Title: Classification of Phthalates According to Their (Q)SAR Predicted Acute Toxicity to Fish: A Case Study.
Publication Year: 2007
JRC N°: JRC35262
Other Identifiers: EUR 22623 EN
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC35262
Type: EUR - Scientific and Technical Research Reports
Abstract: This report presents the preliminary results from a (Q)SAR investigation of the acute toxicity to fish (fathead minnow) for a dataset of phthalate esters. A chemical set of 341 phthalates was compiled by using different searching engines. Their acute toxicity to fathead minnow was calculated with the ECOSAR and TOPKAT software. A good correlation between the predictions from the two programs was established (r2 = 0.81). The chemicals were classified initially into four groups on a basis of their predicted by ECOSAR LC50 values: 1) no reasons for concern (LC50 > 100 mg/L), 2) harmful (10 mg/L < LC50 < 100 mg/L), 3) toxic (1 mg/L < LC50 < 10 mg/L), 4) very toxic LC50 < 1 mg/L). This prediction effort resulted in classification of the vast majority of the phthalates in the “very toxic” group. The reason for this result is that ECOSAR uses linear relationships with the octanol-water partition coefficient (log Kow) for chemicals with log Kow < 5 (warning is issued for chemicals with log Kow > 5). The predictions from TOPKAT (only predictions within the optimum prediction space were considered) correlated relatively well with those from ECOSAR. There were many high molecular weight phthalate esters in the chemical series, which appeared clearly outside the applicability domain of the ECOSAR models. This fact, as well as the understanding that beyond certain limits of hydrophobicity the toxicity of the organic chemicals decreases as a result of reduced bioconcentration, motivated the development of an algorithm for refinement of acute toxicity predictions of the phthalate esters using the bilinear relationship with log Kow. In addition, water solubility limits were considered. Long-term toxicity studies were not considered in this study. Transformation (e.g. biodegradability) of the parent compounds was not considered either. This could potentially be important as, theoretically, the transformation of very hydrophobic chemicals (log Kow > 7) or extremely hydrophobic chemicals (log Kow > 8.0) into more hydrophilic degradation/transformation products may increase the acute toxicity to fish. This case study provides an illustration of how (Q)SAR methods can be used in the development of chemical categories and how (Q)SAR results can be used to perform an initial screening in support of classification and labelling. The results are discussed and interpreted with a view of what constitutes a category, how it can be defined and described, what are its boundaries, and the need to define subcategories that might be useful for deciding on the level of acute toxicological hazard associated with different structural modifications. Due to the preliminary nature of the (Q)SAR models, the results of this study should be regarded as an illustration of the applicability of (Q)SAR methods. The actual model results and rule-based classification scheme will need validation and refinement before they could be considered for regulatory use.
JRC Institute:Institute for Health and Consumer Protection

Files in This Item:
File Description SizeFormat 
5262 - Phthalates_12Jan07.pdf988.35 kBAdobe PDFView/Open

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.