The thyroid hormone system is an important and complex endocrine system. The thyroid system regulates several important physiological processes such as energy metabolism, brain growth and development, and reproduction. The system is conservative across species. Homeostasis of the system could be disrupted by certain industrial chemicals - thyroid disruption chemicals (TDCs). TDCs have been widely identified in aquatic environments and household dust. Exposures to TDCs pose potential threats to human health and ecological system. Considering the relatively high cost of wet lab experiment, it is thus critical to develop computational models for in silico identification of TDCs from industrial chemicals and environmental pollutants.
In our previous studies (SNIC 2014/2-30 & SNIC 2015/7-51 & SNIC 2016/3-68), we have identified 13 novel TDCs targeting the human transthyretin and thyroid receptor using virtual screening tools developed based on molecular docking and molecular dynamics simulations.
Our studies results have been published in ACS Environmental Science & Technology and ACS Chemical Research in Toxicology. However, we also found that the Piscine species e.g. zebrafish and Sear bream are not suitable animal models and underestimate the thyroid toxicities of TDCs due to their large variations with the human thyroid targets.
In this project, we are aimed to (1) investigate the interspecies variations on thyroid targets between human and frogs using bioinformatic tools; (2) develop virtual screening protocol to identify TDCs using human thyroid target structures and tested their thyroid toxicities in both human cell bioassay and frog cell bioassay to reveal the differences.
Frogs are found in oceans all over the world. Several classical TDCs, such as bisphenol A and OH-PCBs have been detected in frog at a relative high level. More importantly, metamorphosis process in frog is mainly regulated by thyroid hormone system. TDC-induced metamorphosis failures will be a clear and obvious readout that can be used to evaluate the thyroid toxicities of environmental pollutants in a high throughput manner.
We also would like to improve virtual screening performance of computational models by using using two strategies. Firstly, usage of physical based re-scoring functions, e.g. LIE and MM-GBSA. Secondly, QM calculations for truncated molecular system of interest, e.g. the ligand binding pocket.
The result of this study can be used to suggest potential TDCs for further toxicological investigations. It also gives a better understanding on the molecular mechanism-of-action of chemical-induced thyroid disruption. The diverse compounds identified help us to summarize structure-activity relationships of TDCs.