The crystal structure of the ligand binding domain (LBD) of the estrogen-related receptor 3 (ERR3) complexed with a steroid receptor coactivator-1 (SRC-1) peptide reveals a transcriptionally active conformation in absence of any ligand. The structure explains why estradiol does not bind ERRs with significant affinity. Docking of the previously reported ERR antagonists, diethylstilbestrol and 4-hydroxytamoxifen, requires structural rearrangements enlarging the ligand binding pocket that can only be accommodated with an antagonist LBD conformation. Mutant receptors in which the ligand binding cavity is filled up by bulkier side chains still interact with SRC-1 in vitro and are transcriptionally active in vivo, but are no longer efficiently inactivated by diethylstilbestrol or 4-hydroxytamoxifen. These results provide structural and functional evidence for ligand-independent transcriptional activation by ERR3.

Retinoids regulate gene expression through binding to the nuclear retinoic acid receptors (RARs) and retinoid X receptors (RXRs). In contrast, no ligands for the retinoic acid receptor-related orphan receptors beta and gamma (ROR beta and gamma) have been identified, yet structural data and structure-function analyses indicate that ROR beta is a ligand-regulated nuclear receptor. Using nondenaturing mass spectrometry and scintillation proximity assays we found that all-trans retinoic acid (ATRA) and several retinoids bind to the ROR beta ligand-binding domain (LBD). The crystal structures of the complex with ATRA and with the synthetic analog ALRT 1550 reveal the binding modes of these ligands. ATRA and related retinoids inhibit ROR beta but not ROR alpha transcriptional activity suggesting that high-affinity, subtype-specific ligands could be designed for the identification of ROR beta target genes. Our results identify ROR beta as a retinoid-regulated nuclear receptor, providing a novel pathway for retinoid action.

Retinoic acid receptors (RARs) are ligand-dependent transcription factors that control a plethora of physiological processes. RARs exert their functions by regulating gene networks controlling cell growth, differentiation, survival, and death. Uncovering the molecular details by which synthetic ligands direct specificity and functionality of nuclear receptors is key to rational drug development. Here we define the molecular basis for (E)-4-[2-[5,6-Dihydro-5,5-dimethyl-8-(2-phenylethynyl)naphthalen-2-yl]ethen-1-yl]benzoic acid (BMS204,493) acting as the inverse pan-RAR agonist and define 4-[5,6-Dihydro-5,5-dimethyl-8-(quinolin-3-yl)naphthalen-2-carboxamido]benzoic acid (BMS195,614) as the neutral RARalpha-selective antagonist. We reveal the details of the differential coregulator interactions imposed on the receptor by the ligands and show that the anchoring of H12 is fundamentally distinct in the presence of the two ligands, thus accounting for the observed effects on coactivator and corepressor interactions. These ligands will facilitate studies on the role of the constitutive activity of RARs, particularly of the tumor suppressor RARbeta, whose specific functions relative to other RARs have remained elusive.

No abstract