Nuclear hormone receptors — ligand-dependent transcription factors that regulate many important physiological processes — have received much attention as therapeutic targets in recent years. The design of agonist drugs is facilitated by crystal structures of agonists bound to the ligand-binding domain of various nuclear hormone receptors. By contrast, the lack of antagonist-bound 'inactive' receptor structures — so far, only available for oestrogen receptor-α, the target of the breast-cancer drug tamoxifen — hinders the rational design of receptor antagonists. Writing in Endocrinology, Baxter et al. describe how agonist-bound structures can be used as the basis for the rational modification of agonists into antagonists, and the use of this approach for developing the first clear antagonist of the thyroid hormone receptor (TR), overactivity of which results in serious heart problems.

Previous structural studies had indicated that folding of the ligand-binding domain (LBD) to encapsulate the agonist might be crucial for the formation of the activated receptor surface. Noting that known antagonists of nuclear receptors seem to be analogues of agonists with a large extension group attached, the authors hypothesized that these antagonists bind in the same general orientation as agonists, but that the extension group acts as a physical barrier that prevents the receptor from attaining an active conformation. Consequently, they used the structures of agonist-bound LBDs of TR to design a potential antagonist by adding an extension group to an agonist analogue of the thyroid hormone triiodothyronine (T3). Binding assays and cell-culture experiments showed that the designed compound blocks binding of thyroid hormone and thyroid-hormone mediated responses.

So, this study shows that TR antagonist activity can be obtained, and is a first step towards evalulating the potential of TR blockade as a therapeutic approach. More generally, the structure-based approach that was used could perhaps be widely applicable to the design of nuclear-receptor antagonists, although as the authors note, there is a lot more to be learned about how this antagonism results.