You are here:
Integrating in silico and in vitro data to identify putative thyrotropin-releasing hormone receptor ligands
Citation:
Shobair, Mahmoud A., D. Chang, Chris Grulke, K. Paul-Friedman, AND A. Richard. Integrating in silico and in vitro data to identify putative thyrotropin-releasing hormone receptor ligands. ACS Fall meeting, acs.org, San Francisco, California, August 16 - 20, 2020. https://doi.org/10.23645/epacomptox.12822230
Impact/Purpose:
Application of high-throughput screening (HTS) to toxicology faces the challenge of relating HTS data to molecular initiating events (MIEs) and adverse outcome pathways (AOP) even when the relationship between the HTS output and MIE is indirect. Thyrotropin-Release Hormone Receptor (TRHR) activation is a MIE in the AOP network for thyroid hormone (TH) disruption, but the ability of environmental chemicals to perturb TRHR signaling is unknown. A Tox21 HTS biochemical assay is available, but no orthogonal or confirmatory Tox21 or ToxCast assays can be used to differentiate TRHR true positive responses. To test the hypothesis that environmentally-relevant chemicals can directly interact with TRHR, we developed an in silico cheminformatic workflow to identify chemicals predicted to interfere with TRH binding.
Description:
Application of high-throughput screening (HTS) to toxicology, as in the cross-federal Tox21 partnership or EPA’s ToxCast program, faces the challenge of relating HTS data to molecular initiating events (MIEs) and adverse outcome pathways (AOP) even when the relationship between the HTS output and MIE is indirect. Thyrotropin-Release Hormone Receptor (TRHR) activation is a MIE in the AOP network for thyroid hormone (TH) disruption, but the ability of environmental chemicals to perturb TRHR signaling is unknown. A Tox21 HTS biochemical assay is available, but no orthogonal or confirmatory Tox21 or ToxCast assays can be used to differentiate TRHR true positive responses. To test the hypothesis that environmentally-relevant chemicals can directly interact with TRHR, we developed an in silico cheminformatic workflow to identify chemicals predicted to interfere with TRH binding. The tiered approach has the following steps: 1) identify strong signals using structure-based chemotype enrichment analysis; 2) eliminate noise from cytotoxicity or assay interference; 3) identify other sources of interference such as perturbation in calcium signaling; and 4) prioritize potential TRHR modulators by likelihood of binding. For in silico binding model training, we curated a reference structure-activity dataset from literature reports of chemical concentrations required to displace binding of radiolabeled TRH in competitive binding assays. The dataset is balanced between actives and inactives and mainly contains TRH derivatives and psychoactive drugs. We used 3D pharmacophore modeling to discriminate between binders and non-binders. Candidate binders were visualized in a 3D homology model. Preliminary results suggest that < 5% of the actives in the Tox21 biochemical assay contain TRH-like binding features, likely due to the conservation of the TRHR ligand binding site and the structural similarity of these substances to known TRHR modulators, benzodiazepines and neuropeptides. Our tiered in silico workflow increases the value of in vitro data for chemical prioritization by grounding the results in physical determinants that are directly related to experimental target binding results. Our findings suggest that combining structure-based methods and data enrichment analysis can be applied more broadly to increase confidence in HTS results and inform further screening and hazard characterization. This abstract does not reflect US EPA policy.
URLs/Downloads:
DOI: Integrating in silico and in vitro data to identify putative thyrotropin-releasing hormone receptor ligands
ACS_TALK_3408391_SLIDES_SHOBAIR.PDF (PDF, NA pp, 1418.285 KB, about PDF)