Heat shock factor 1 (HSF1), the transcriptional activator of the heat shock genes, is increasingly implicated in cancer. We have shown that HSF1 binds to the corepressor metastasis-associated protein 1 (MTA1) in vitro and in human breast carcinoma samples. HSF1-MTA1 complex formation was strongly induced by the transforming ligand heregulin and complexes incorporated a number of additional proteins including histone deacetylases (HDAC1 and 2) and Mi2a, all components of the NuRD corepressor complex. These complexes were induced to assemble on the chromatin of MCF7 breast carcinoma cells and associated with the promoters of estrogen-responsive genes. Such HSF1 complexes participate in repression of estrogen-dependent transcription in breast carcinoma cells treated with heregulin and this effect was inhibited by MTA1 knockdown. Repression of estrogen-dependent transcription may contribute to the role of HSF1 in cancer.

No abstract

Administration of RU486 to late pregnant rats results in preterm delivery 24 h after treatment and the induction of a luteolytic process after labor. We investigated whether functional changes occurring within the corpora lutea after RU486 treatment were associated with morphologic features of apoptotic cell death. Rats on d 18 of pregnancy were treated with RU486 (5 mg/kg) at 10:00 am and killed 72 h after. We studied the number of apoptotic cells in paraffin sections of the corpora lutea by routine hematoxylin and eosin (H&E) staining, and by in situ 3' end labeling (TdT-mediated dUTP nick-end labeling [TUNEL]). The corpora lutea were also processed for electron microscopy to study ultrastructural changes after RU486 treatment. The number of cells showing apoptotic nuclei in H&E-stained sections was higher in RU486-treated animals than in controls (vehicle-treated rats). The quantification of the number of apoptotic nuclei within the corpora lutea performed by TUNEL confirmed the higher number of apoptotic nuclei in animals receiving the antigestagen compared with controls. Ultrastructurally, the luteal cells undergoing apoptosis presented a highly deteriorated cytoplasmic organization The nuclei, in an initial step of regression, displayed condensation of the chromatin, a prominent nucleolus, and a perinuclear space. In an advanced step of degeneration, the nuclei showed evidence of large irregular aggregates of condensed chromatin. Prostaglandin F2,alpha(PGF2alpha), which mediates the luteolytic action of RU486, mimicked the effect of the antigestagen on the induction of apoptosis when administered to rats on d 18 of pregnancy (100 microg at 9:00 am and 1:00 pm), which were killed 72 after the last injection. In conclusion, the present results indicate that functional luteolysis in rats is associated with structural luteal regression with the morphologic features of apoptotic cell death, as demonstrated by studying the luteolytic process induced by the administration of the antigestagen RU486.

In previous studies, we found that the human estrogen-regulated heat shock protein (hsp) 27 (human homologue of rat hsp25) is modulated in the endometrium during the different phases of the menstrual cycle and that it is present in endometrial predecidual cells and in decidual cells attached to the placenta. In the present report, we describe the cell type-specific pattern of hsp25 expression in the rat uterus during the periimplantation period as well as during early and late decidualization and placentation. The hsp25 expression pattern was also analyzed in pseudopregnant rats with deciduomas. Immunocytochemistry was performed with an antibody generated against a chimeric hybrid protein containing the N-terminal of the murine hsp25 and the C-terminal of the human hsp27. During pregnancy at the time of implantation, hsp25 was expressed in the endothelial cells of the endometrial vessels and in the luminal epithelium of the antimesometrial region. As pregnancy advanced, hsp25 appeared in predecidual/decidual cells close to the implantation region and then expanded to the mesometrial region. This expression pattern was very similar during pseudopregnancy. Hsp25 was strongly expressed in trophoblastic giant cells beginning on Day 11 of gestation; less expression was noted in the junctional and labyrinth zones of the chorioallantoic placenta (in some cells lining the vascular spaces). In all the disparate cell types that expressed hsp25, the presence of the protein did not correlate with cell proliferation or with apoptosis but with the state of differentiation. Some placental PRL-family members with molecular weights similar to that of hsp25 are also present in antimesometrial decidua and in differentiated trophoblast giant cells; therefore, in this study we eliminated the possibility that our antibody was recognizing prolactin. We also determined that the hybrid hsp25/27 protein did not bind prolactin receptors, and noted that the hsp25 immunostaining pattern was not identical to that of decidual prolactin. In conclusion, the striking cell type-specific timing of hsp25 expression points to hsp25 as a molecule that is important during the implantation, decidualization, and placentation processes.

No abstract