N-terminal variants of thyroid hormone receptor beta: differential function and potential contribution to syndrome of resistance to thyroid hormone.

The human syndrome of resistance to thyroid hormone (RTH) is associated with dominant mutations in the thyroid hormone receptor beta (TR beta) gene that generate mutant receptors with impaired binding for T3. Although the TR beta gene differentially expresses two N-terminal variant receptors, TR beta 1 and TR beta 2, functional analyses of RTH mutants have focused exclusively on TR beta 1. Since TR beta 2 is expressed in tissues that are malfunctional in RTH, the role of mutations in the context of TR beta 2 was examined. We compared the functional properties of corresponding RTH mutations in the common C-terminal domain of both TR beta 1 and TR beta 2. Wild type TR beta 1 and TR beta 2 bound similarly as homodimers and as heterodimers with retinoid X receptors to T3-responsive elements consisting of a direct repeat with 4-base pair spacing or an everted repeat. Homodimers, but not monomers or heterodimers, of both receptor subtypes were dissociated by the addition of T3. However, TR beta 2 formed at least 10-fold more stable homodimers than TR beta 1 on a palindromic repeat element, indicating that the N termini of TR beta 1 and TR beta 2 differentially influence dimerization on DNA. The RTH-like mutants of both TR beta 1 and TR beta 2 were equally insensitive to T3. They were defective in T3 binding but still bound DNA like their wild type counterparts except that the T3-dependent dissociation of homodimers from DNA was severely reduced. Wild type TR beta 1 and TR beta 2 mediated T3-inducible transactivation in cotransfection assays; this, however, was abolished in both mutants. TR beta 1 mediated more sensitive T3-dependent transcriptional suppression than TR beta 2 through the negative T3 response region of the TSH beta gene. Again, the mutation abolished T3-dependent suppression by both mutants. Furthermore, both mutants inhibited T3-inducible transcriptional activation by different wild type TR alpha and beta variants. These results indicate that both mutants have the potential to contribute to the pathogenesis of RTH and suggest that a reassessment of previous models of RTH is required to take into account the inhibitory activity of both TR beta 2 and TR beta 1 mutants.