Search

PRL has been shown to stimulate mRNA expression of both ERalpha and ERbeta in the rat corpus luteum and decidua of pregnancy. To investigate whether PRL may stimulate ER expression at the level of transcription and which transcription factors may mediate this stimulation, we have cloned the 5'-flanking regions of both rat ER genes. A constitutively active PRL receptor (PRL-R(CA)) stimulated both ERalpha and ERbeta promoter activity, indicating that PRL is acting to stimulate ER transcription. Putative signal transducer and activator of transcription (Stat)5 response elements were identified at -189 in the ERalpha promoter and at -330 in the ERbeta promoter. Mutation of these response elements or overexpression of dominant negative Stat5 prevented stimulation of ERalpha and ERbeta promoter activity, indicating that PRL regulation of ER expression requires both intact Stat5 binding sites as well as functional Stat5. Interestingly, either Stat5a or Stat5b could stimulate ERalpha transcription while stimulation of ERbeta occurred only in the presence of Stat5b. Through mutational analysis, a single nucleotide difference between the ERalpha and ERbeta Stat5 response elements was shown to be responsible for the lack of Stat5a-mediated stimulation of ERbeta. These findings indicate that PRL stimulation of ER expression occurs at the level of transcription and that PRL regulation of ERalpha can be mediated by either Stat5a or Stat5b, while regulation of ERbeta appears to be mediated only by Stat5b.

Several strains of laboratory rats have a high background incidence of mammary tumors and develop a persistent, anovulatory estrus condition at about 12 mo of age. The increased tumor incidence is believed to be associated with elevated estradiol (E2) and prolactin during the period of persistent estrus. A pharmacodynamic estrus cycle (PD-EC) model for the Sprague-Dawley rats has been developed in an attempt to analyze the physiological basis of early-onset persistent estrus and to examine the potential sites of interactions in the hypothalamic-pituitary-ovarian axis for endocrine-modulating xenobiotics that accelerate the onset of persistent estrus. This initial estrus cycle model focused solely on cyclical changes in E2 and luteinizing hormone (LH). An LH surge was scheduled when a hypothetical estrus cycle-related protein (EC-RP) under transcriptional control by the E2 receptor reached a critical concentration. In the model, aging-related cumulative hypothalamic E2 exposure impaired the LH surge by reducing the rate of production of the EC-RP. The progressively decreasing intercycle resynthesis rate leads first to longer, variable-length cycles and finally to persistent estrus at about 12 mo of age. This model construct is consistent with early-onset persistent estrus related to neonatal E2 exposures, with acyclicity associated with high-dose E2 exposure in the adult, and with persistent estrus conditions associated with exposures to xenobiotic endocrine modulators that are either weak E2 antagonists or weak E2 agonists. With further development these pharmacodynamic estrus cycle models should be useful in aiding risk assessments for compounds causing mammary-tissue tumors associated with persistent estrus states.

Our recent studies demonstrated that the continuous administration of a GnRH agonist (GnRH-Ag; WY 40972) induces abortion in rats by suppressing plasma progesterone (P) levels within 24 h. This fall in P levels is not accompanied by a fall in ovarian venous plasma testosterone (T) or estradiol (E) levels. In this study an attempt was made 1) to determine whether the suppression of P by GnRH-Ag is due to decreased estrogen present in the corpora lutea (CL) and/or a decrease in luteal receptors of E, and 2) to investigate the effects of GnRH-Ag on the nocturnal surges of PRL. Rats were treated continuously on days 7-11 of pregnancy with 5 micrograms/day GnRH-Ag using an osmotic minipump. Ovarian blood samples were obtained on day 8; at autopsy CL were harvested and incubated with medium 199 for 4 h at 37 C under an atmosphere of 95% oxygen-5% carbon-dioxide. Additional rats were killed on day 8 or 10; CL were isolated from the ovary and pooled within the group for the measurement of nuclear and cytosolic E receptors. In other experiments, on days 8, 9, 10, and 11 of pregnancy, blood samples (0.3 ml) were collected via an indwelling intraatrial Silastic cannula at 0330, 0500, or 0600 h for the measurement of PRL and P. While the net synthesis of P by CL in the GnRH-Ag-treated rats decreased to 48 +/- 12 from 224 +/- 47 ng/CL in controls, T and E levels were not different from their respective control values. Steroid levels in ovarian venous plasma reflected a similar response. Nuclear E receptors levels were 82 and 80 in controls and 39 and 41 fmol/mg DNA in the treated group on days 8 and 10, respectively. Nocturnal surges of PRL in plasma were detected at 0330 h in controls as well as in treated rats. However, plasma PRL levels at 0330 h were 101 +/- 24, 120 +/- 22, 196 +/- 40, and 103 +/- 13 ng/ml in controls and 44 +/- 8, 50 +/- 10, 29 +/- 13, and 20 +/- 9 in the GnRH-Ag-treated group on days 8, 9, 10, and 11, respectively. These results suggest that GnRH-Ag has no effect on the ability of the luteal synthesis of T and E and that the antipregnancy effect of GnRH-Ag may be at the level of the CL due to the direct inhibitory effect of GnRH-Ag on the luteal synthesis of P, which, in turn, results in decreased nocturnal surges of PRL and a fall in E receptors in the CL. Alternatively, GnRH-Ag treatment could suppress ovarian or luteal receptors for PRL, which, in turn, lower luteal E receptors, leading to a fall in luteal synthesis and release of P.