The human progesterone receptor form B (hPR-B) was expressed in Saccharomyces cerevisiae together with a specific reporter plasmid. To understand the mechanism underlying antagonist ligand activity, libraries of hormone binding domain (HBD)-mutated hPR-B molecules were prepared. A mutant receptor was identified that had lost the ability to bind either progesterone or R5020; it could still bind RU486 and, surprisingly, fully activated transcription in the presence of this "antagonist" and other antiprogestins. When this receptor mutant was assayed in mammalian cells, RU486 again demonstrated agonistic activity. Sequence analysis indicated that the mutant phenotype was due to truncation of the carboxy (C)-terminal 42 aa. We conclude that amino acids in the extreme C-terminal region are required for the receptor to bind progesterone, while antagonists bind to a site located more N-terminal of the HBD. Our results suggest that the extreme C-terminal region of the receptor contains an inhibitory function that silences receptor transactivation in the absence of agonist and in the presence of antagonist.
PIP: The function of the hormone-binding domain of the human progesterone receptor was examined in a yeast cell system and in mammalian HeLa cells using mutant receptors, progesterone, the progesterone agonist R502, and the antagonists RU-486, Org31806, and Org31376. The hormone-binding domain, located on the carboxy terminal of the peptide, is known to initiate a conformational change in the receptor upon binding an agonist or antagonist, then shedding of associated proteins including the heat shock protein, dimerization of the receptor, and finally, binding to DNA, leading to transcription. Binding of a progesterone antagonist such as RU-486 elicits all these events except transcription. First the progesterone receptor was inserted in yeast with a plasmid, and a set of mutants were generated, using beta galactosidase as an indicator. A mutant progesterone receptor, U-P1, was selected for mechanistic studies, that binds and was activated by antagonists, but was inactive with progesterone. This receptor had a deletion at base 2636, resulting in a shift of reading frame so that a stop codon 36 nucleotides downstream caused truncation of 54 amino acids at the C-terminus and addition of 12 novel amino acids. Western blot analysis confirmed the expected molecular weight. The mutant receptor was active with RU-486, suggesting that the C-terminus may be responsible for poor transcription with RU-486, suggesting in normal receptors. 2 other truncated mutants were inactive with progesterone. These data suggested that the terminal 42 amino acids of the progesterone receptor are needed to bind progesterone, and that the antagonist is contacting different amino acids than the native receptor, possibly inducing a different conformational change. The activity of the UP-1 mutant was also confirmed in HeLa cells, with the chloramphenicol acetyltransferase reporter system. The results were interpreted to mean that progesterone agonists and antagonists contact at least some different amino acids in the hormone binding domain of the receptor, and that the conformational changes resulting from binding these agents are different. It appears that the C-terminus of the receptor contains an inhibitory domain which, when removed, turns antagonists into agonists.