Quantitative structure-activity relationships for fish early life stage toxicity

Abstract

Early life stage (ELS) toxicity experiments were carried out for both chlorobenzene and aniline derivatives under semistatic conditions using zebra fish ( Brachydanio rerio). The parameters studied included survival, embryo-hatchability and growth. From these effect parameters and the n-octanol/water partition coefficient ( P oct) high quality quantitative structure-activity relationships (QSARs) could be derived. The aniline derivatives appeared to have a higher toxicity than the chlorobenzenes, which have a nonspecific (narcotic) mode of toxic action. For the narcotizing chemicals long-term 28-day ELS toxicity could be accurately predicted on the basis of short-term (7-day) experiments. For the anilines, which are assumed to exert their toxic effects after metabolic activation, the predictions for long-term toxicity on the basis of short-term experiments were found to be inaccurate. For these chemicals it seems necessary either to include detailed microscopic observations for sublethal effects or to prolong the exposure period for the free-feeding larval stages. Differences in sensitivity between B. rerio (this study) and fathead minnow, Pimephales promelas (data reported in literature) were found to be small. A QSAR for ELS-toxicity for 30 narcotizing chemicals for the two species was derived. Available data on the acute toxicity of guppy, Poecilia reticulata show that for chemicals with a nonspecific mode of toxic action the mean ‘acute/chronic’ toxicity ratio is approximately 15.