Vascular endothelial growth factor (VEGF) is potentially an important regulator of angiogenesis, particularly during the extensive tissue growth and remodeling that occur in utero. In the present study, we have investigated the role of VEGF during human fetal development by analyzing the distribution of VEGF messenger RNA as well as the tissue- and cell-specific localization of VEGF peptide in the human midgestation (16-22 weeks) fetus. As a comparison, we conducted parallel studies on several human adult tissues. Messenger RNA encoding VEGF was detected in all fetal tissues studied and was most abundant in human fetal lung, kidney, and spleen; moderately abundant in heart, adrenal, pancreas, intestine, liver, testis, skin, muscle, and brain; and minimally detected in thymus and placenta. VEGF peptide, detected by immunohistochemistry, always was intracytoplasmic and localized principally in epithelial cells and myocytes, including the smooth muscle cells lining blood vessels. VEGF was not detected in vascular endothelial cells. As the cellular localization of VEGF in several human adult tissues was similar to that found in the cognate fetal tissues, VEGF is probably important not only in angiogenesis, but also in the maintenance of existing vessels. As VEGF was localized primarily in epithelial cells and myocytes and not in endothelial cells, these data are consistent with a paracrine mechanism of action whereby VEGF secreted by nonendothelial cells modulates activities in adjacent vascular endothelium.

Our study compared 84 patients with polycystic ovary syndrome (PCOS) with 84 control patients who had normal ovaries and who were matched for the main determinants of success in in-vitro fertilization (IVF) and embryo transfer. Serum concentrations of oestradiol and progesterone on the day of human chorionic gonadotrophin (HCG) injection were significantly higher in PCOS than in normal patients (oestradiol 2016 +/- 1.8 pg/ml versus 1456 +/- 40.9 pg/ml, P < 0.01; progesterone 1.6 +/- 0.1 ng/ml versus 1.2 +/- 0.1 ng/ml, P = 0.03). Furthermore despite oocytes from PCOS patients having a reduced fertilization rate compared with normal patients (61.8 +/- 4.1% versus 73.5 +/- 4.3%, P = 0.03), the differences in pregnancy rate (22.6 versus 19%) and miscarriage (31.5 versus 18.7%) were not statistically significant. In PCOS patients, a critical breakpoint was identified at serum progesterone concentrations of 1.2 ng/ml on the day of HCG injection. The PCOS patients with progesterone > or = 1.2 ng/ml showed a higher pregnancy and miscarriage rate than PCOS patients with progesterone < 1.2 ng/ml (26.6 versus 17.9%, P < 0.01; and 41.7% versus 14.3%, P < 0.01 respectively). These findings suggest that premature progesterone production does not have an adverse effect on pregnancy rate in PCOS, but on the contrary, may be a predictor for success in IVF/embryo transfer.

The actions, localization, and regulation of activin in the human ovary are unknown. Therefore, the aims of this study were (a) to define the effects of recombinant activin-A and its structural homologue, inhibin-A, on mitogenesis and steroidogenesis (progesterone secretion and aromatase activity) in human preovulatory follicular cells; (b) to localize the activin-A dimer in the human ovary by immunohistochemistry; and (c) to examine regulation of intracellular activin-A production in cultured human follicular cells. In addition to stimulating mitogenic activity, activin-A causes a dose-and time-dependent inhibition of basal and gonadotropin-stimulated progesterone secretion and aromatase activity in human luteinizing follicular cells on day 2 and day 4 of culture. Inhibin-A exerts no effects on mitogenesis, basal or gonadotropin-stimulated progesterone secretion and aromatase activity, and does not alter effects observed with activin-A alone. Immunostaining for dimeric activin-A occurs in granulosa and cumulus cells of human ovarian follicles and in granulosa-lutein cells of the human corpus luteum. cAMP, and to a lesser degree human chorionic gonadotropin and folliclestimulating hormone, but not inhibin-A, activin-A, or phorbol 12-myristate 13-acetate, increased the immunostaining for activin-A in cultured granulosa cells. These results indicate that activin-A may function as an autocrine or paracrine regulator of follicular function in the human ovary. (J. Clin. Invest. 1992.

The use of metformin with GnRH antagonist improves the outcome of ovarian stimulation in IVF-ET cycles in PCOS patients.