The goal of this project is to understand the mechanism by which the progesterone receptor (PR) interacts with and remodels chromatin at target genes in vivo using the mouse mammary tumor virus (MMTV) promoter as a model system. In the non-activated state, this promoter has a chromatin structure repressive to transcription when it exists in cells as replicating chromatin. Upon binding of the liganded glucocorticoid receptor (GR), the promoter undergoes a chromatin remodeling event which is mechanistically involved in the activation of transcription. Our previous work has shown that the GR and PR have different requirements for chromatin remodeling at the MMTV promoter even though they bind to the same DNA sequences in the promoter. Our observations may form the basis for a mechanism by which the GR and PR control expression of distinct sets of target genes in vivo. This is particularly relevant in the mammary gland where the GR and PR can coexist in the same cell types. We have established that the PR exists in two distinct functional states in cultured mammary adenocarcinoma cells. In one state, it can neither remodel chromatin nor activate transcription at the MMTV promoter; thus its action may be restricted to target genes which do not require remodeling prior to activation. In addition, this form of the PR can be activated by other signal transduction pathways in a progestin-independent fashion. In the second functional state, the PR is able to remodel and activate the MMTV promoter in chromatin, but is refractory to ligand-independent activation. Thus, this form of the PR responds only to its ligand but would be able to activate target genes even in a repressive chromatin environment. We have also shown that the PR can be converted from the first state to the second by some form of cellular processing. This may represent a mechanism by which cells can restrict or expand the activity of the PR in vivo. To elucidate the biochemical basis for the two distinct functional states, we have developed an immunoprecipitation method to isolate the PR in its native forms and examine associated proteins and post-translational processing. This has been done in collaboration with Dr. David Smith (Mayo Clinic Scottsdale) who has provided us with reagents and expertise invaluble to the project. We have focused initially on interactions of the unliganded PR with chaperone proteins and immunophilins. We find that there are two classes of interactions based on nuclear binding properties of the unliganded PR. In the fraction of PR loosely-bound to the nucleus, the PR is largely monomeric and associated with hsp90 and the co-chaperone protein, p23. Large immunophilins are absent from this complex. In contrast, the fraction of PR tightly bound to the nucleus appears to exist in two forms. In one, it is associated with hsp90, p23, and the large immunophilin FKBP51. In the other it is dimerized, an event which normally occurs after ligand addition. The PR forms both classes of interactions in the two functional states, but the equilibrium between loosely-bound and tightly-bound PR is significantly different. We speculate that this altered balance indicates that PR/chaperone interactions can be regulated by means other than the presence of progestins. In addition, the balance of these interactions may have a significant impact on the function of the PR. Current efforts are concentrated on further characterization of the tightly-bound PR complexes because it is the predominant class when the PR in unable to productively interact with organized chromatin. Future studies include examination of the role of the large immunophilin and PR phosphorylation in setting the equilibrium of complexes.
