Raphaëlle Lesage (1), Pavel Balazki (1), Stephanie Melching-Kollmuss (2), Eric Fabian (2), Brandy Williamson Riffle (2), Stefan Stinchcombe (2)
(1) ESQlabs GmbH, Saterland, Germany, (2) BASF SE, Limburgerhof, Germany
Objectives: The current practice of medicines or chemicals toxicity assessment still relies on extensive preclinical studies. The assessment of potential endocrine effects, such as thyroid hormones (TH) disturbance, is hampered by disparities observed between human and animal responses, notably in species like rat (1–4) . This underscores the need to explore the mechanism leading to species differences. The purpose of this study is to develop a quantitative systems toxicology (QST) platform describing the regulation of thyroid hormones (TH) in rats and humans to assess species-specific effects of TH disrupting chemicals. The platform should include mechanistic description of processes with known species differences, such as binding to plasma proteins, or elimination pathways of the TH. We also aim at testing the platform via modelling Phenobarbital, a compound with known effects on the TH axis (5).
Methods: A published physiologically based kinetic (PBK) TH model was leveraged and adapted with new mechanisms to account for species-specific differences with the OSP suite (PK-Sim and Mobi) version 11.2 (6). The initial model included the pituitary gland synthesis of thyroid stimulating hormone (TSH), TH synthesis (T3 & T4) and induction by TSH as well as feedback loops. It was adapted to account for mechanisms such as the UDPGT-mediated clearance of TH, TSH circadian rhythm, and mechanistic binding of TH to important plasma transport proteins, namely Albumin, Transthyretin (TTR), and Thyroid Binding Globulin (TBG). A common model structure was established and parametrized for adult human and rat. To test the platform, a PBK model of Phenobarbital was developed, a compound known for disrupting the thyroid hormone axis via induction of UDPGT. Literature and legacy data were used to assess the model’s performance and predictive power.
Results: Albumin, TTR, and TBG, the 3 TH plasma binding proteins, were successfully implemented as endogenous large molecules and their species-specific turnover could be reproduced with parameters falling within the ranges of literature-reported values. Notably, TBG is not present in rat, while it greatly contributes to TH binding in human. Similarly, species-specific turnover of TH and the free fractions in plasma could be reproduced based on literature data. T3 is eliminated through glucuronidation by UGT enzymes (30% in rats) and residual clearance by deiodination in all tissues. In humans, UGT does not play a significant role in T3 clearance. T4 is converted to T3 by iodothyronine deiodinase (30%) and to reverse T3 (30%). It also undergoes glucuronidation by UGT (25%) and sulfation, the latter implemented as residual liver clearance. Modeling Phenobarbital-mediated induction of UDPGTs confirmed the capacity of the model to predict the differential impact of a chemical on TH levels between rats and humans.
Conclusions: In summary, the model captures dynamic binding of TH to plasma proteins and reproduces published data on intravenous TH administrations and feedback loops governing TH/TSH regulation. By accounting for species-specific binding mechanisms and hormones degradation pathways, this model is able to predict species-specific thyroid toxicity of compounds. This is exemplified by our Phenobarbital case study. Overall, this PBK study demonstrated mechanistically that species-specific expression of plasma proteins and binding affinities can explain essential differences in TH disruption after chemical or drug exposure. The QST TH model presents as a modular platform, which may be coupled to any PB(P)K model to evaluate the thyroid toxicity potential of various compounds and look into parameter distributions to study variations across populations for the safe development of drugs and chemicals.
References:
[1] Wiemann C, Melching-Kollmuss S, Hambruch N, Wiss L, Stauber F, Richert L. Boscalid shows increased thyroxin-glucuronidation in rat but not in human hepatocytes in vitro. J of Applied Toxicology. 2023 Jun;43(6):828–44.
[2]Foster JR, Tinwell H, Melching-Kollmuss S. A review of species differences in the control of, and response to, chemical-induced thyroid hormone perturbations leading to thyroid cancer. Arch Toxicol. 2021 Mar 1;95(3):807–36.
[3] Hernández AF, Bennekou SH, Hart A, Mohimont L, Wolterink G. Mechanisms underlying disruptive effects of pesticides on the thyroid function. Current Opinion in Toxicology. 2020 Feb 1;19:34–41.
[4] Wu KM, Farrelly JG. Preclinical Development of New Drugs that Enhance Thyroid Hormone Metabolism and Clearance: Inadequacy of Using Rats as an Animal Model for Predicting Human Risks in an IND and NDA: American Journal of Therapeutics. 2006 Mar;13(2):141–4.
[5] Plummer S, Beaumont B, Elcombe M, Wallace S, Wright J, Mcinnes EF, et al. Species differences in phenobarbital-mediated UGT gene induction in rat and human liver microtissues. Toxicology Reports. 2021;8:155–61.
[6] PB-QST model of thyroid hormones – Release version 1.0 · Open-Systems-Pharmacology [Internet]. [cited 2023 Sep 22]. Available from: https://github.com/Open-Systems-Pharmacology/Thyroid-Hormones-PB-QSP-Model/releases/tag/v1.0
Reference: PAGE 32 (2024) Abstr 11172 [www.page-meeting.org/?abstract=11172]
Poster: Drug/Disease Modelling - Other Topics