Development of QSARs for the toxicity of chlorobenzenes to the soil dwelling springtail Folsomia candida

Abstract

To meet the goals of Registration, Evaluation, Authorisation, and Restriction of Chemicals (REACH) as formulated by the European Commission, fast and resource-effective tools are needed to predict the toxicity of compounds in the environment. We developed quantitative structure-activity relationships (QSARs) for the toxicity of nine chlorinated benzenes to the soil-dwelling collembolan Folsomia candida in natural LUFA2.2 (Landwirtschaftliche Untersuchungs und Forschungsanstalt [LUFA]) standard soil and in Organisation for Economic Co-operation and Development artificial soil. Toxicity endpoints used were the effect concentrations causing 10% (EC10) and 50% (EC50) reduction in the reproduction of the test organism over 28 d, while lethal effects on survival (LC50) were used for comparisons with earlier studies. Chlorobenzene toxicity was based on concentrations in interstitial water as estimated using nominal concentrations in soil and literature soil-water partition coefficients. Additionally, for LUFA2.2 soil the estimated concentrations in interstitial water were experimentally determined by solid-phase microextraction measurements. Measured and estimated concentrations showed the same general trend, but significant differences were observed. With the exception of hexachlorobenzene, estimated EC10 and EC50 values were all negatively correlated with their logK(ow) and QSARs were developed. However, no correlation for the LC50 could be derived and 1,2,4,5-tetrachlorobenzene and hexachlorobenzene had no effect on adult survival at all. The derived QSARs may contribute to the development of better ecotoxicity-based models serving the REACH program.

Toxicity in relation to mode of action for the nematode Caenorhabditis elegans: Acute‐to‐chronic ratios and quantitative structure–activity relationships

“…For a different group of nonpolar narcotic chemicals, the PAHs, the identical relationship (slope -0.92) of acute and chronic toxicity in C. elegans with the log K OW was found [43]. Indeed, for nonpolar narcotic chemicals, a value for the slope close to unity has often been observed for acute and chronic toxicity endpoints in a number of species [2,6,9,11,12,[19][20][21]49]. For the narcotic chemicals tested with C. elegans, the fact that for all toxicity measures their relation to the log K OW value is the same (the same slope for all toxicity endpoints measured) implies that the ACR of the tested substances is independent of their log K OW .…”

“…In addition, the ECOSAR program from the US Environmental Protection Agency provides QSARs for specifically acting compounds , although this option is lacking in the European Union . For larger sized sediment and soil animals (earthworms, springtails, chironomids) , preliminary QSARs exist for small subsets of selected chemicals, but none are available for small benthic/soil animals. Following the approaches for organic compounds, there have also been attempts to relate metal toxicity to ion characteristics .…”

Determining High-Quality Critical Body Residues for Multiple Species and Chemicals by Applying Improved Experimental Design and Data Interpretation Concepts

“…Experiments were designed so that the mass of the chemical sorbed to the surrogate biological phase was always more than 30% of the total mass in the system . Polyacrylate-water partition coefficients required for the derivation of aqueous concentrations had either been determined previously or were measured as described in ref . Newly determined values are listed in Table S5.…”

“…Yet, CBRs for polar and nonpolar compounds, which differ significantly on a wet weight basis, − seem to converge when expressed as estimated concentrations in membrane lipids. Most probably, this can be explained by a relatively high affinity of polar compounds for membrane lipids as compared to that of nonpolar compounds. − Differences in whole body-based CBR values for a specific chemical in different organisms may be related to differences in the composition of the organisms, i.e., the ratio of polar membrane lipids to nonpolar storage lipids. In addition, other biological compartments such as proteins may act as sorption domains and may thus complicate data interpretation. − As such, these compartments should preferably be included in data analysis.…”