Purpose. To determine whether human retinal pigment epithelial (hRPE) cells produce activin, a growth factor in the transforming growth factor beta family, and to characterize growth regulatory effects of activin on retinal pigment epithelium.Methods. mRNA expression was examined using polymerase chain reaction with primers specific for the beta A and beta B chains of activin and by slot blot analysis with a probe specific for the beta A chain. Protein localization was determined immunocytochemically using antibodies specific for the beta A chain of activin and intact activin A. The effect of activin A on DNA synthesis was studied by measuring (H-3) thymidine incorporation after cells were exposed to recombinant human activin A (rhA). Growth regulatory effects of rhA on hRPE cells were examined with cell growth assays.Results. beta A mRNA was expressed constitutively in 8/8 cell lines tested. beta B mRNA was not expressed in any of the six cell lines tested but was expressed in human ovarian granulosa cell controls. Positive immunostaining was observed for both the beta A chain and intact activin A. (H-3) thymidine incorporation was inhibited 44% (P < 0.025), 45% (P < 0.025), and 44% (P < 0.015) when RPE cells were exposed to 100 ng/ml rhA and grown in serum-free medium, medium with 0.5% serum, and 1% serum, respectively. Cell growth was inhibited 33.2% (P = 0.0001) after RPE cells were exposed to 100 ng/ml rhA for 8 days.Conclusions. These results suggest that activin A can act as an autocrine-paracrine growth regulator in RPE cells and may help control cellular growth in ocular development and proliferative eye disease.

A reverse hemolytic plaque assay was used to measure the GH responses of fetal pituitary cells to GHRH, SRIH, T3 and glucocorticoids. Cells from eight human abortuses (18-22 weeks' gestation) showed accelerated plaque formation after treatment with 10(-7) mol/L GHRH-(1-44) [25.6 +/- 0.6% (+/- SE) of cells formed plaques (PFC); mean area, 14.5 +/- 2.7 X 10(4) micron2; all at 1 h], while 10(-7) mol/L SRIH-(1-28) slowed plaque formation (8.6 +/- 0.6% PFC; mean area, 4.2 +/- 0.8 X 10(4) micron2) vs. control (13.7 +/- 0.7% PFC; mean area, 5.3 +/- 0.8 X 10(4) micron2; all at 1 h). The proportion of PFC was equal in GHRH-treated and control groups by 4 h, suggesting that GHRH affects the amount of GH secreted per somatotroph rather than the number of cells that are preferentially responsive to GHRH. Qualitatively similar data were obtained using pituitary cells from four near-term rhesus fetuses. When cells were cultured in defined medium for 3 days, supplementation with T3 reduced basal GH secretion and attenuated the responses to GHRH. Culture with dexamethasone increased basal GH secretion and restored the responsiveness to GHRH. Dexamethasone also caused a shift in plaque area frequency distributions to patterns similar to those in serum-supplemented medium. We conclude that fetal somatotrophs are responsive to SRIH, GHRH, T3, and dexamethasone. Furthermore, glucocorticoids can maintain a subpopulation of fetal somatotrophs in the GHRH-responsive state.

Departments of Radiology and Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, CA 94143

Previous studies demonstrated that high doses of dopamine administered as a constant infusion were capable of suppressing PRL secretion in both normal and hyperprolactinemic individuals. The present study was designed to examine the dose-response relationship of PRL suppression to low doses of dopamine (between 0.06 and 4 micrograms/kg X min) infused in a step-wise fashion. Ten hyperprolactinemic patients with PRL-secreting pituitary adenomas and 10 normally menstruating women in the early follicular phase of a cycle were studied. For hyperprolactinemic women, the mean (+/- SD) serum PRL level (by RIA) was 112 +/- 101 ng/ml, with a range of 26-386 ng/ml, the mean estradiol level (by RIA) was 28 +/- 8 pg/ml, and the mean baseline dopamine level by carboxy-o-methyl transferase assay (COMT) was 357 +/- 237 pg/ml (2.33 +/- 1.54 X 10(-9) M). The normal women had a mean PRL of 8.1 +/- 3.5 ng/ml, a mean estradiol level of 42 +/- 12 pg/ml, and a mean baseline dopamine level of 317 +/- 223 pg/ml (2.07 +/- 1.45 X 10(-9) M). Concentrations of dopamine achieved at the lowest doses infused were less than 10 times baseline levels and were in the nanomolar range. The measured IC50 (the concentration of dopamine required to achieve 50% of maximal suppression) was 6 +/- 3 X 10(-9) M for normal women and 14 +/- 4 X 10(-9) M for hyperprolactinemic patients. The apparent inhibition constant determined by a nonlinear least squares procedure, was 3 +/- 3 X 10(-9) M for normal women and 11 +/- 12 X 10(-9) M for women with hyperprolactinemia. Both groups exhibited dose-dependent suppression of PRL to about 80% of initial values. Under these experimental conditions, the data do not support the hypothesis that there is a marked loss in sensitivity to dopamine in patients with PRL-secreting adenomas.

This article deals with the formation and metabolism of steroid hormones in pregnancy and the proposed functions of these hormones; the formation of the polypeptide and neuropeptide hormones of fetus and placenta and their regulatory roles in pregnancy homeostasis; and the fetal thyroid gland and its regulation. It also focuses on the role played by the endocrine system in allowing the pregnancy to function as a heterograft and in initiating the process of labor.