Correlation of the toxicity of organic compounds to tadpoles using the Abraham model.

The narcotic and lethal concentrations of organic compounds have been compiled for several tadpole species (Rana temporaria, Rana pipiens, Rana japonica, Xenopus laevis and Rana brevipoda porosa). The narcotic and lethal concentrations have been correlated using the Abraham solvation parameter model to yield an equation that can be used to predict the narcotic concentrations of additional nonpolar and polar narcotic compounds to R. temporaria, and a more general correlation that should be applicable to different species of tadpoles. The more general equation is based on 240 experimental data points. A training set of 123 compounds could be fitted with the Abraham solvation parameters with R(2)=0.931 and S.D.=0.343 log units. The training equation predicted the test set of 122 values with AE=-0.022 log units, S.D.=0.300 log units and an average absolute error, AAE, of 0.227 log units. The structural features that are important in narcosis of tadpoles have been examined; it is concluded that hydrogen bond basicity reduces narcotic activity of compounds and that compound size increases narcotic activity. The solvation parameter model enables narcosis of tadpoles to be compared to various other biological processes and to physicochemical processes that might be used as models for narcosis.

Correlating toxicities of organic compounds to select protozoa using the Abraham model.

The Abraham solvation parameter model is used to construct mathematical correlations for describing the nonspecific toxicity of organic compounds to three protozoas (Entosiphon sulcantum, Uronema parduczi and Chilomonas paramecium). The derived mathematical correlations describe the observed published toxicity data to within an overall average standard deviation of approximately 0.35 log units. The correlations can be used to estimate aquatic toxicities of organic chemicals to the three aquatic organisms studied, and to help in identifying compounds whose toxic mode of action might involve chemical specific reactivity, rather than nonpolar or polar narcosis. A principal component analysis of the correlation equations found in this work shows that no water-solvent system we have investigated is a good model for nonspecific aquatic toxicity towards the three protozoas. Furthermore, correlation equations for nonspecific aqueous toxicity towards various biological systems, that we have found in this work and in previous studies, cover such a wide range that no single water-solvent system could ever be a good model for all the biological systems.