Evaluation of a body-residue-based one-compartment first-order kinetic model for estimating the toxicity of mixtures of chlorinated organic contaminants to rainbow trout, Oncorhynchus mykiss

Abstract

A general lack of understanding and inadequate predictive models of toxicant interactions have made it difficult for regulatory agencies to provide water quality guidelines that effectively consider chemical mixtures. Given the amounts and types of compounds found in aquatic ecosystems, research in this area is important in order to ensure that mixtures of chemicals are addressed in a way that provides adequate protection for aquatic organisms. A whole body-residue-based one-compartment first-order kinetic (1CFOK) model was used to examine the combined effects of chlorinated phenols and benzenes on rainbow trout (Oncorhynchus mykiss). The use of a kinetic model allows the separation and study of two important aspects involved in the interaction of toxicant mixtures: the influence of one chemical on the kinetics of the other i.e. the effect on uptake and/or depuration, and the influence of one chemical on the resulting toxicity of the other once they have reached their respective sites of toxic action. This approach is relatively well established but only recently have technological advances permitted an extensive analytical evaluation. The overall objective of this research was to assess the ability of the body-residue approach to predict the toxicity of binary mixtures of toxicants to fish.

In order to use a 1CFOK model to predict chemical loading into fish tissue and the potential effects of those residues on organism health, a number of biological and chemical parameters must be determined for each of the compounds to be studied. These parameters include: rate constants for chemical uptake and depuration, bioconcentration factors, and critical body-residues for the effects being studied. These parameters must be derived within the system in which the mixture studies will be done in order to compensate for potential modifying factors. The rate constants for chemical uptake and depuration are derived using kinetic experiments in which the body-residues resulting from waterborne exposure were tracked over time. The critical body-residues and times-to-death are defined using a standard toxicity test format with the addition of quantifying chemical residues in samples of the dead fish.

Based on the results, the 1CFOK model was shown to adequately describe the uptake and depuration of 1,2,4-trichlorobenzene, 2,4,5-trichlorophenol, 2,4,6-trichlorophenol and pentachlorophenol in rainbow trout larvae. Suitable measures of k1, k2 and the BCF were derived from each compound which could subsequently be used to predict the chemical concentration in the organism at any time for acute exposure to multiple chemicals. A critical body-residue was used to define the chemical toxicity in terms of the chemical concentration in the fish. The critical body-residues for 1,2,4-trichlorobenzene, 2,4,5-trichlorophenol, 2,4,6-trichlorophenol and pentachlorophenol were shown to be independent of the exposure time and water concentration, and thus could be used, in conjunction with the 1CFOK modelling to accurately predict lethality in an exposured group of organisms.

For binary mixture scenarios, the model was used to track the concentration of each chemical in the whole organism over time. By expressing this predicted value as a ratio of the chemicals respective critical body-residue and summing these ratios, it is possible to predict both if death will occur and when it will occur. It was possible to validate the models predictions by performing the mixture experiments and noting the time to death, the number of mortalities and by quantifying the body-residues in each of the organisms at the time of death. A comparison of the results of the mixture toxicity tests with those predicted from the model demonstrate that the effects on larval rainbow trout of binary mixtures of 1,2,4-trichlorobenzene, 2,4,5-trichlorophenol, 2,4,6-trichlorophenol and pentachlorophenol were accurately modelled, both in terms of exposure and toxicity, using the 1CFOK model and a body-residue-based toxicity endpoint. For the most part, the interaction of the chemicals in the mixture could be described as additive. Deviations from this were noted for the mixture of 2,4,5-trichlorophenol and 2,4,6-trichlorophenol but were marginal. The exact nature of the interaction was not determined but changes in the rates of chemical uptake and depuration by the fish were not responsible. It was concluded that the potential toxicity to fish of mixtures of chlorinated phenols and benzenes should be adequately predicted by assuming addition.

In addition, the research was completely dependent on the ability to quantify chemical residues in the organism. It became rapidly apparent that a new approach to residue analysis was required. Traditional methods of organic extraction using separatory funnels were both inefficient and labour intensive making analysis of the thousands of tissue samples an impossible task. The Cryo-extraction protocol was developed in order to accurately and rapidly accomplish tissue residue analysis. Through the systematic assessment of the extraction efficiency, reproducibility and linearity, the cryo-extraction protocol for quantifying the chemical body-residues of chlorinated benzenes and phenols in fish. The cryo-extraction protocol provided a faster, more efficient and more reliable approach.