The function of retinoid X receptors on negative thyroid hormone response elements

Retinoid X receptors (RXRs) form heterodimers with thyroid hormone receptors (TRs). RXRs increase DNA binding affinity of TRs and T3-mediated transactivation on positive T3 response elements (TREs). However, the role of RXRs on negative TREs, and the relation of RXRs to the dominant negative effect of mutant TRs, are not defined. To clarify the function of RXRs on negative TREs, we performed transient cotransfection studies using the rat glycoprotein hormone alpha promoter fused to luciferase gene (alphaLuc), and human TRH promoter fused to luciferase gene (TRH-Luc) as reporters. We found that the JEG-3 cell-alphaLuc system was very sensitive to TR regulation. Using TRbeta1 wild-type (WT) expression vector, 6.2 ng/well (170 ng/10 cm dish), and 0.2 ng/well (11 ng/10 cm dish) caused maximal, and half maximal, inhibition of Luc activities in the presence of 1 nM T3. A T3 dose dependent inhibition study was also performed. From these studies, we determined that the appropriate conditions in which to study alphaLuc transactivation, in a linear portion of the dose response curve, was using 0.8 ng/well TRbeta1 expression vector and 0.1 nM T3. Under these conditions, TRbeta1 mutant R316H (GH), but not G345R (Mf), showed a weak dominant negative effect at a 1:1 ratio in the presence of 0.1 nM T3 although neither mutant had detectable T3 binding affinity. Moreover this dominant negative effect of R316H on the alphaLuc reporter was enhanced in the presence of RXRgamma. Mutant G345R showed a stronger dominant negative effect than did R316H when using a double palindromic TRE fused to herpes simplex thymidine kinase-Luc reporter as a positive TRE. These results conform to the clinical features of R316H which is associated with apparent pituitary resistance of thyroid hormone (PRTH). Mutant R316H also showed a weak dominant negative effect with TRH-Luc at a 1:1 ratio in the absence or presence of RXRgamma. However RXRgamma did not enhance the dominant negative effect as it did using alphaLuc reporter gene. Electrophoretic gel mobility shift assay (EMSA) showed that RXR alpha augmented the DNA binding affinity of wild type and R316H TRs as heterodimers on the previously reported negative TREs of glycoprotein hormone alpha promoter, suggesting that RXR does not produce its response by removing TRs from these TREs. RXR alpha augmented DNA binding affinity of TRbeta1WT, and R316H showed a weaker heterodimer band than did the wild type in EMSA. Using the TRH-Luc reporter, basal activity was increased by wild type TRbeta1. However a TRbeta1 DNA binding domain mutant, (C127S) which can not bind to DNA, did not increase the basal activity. This indicates that DNA binding of the TR is required for increasing basal activity of TRH promoter. These results indicate that (1) RXR-TR heterodimers play a role in basal transactivation and T3 suppression of negatively regulated genes, and (2) RXRs increase the dominant negative effect of some mutant TRs on specific negative TREs. (3) This effect occurs without removing TRs from the TRE. (4) The differential dominant negative effect of mutant R316H (negative TRE > positive TRE) may explain, at least in part, the presentation of R316H as PRTH. (5) Augmentation of basal activity by wild type TRs on a negative TRE requires DNA binding.