Evolution of Nuclear Receptors and Ligand Signaling: Toward a Soft Key–Lock Model?

Abstract Nuclear receptors (NRs) are a family of ligand-regulated transcription factors that modulate a wide variety of physiological functions in a ligand-dependent manner. The first NRs were discovered as receptors of well-known hormones such as 17β-estradiol, corticosteroids, or thyroid hormones. In these cases a direct activation of the receptor transcriptional activity by a very specific ligand, with nanomolar affinity, was demonstrated, providing a strong conceptual framework to understand the mechanism of action of these hormones. However, the discovery that some NRs are able to bind different ligands with micromolar affinity was a first sign that the univocal relationship between a specific receptor (e.g., TR) and a specific ligand (e.g., thyroid hormone) should not be generalized to the whole family. These discussions about the nature of NR ligands have been reinforced by the study of the hormone/receptor couple evolution. Indeed when the ligand is not a protein but a small molecule derived from a biochemical pathway, a simple coevolution mechanism between the ligand and the receptor cannot operate. We and others have recently shown that the ligands acting for a given NR early on during evolution were often different from the classical mammalian ligands. This suggests that the NR/ligand evolutionary relationship is more dynamic than anticipated and that the univocal relationship between a receptor and a specific molecule may be an oversimplification. Moreover, classical NRs can have different ligands acting in a tissue-specific fashion with significant impact on their function. This also suggests that we may have to reevaluate the pharmacology of the ligand/receptor couple.