dietary bioaccumulation test data from OECD-305 dietary tests. The model framework and ADME-B calculator are illustrated by analysis of fish dietary bioaccumulation test data for 238 substances representing different structural classes and susceptibilities to biotransformation. The ADME of the chemicals is determined from dietary bioaccumulation tests and bioconcentration factors, biomagnification factors, and somatic and intestinal biotransformation rates. The 2-compartment fish toxicokinetic model can account for the effect of the exposure pathway on bioaccumulation, which the one-compartment model cannot. This insight is important for applying a weight-of-evidence approach to bioaccumulation assessment where information from aqueous and dietary test endpoints can be integrated to improve the evaluation of a chemical's bioaccumulation potential. Environ Toxicol Chem 2020;39:171-188.The depuration rate constant of the test (k BT ) and nonmetabolizable reference (k BT,R ) chemicals is the sum of the rate constants for respiratory elimination (k B2 ), fecal excretion (k BE ), somatic biotransformation (k BM , with k BM = 0 for k BT,R ) and growth dilution (k GD ; Equation 5 in Table 1). It can be derived wileyonlinelibrary.com/ETC © 2019 SETAC Interpreting OECD-305 dietary bioaccumulation tests-Environmental Toxicology and Chemistry, 2020;39:171-188 173 © 2019 SETAC wileyonlinelibrary.com/ETC Application and evaluation of the ADME-B calculatorThe results of the application of the ADME-B calculator to the dietary bioaccumulation test data of 238 test chemicals in wileyonlinelibrary.com/ETC

In vitro bioassays to estimate biotransformation rate constants of contaminants in fish are currently being investigated to improve bioaccumulation assessments of hydrophobic contaminants. The present study investigates the relationship between chemical substrate concentration and in vitro biotransformation rate of 4 environmental contaminants (9-methylanthracene, pyrene, chrysene, and benzo[a]pyrene) in rainbow trout (Oncorhynchus mykiss) liver S9 fractions and methods to determine maximum first-order biotransformation rate constants. Substrate depletion experiments using a series of initial substrate concentrations showed that in vitro biotransformation rates exhibit strong concentration dependence, consistent with a Michaelis-Menten kinetic model. The results indicate that depletion rate constants measured at initial substrate concentrations of 1 μM (a current convention) could underestimate the in vitro biotransformation potential and may cause bioconcentration factors to be overestimated if in vitro biotransformation rates are used to assess bioconcentration factors in fish. Depletion rate constants measured using thin-film sorbent dosing experiments were not statistically different from the maximum depletion rate constants derived using a series of solvent delivery-based depletion experiments for 3 of the 4 test chemicals. Multiple solvent delivery-based depletion experiments at a range of initial concentrations are recommended for determining the concentration dependence of in vitro biotransformation rates in fish liver fractions, whereas a single sorbent phase dosing experiment may be able to provide reasonable approximations of maximum depletion rates of very hydrophobic substances.

Incorporating biotransformation in bioaccumulation assessments of hydrophobic chemicals in both aquatic and terrestrial organisms in a simple, rapid, and cost-effective manner is urgently needed to improve bioaccumulation assessments of potentially bioaccumulative substances. One approach to estimate whole-animal biotransformation rate constants is to combine in vitro measurements of hepatic biotransformation kinetics with in vitro to in vivo extrapolation (IVIVE) and bioaccumulation modeling. An established IVIVE modeling approach exists for pharmaceuticals (referred to in the present study as IVIVE-Ph) and has recently been adapted for chemical bioaccumulation assessments in fish. The present study proposes and tests an alternative IVIVE-B technique to support bioaccumulation assessment of hydrophobic chemicals with a log octanol-water partition coefficient (K ) ≥ 4 in mammals. The IVIVE-B approach requires fewer physiological and physiochemical parameters than the IVIVE-Ph approach and does not involve interconversions between clearance and rate constants in the extrapolation. Using in vitro depletion rates, the results show that the IVIVE-B and IVIVE-Ph models yield similar estimates of rat whole-organism biotransformation rate constants for hypothetical chemicals with log K ≥ 4. The IVIVE-B approach generated in vivo biotransformation rate constants and biomagnification factors (BMFs) for benzo[a]pyrene that are within the range of empirical observations. The proposed IVIVE-B technique may be a useful tool for assessing BMFs of hydrophobic organic chemicals in mammals. Environ Toxicol Chem 2017;36:1934-1946. © 2016 SETAC.