Endocrine Abstracts

The glucocorticoid receptor antagonist, RU486, binds to the glucocorticoid receptor (GR) with high affinity, and allows both nuclear localisation and also DNA binding. However, the DNA bound GR does not activate target gene transcription. This suggests that the ligand-binding domain of the GR is required to interpret agonist and antagonist ligands, and transmit those signals to the basal transcriptional machinery. Adaptor proteins play an important role in transducing the effect of DNA bound nuclear receptors on gene transcription; these include the co-activators SRC1 and GRIP1, and the co-repressor NCoR.

Mammalian two hybrid and GST pulldowns were used to show that RU486 failed to recruit of SRC-1 to the GR ligand binding domain, whereas dexamethasone efficiently promoted recruitment. However, the N terminal of the GR contains a ligand independent activation function (AF-1), and the absence of SRC-1 recruitment to the C terminal AF-2 might be insufficient to explain why RU486 is an antagonist. No repressor for the GR has been identified.

In a mammalian two-hybrid assay 100nM RU486 induced a four-fold induction of NcoR recruitment to the GR ligand-binding domain. Dexamethasone did not allow any induction, and there was no interaction between NcoR and GR in the absence of ligand. Previously we have defined tyrosine 735 within the GR ligand-binding domain as important for co-activator recruitment in response to agonist binding. Substitution to serine abolished interaction between the GR and SRC-1. Interestingly, the tyrosine to serine mutation enhanced recruitment of NcoR to the GR in response to RU486 from 4 fold to more than 8 fold. This suggests that the GR surfaces for interaction with SRC-1 and NcoR are separate and to an extent mutually exclusive. These data suggest that RU486 works as an antagonist both by failing to allow co-activator recruitment to the GR, but also by recruiting a powerful transcriptional repressor, NCoR, which is likely to explain the inhibition of AF-1 function.