Because progesterone antagonists are growth inhibitors, they are in Phase III clinical trials for the treatment of breast cancer. However, when cellular cAMP levels are elevated, some antiprogestins inappropriately activate transcription. We have proposed that hormone "resistance" may result from such unintended stimulation of breast cancer by antagonists. In transient expression systems, the two natural isoforms of human progesterone receptors (PR), B-receptors and truncated A-receptors, have dissimilar effects on agonist-mediated transcription. We show here that in the presence of 8-Br-cAMP, antiprogestin-occupied B-receptors but not A-receptors become transcriptional activators. Therefore, we developed new model systems to study each PR isoform independently in a breast cancer setting: (a) a stable PR-negative monoclonal subline (T47D-Y) of PR-positive T47D breast cancer cells was selected by flow cytometric PR screening. T47D-Y cells are PR-negative by immunoassays, by ligand binding assay, by growth resistance to progestins, by failure to bind a progesterone response element (PRE) in vitro, and by failure to transactivate PRE-regulated promoters; and (b) T47D-Y cells were stably transfected with expression vectors encoding one or the other PR isoform, and two monoclonal cell lines were selected that express either B-receptors (T47D-YB) or A-receptors (T47D-YA) at levels equal to those seen in natural T47D cells. The ectopically expressed receptors are properly phosphorylated, and like endogenously expressed receptors, they undergo ligand-dependent down-regulation. The expected B:B or A:A homodimers are present in cell extracts from each cell line, but A:B heterodimers are missing in both. In the presence of agonists, cAMP-dependent, transcriptional synergism of PRE-regulated promoters is seen in both cell lines. By contrast, in the presence of the antiprogestins RU486 or ZK112993, inappropriate transactivation occurs in YB cells but not in YA cells. The class of antiprogestins represented by ZK98299, which blocks PR binding to DNA, does not activate transcription in either cell line. We propose that these new cell lines are physiological models for the study of PR isoform-specific antiprogestin resistance in breast cancer.