SETAC Europe 35th Annual Meeting

In Silico Tools for Endocrine Disruption Assessment: A Focus on Human Transthyretin Disruptors

Abstract: Endocrine Disrupting Chemicals (EDCs) are a structurally heterogeneous group of substances of widespread use and application, whose identification and assessment has become prioritary due to their ability to negatively affect the endocrine system in living organims. In this context, the use of in silico approaches, such as Quantitative Structure-Activity Relationships (QSARs), is suggested to fill data gaps and to support the identification of substances with endocrine activity.
This work aims to develop new QSAR models for the prediction of the binding affinity of exogenous chemicals with human transthyretin (hTTR), identified as a relevant molecular initiating event leading to thyroid hormone (TH) system dysfunctions. The modelled endpoint was the logarithm of the Relative competitive Potency (RP), which reflects the ability of a compound to displace the TH thyroxine (T4) from the hTTR. The here proposed models were developed using three new datasets that include, to our knowledge, all the currently available experimental values of RP published in the literature and measured with three different in vitro assays: the radiolabeled [125I]-T4 binding assay (RLBA), the 8-anilino-1-naphtalenesulfonic acid (ANSA) based binding assay, and the fluorescence conjugate isothiocyanate (FITC)-T4 based binding assay. Each dataset is composed of data measured by the same in vitro assay. Theoretical molecular descriptors were generated from harmonized SMILES (Simplified Molecular Input Line Entry System) and were used as variables in QSAR modeling. QSARs development followed the OECD guidelines, in order to guarantee for the internal statistical reliability and the external predictivity of the models when applied to new chemicals. The mechanistic interpretation of the selected molecular descriptors provided insights on the structural features that promote the displacement of T4 from hTTR. Due to the presence of different chemical structures in the three datasets, each QSAR was applied through the QSAR-ME Profiler software to fill missing values in the other two datasets. Compounds with reliable predictions were ranked by PCA analysis according to their experimental or predicted RPs, in order to discriminate stronger and weaker hTTR binders.
The here proposed QSARs can be applied to support the identification of chemicals with potential TH system-disrupting activity, which could represent a severe threat for human health and wildlife.

Disclaimer/ Disclosure (optional):

Author(s):

Marco Evangelista QSAR Research Unit in Environmental Chemistry and Ecotoxicology, Department of Theoretical and Applied Sciences (DiSTA), University of Insubria Italy This Author Is the Presenter

Nicola Chirico QSAR Research Unit in Environmental Chemistry and Ecotoxicology, Department of Theoretical and Applied Sciences (DiSTA), University of Insubria Italy

Ester Papa QSAR Research Unit in Environmental Chemistry and Ecotoxicology, Department of Theoretical and Applied Sciences (DiSTA), University of Insubria Italy