Evaluation of gene expression endpoints in the context of a Xenopus laevis metamorphosis-based bioassay to detect thyroid hormone disruptors.

Thyroid hormones (TH) are important in growth, development and the maintenance of proper cellular metabolism in vertebrates. Amphibian metamorphosis is completely dependent on TH and forms the basis of a screen for thyroid axis disrupting chemicals that currently relies on external morphological endpoints and changes in thyroid gland histology. The requirement for TH-dependent gene expression makes it possible to augment this screen through the addition of molecular endpoints. In order to do this, gene selection, choice of sampling time, tissue sensitivity, and their relationship to morphological change must all be considered. We exposed stage 54 Xenopus laevis tadpoles to a concentration series of the THs, thyroxine (T4) and 3,5,3'-triiodothyronine (T3), and three known TH antagonists, methimazole, propylthiouracil (PTU), and perchlorate. The agonists significantly accelerated metamorphosis as defined by developmental stage attained after 14 days. In contrast, the TH antagonists significantly delayed metamorphosis at 14 days and caused an increase in thyroid gland size at day 8. We assessed the changes in steady-state mRNA levels of thyroid hormone receptor alpha- and beta-isoforms and the basic transcription element binding (BTEB) protein by quantitative real-time polymerase chain reaction. Three tissues (brain, tail and hindlimb) were analyzed at 24, 48 and 96 h and we found that TH receptor, TRbeta, and BTEB were the most sensitive gene transcripts for the TH agonists, whereas only TRalpha displayed significant changes upon antagonist exposure. We detected differences in tissue-specific responses between the two agonists. We matched the concentrations of T3 and T4 that elicited similar biological responses at 14 days and compared the induction of gene expression. At 96 h, the TRbeta and BTEB expression response to T3 and T4 was similar in the tail. In contrast, T3 elicited no concentration-dependent changes in TRbeta and BTEB expression in the brain, whereas T4 elevated their expression. The tail showed the highest correlation between TH concentration and morphological outcome whereas the brain was the most sensitive to antagonist treatment. Only methimazole and perchlorate showed significant changes in TRalpha gene expression in the brain whereas PTU did not suggesting differences in cellular mechanisms of action. The greatest effect on gene expression occurred within 48 h with many of the hormone-dependent changes disappearing by 96 h. This study accentuates the need to examine multiple tissues and provides critical information required for optimization of exposure regimens and endpoint assessments that focus on the detection of disruption in TH-regulatory systems.