Office of Research and Development Publications

A High-Throughput Approach to Identify and Prioritize Putative Thyroid-Stimulating Hormone Receptor Agonists and Antagonists

Citation:

Shobair, M., M. Nelms, C. Deisenroth, G. Patlewicz, AND K. Paul-Friedman. A High-Throughput Approach to Identify and Prioritize Putative Thyroid-Stimulating Hormone Receptor Agonists and Antagonists. Presented at Society of Toxicology annual meeting, Baltimore, MD, March 10 - 14, 2019. https://doi.org/10.23645/epacomptox.7848404

Impact/Purpose:

Poster for presentation at Society of Toxicology annual meeting in March 2019. Due to human health effects resulting from altered thyroid hormone levels, it is important to evaluate whether environmental chemicals can disrupt thyroid function via TSHR-mediated signaling pathways. This study was done using a data-driven strategy, with categorical and quantitative features; identifying structure-activity relationships using clustering, and diversity maximization.

Description:

The thyroid-stimulating hormone receptor (TSHR) is a G protein-coupled receptor that signals through adenylate cyclase to increase intracellular 3’,5’- cyclic adenosine monophosphate (cAMP), resulting in increased thyroid hormone (TH) production in thyroid follicular cells. Due to human health effects resulting from altered TH levels, it is important to evaluate whether environmental chemicals can disrupt thyroid function via TSHR-mediated signaling pathways. As part of the Tox21 collaboration, HEK293-TSHR cells were used in a 1536-well assay format to demonstrate agonism or antagonism of the TSHR, using cAMP as a marker of TSHR activation. Homogeneous time-resolved fluorescence technology was used to quantify cAMP using a competitive immunoassay between native and dye-labeled cAMP. Out of the 7,872 tested chemicals, 6% agonist, 4% antagonist, and 0.6% agonist-antagonist hits were identified, for a total of ~10% putative active chemicals. Because receptor binding is highly specific, we hypothesized that many of the hits were false positives. Thus, we developed a novel prioritization scheme to select chemicals for screening in biologically-relevant follow-up assays. Chemicals (558/778 active chemicals) were clustered by structural similarity using ChemoTyper ToxPrint fingerprints. The priority score (within cluster and for non-clustered chemicals) was penalized for: i) activity in other ToxCast cAMP enzymatic assays, ii) promiscuity according to ToxCast total assay hit rate, iii) signal interference by autofluorescence, and iv) cytotoxicity. Highly-ranked agonist clusters contain phenols, organochlorine insecticides, and retinoids. Cytotoxicity contributed significantly to the antagonist priority rank, with as many as 68% of antagonists suspected to be cytotoxic in the active concentration ranges. The prioritization scheme has identified 69/778 active chemicals that are structurally diverse for additional testing. Using this scheme, secondary screening of identified priority chemicals will be combined with structural prioritization to create an integrated predictive tool for TSHR activity. This abstract does not necessarily reflect U.S. EPA policy.