Determinants for the repression of estrogen receptor transcriptional activity by ligand-occupied progestin receptors

There is considerable evidence for cross-talk between the estrogen and progestin signaling pathways, including examples of repression or attenuation of estrogen-stimulated endpoints by progestin receptor (PR) agonists and antagonists. We have previously described an experimental system for examining aspects of this cross-talk, namely the repression of estrogen receptor (ER) transcriptional activity by liganded PR (Kraus, W. L., Weis, K. E., Katzenellenbogen, B. S., Mol. Cell. Biol. 15 (1995) 1847-1857). Under promoter and cell type conditions where liganded PR was not a good activator of transcription, PR isoforms were shown to act as potent ligand-dependent repressors of ER transcriptional activity. In the current study, we have identified multiple determinants of this repression by systematically manipulating potentially important variables in this system (e.g. PR A:PR B ratio, sequence of the response elements, receptor structure, and ligand type). Alterations in several of these parameters had profound effects on the ability of PR to repress the activity of ER. Decreases in the PR A:PR B ratio and changes in the sequence of the progestin response element in the reporter gene construct abolished the repressive action of agonist-occupied PR A on ER transcriptional activity. In addition, point or deletion mutations in the amino-terminal A/B region of ER, including a triple point mutation which eliminates phosphorylation sites previously shown to be important in the activity of the receptor, made the ER more sensitive to the repressive actions of liganded PR. The PR ligands that promoted the most potent repression of ER activity were those with 11β phenyl substitutions, suggesting that the phenyl moiety in the 11β position is the important structural feature leading to strong repression. Interestingly, changes in the structure of the ER ligand and the sequence of the estrogen response element did not influence the magnitude of repression by PR. The fact that alterations in these check points along the estrogen signaling pathway had little or no effect on the magnitude of repression suggests that liganded PR interferes with the ability of ER to interact productively with the transcriptional machinery; in other words, PR-mediated repression occurs downstream of the events leading to the ligand-dependent conversion of ER to a transcriptionally active form. Our results indicate that a number of parameters which are naturally varied in vivo, such as the sequence of PR DNA binding sites and the PR A:PR B ratio, can dramatically alter the repression of ER activity by liganded PR, and may explain the differential affects of progestin-occupied PR on the expression of different estrogen regulated genes.