An amphioxus orthologue of the estrogen receptor that does not bind estradiol: Insights into estrogen receptor evolution
1 Institut de Génomique Fonctionnelle de Lyon, Molecular Zoology team, Université de Lyon, Ecole Normale Supérieure de Lyon, Université Lyon 1, CNRS, INRA, Institut Fédératif 128 Biosciences Gerland Lyon Sud, France
2 Karolinska Institutet, Department of Biosciences and Nutrition, S-141 57 Huddinge, Sweden
3 CNRS, UMR 7628, Modèles en Biologie Cellulaire et Evolutive, Observatoire océanographique, F-66651, Banyuls/mer, France
BMC Evolutionary Biology 2008, 8:219 doi:10.1186/1471-2148-8-219Published: 25 July 2008
The origin of nuclear receptors (NRs) and the question whether the ancestral NR was a liganded or an unliganded transcription factor has been recently debated. To obtain insight into the evolution of the ligand binding ability of estrogen receptors (ER), we comparatively characterized the ER from the protochordate amphioxus (Branchiostoma floridae), and the ER from lamprey (Petromyzon marinus), a basal vertebrate.
Extensive phylogenetic studies as well as signature analysis allowed us to confirm that the amphioxus ER (amphiER) and the lamprey ER (lampER) belong to the ER group. LampER behaves as a "classical" vertebrate ER, as it binds to specific DNA Estrogen Responsive Elements (EREs), and is activated by estradiol (E2), the classical ER natural ligand. In contrast, we found that although amphiER binds EREs, it is unable to bind E2 and to activate transcription in response to E2. Among the 7 natural and synthetic ER ligands tested as well as a large repertoire of 14 cholesterol derivatives, only Bisphenol A (an endocrine disruptor with estrogenic activity) bound to amphiER, suggesting that a ligand binding pocket exists within the receptor. Parsimony analysis considering all available ER sequences suggest that the ancestral ER was not able to bind E2 and that this ability evolved specifically in the vertebrate lineage. This result does not support a previous analysis based on ancestral sequence reconstruction that proposed the ancestral steroid receptor to bind estradiol. We show that biased taxonomic sampling can alter the calculation of ancestral sequence and that the previous result might stem from a high proportion of vertebrate ERs in the dataset used to compute the ancestral sequence.
Taken together, our results highlight the importance of comparative experimental approaches vs ancestral reconstructions for the evolutionary study of endocrine systems: comparative analysis of extant ERs suggests that the ancestral ER did not bind estradiol and that it gained the ability to be regulated by estradiol specifically in the vertebrate lineage, before lamprey split.