The origin of the ligand-binding ability of nuclear receptors is still a matter of discussion. Current opposing models are the early evolution of an ancestral receptor that would bind a specific ligand with high affinity and the early evolution of an ancestral orphan that was a constitutive transcription factor. Here we review the arguments in favour or against these two hypotheses, and we discuss an alternative possibility that the ancestor was a ligand sensor, which would be able to explain the apparently contradictory data generated in previous models for the evolution of ligand binding in nuclear receptors.
Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.