Search

In spite of the importance of the corpus luteum in human reproduction, little is known about its formation after ovulation and during regression in the absence of conception. This is largely due to constraints on the availability of normal human tissue; therefore an appropriate model which could be studied and could provide information applicable to the human was sought. The baboon (Papio), a non-human primate, has been determined to be one such model. Thus, in the past several years our studies have examined the role of luteal peptides in corpus luteum function, and, when possible, we have attempted to examine corpora lutea from the human and baboon in parallel. Although a milk-ejection factor was recognized to be present in luteal tissue in 1910 (Ott and Scott, Proc. Soc. Exp. Biol. Med., Vol. 8, p. 49), the role of oxytocin in luteal physiology has not been easy to ascertain. This is in part due to the methodologies employed to assess its role. Our studies summarized below suggest that oxytocin does not directly affect luteal steroidogenesis but that it may play a role in cell to cell communication involving the expression of the gap junction proteins, the connexins. In view of the fact that oxytocin, its receptor, gap junctions and associated proteins are not unique to the human and non-human primates, the model of luteal development and demise proposed may be applicable to most species.

The synthesis and secretion of progesterone in the corpus luteum are regulated by both endocrine and paracrine/ autocrine factors which affect the steroidogenic cells. Evidence suggests that these cells communicate via cell-cell junctional proteins, the connexins. Previously we have shown that connexin-43 is expressed in both human and baboon (Papio hamadryus anubis) corpora lutea, with differential expression throughout luteal development, but is not detectable in corpora albicantia. We have examined the effect of human chorionic gonadotropin (hCG), oxytocin, clomiphene citrate and the anti-progesterone onapristone on expression of connexin-43 protein in the early luteal phase 1-5 days after the mid-cycle luteinizing hormone (LH) surge (LH+ 1-5 days), the mid-luteal phase 6-10 days after the LH surge (LH+ 6-10 days), and the late luteal phase 11-15 days after the LH surge (LH+ 11-15 days) in corpora lutea obtained from normal adult cycling females. Connexin-43 was localized by immunohistochemistry in cultured cells from all the three stages. Western blot analysis of the treated cells indicated the presence of two bands at 43 and 45 kDa. The band at 45 kDa was found to be phosphorylated connexin-43, indicating the presence of functional gap junctions. hCG (10 IU/ml) stimulated the expression of connexin-43 throughout luteal development; however, maximum expression occurred in the early luteal phase with a significantly greater expression of the non-phosphorylated protein. In contrast, in the mid-luteal phase, the expression of the phosphorylated protein was predominant. Oxytocin (200 mU/ml) also stimulated connexin-43 expression throughout luteal development with similar effects on the phosphorylated and non-phosphorylated protein in the early and mid-luteal phase; however, compared with hCG, oxytocin had a greater effect on mid-luteal phase connexin-43 expression. In the presence of both hCG and oxytocin, the expression of connexin-43 was significantly higher than the control only in the late luteal phase. Both clomiphene citrate and onapristone suppressed connexin-43 expression, and concomitant addition of hCG did not counteract their effect. In the context of our previous studies, it is concluded that, together with LH/hCG and the steroid hormones, oxytocin is involved in cell-cell contact-dependent communication in the corpus luteum.

Interleukin-1 beta (IL-1 beta) modulates steroidogenesis and prostaglandin (PG) secretion by dispersed luteal cells of some non-primate species. To determine if IL-1 beta affects progesterone (P) and PGF2 alpha secretion by the baboon corpus luteum (CL), we microretrodialyzed the intact CL for 48 h; 12 h media (baseline), 12 h with IL-1 beta (3 IU/h), 12 h media only (post- IL-1 beta) and 12 h with cAMP (5 nmol/h). Four CL from the midluteal phase (LH + 8 days) were studied. P was measured by a sensitive and specific radioimmunoassay and PGF2 alpha by an enzyme immunoassay in the 10-min fractions of retrodialysates collected. P secretions were analyzed for peaks by PC Pulsar (3.0) and total P retrieved/12 h for each experimental segment was calculated. P secretion was pulsatile. Pulses of P declined from 8.2 +/- 1.2/12 h (mean +/- SEM) before to 5.0 +/- 1.2 h after IL-1 beta treatment (P = 0.022), but increased to 10.2 +/- 4.3/12 h with cAMP. Interpulse interval increased significantly from 92 +/- 23 min (baseline) to 137 +/- 31 min (p = 0.025) after IL-1 beta treatment. Total P secreted decreased significantly from 2471 +/- 515 nmol/12 h (baseline) to 1480 +/- 167 nmoles/12 h during IL-1 beta and 788 +/- 85 nmoles/12 h after IL-1 beta (P = 0.015). P was immediately suppressed after starting IL-1 beta in 2 CL but declined only towards the end of treatment in the other 2 CL. PGF2 alpha secretion increased during IL-1 beta with a further increase after IL-1 beta, while P secretion was progressively inhibited. Therefore, IL-1 beta is luteolytic to the primate midluteal phase CL by inhibiting P while simultaneously stimulating PGF2 alpha secretion, demonstrating paracrine-autocrine interaction within the luteal tissue.

We have recently shown the presence of E-cadherin and of alpha- and gamma-catenins in human and baboon corpora lutea. These are components of adherens junctions between cells. The cytoplasmic catenins link the cell membrane-associated cadherins to the actin-based cytoskeleton. This interaction is necessary for the functional activity of the E-cadherins. Our aim therefore was to determine the presence of alpha-actin in the baboon corpus luteum, to further establish whether the necessary components for E-cadherin activity are present in this tissue. An antibody specific for the smooth muscle isoform of actin, alpha-actin, was used for these studies. The results using immunohistochemistry show that (a) alpha-actin is present in steroidogenic cells of the active corpus luteum, theca externa of the corpus luteum, cells of the vasculature, and the tunica albuginea surrounding the ovary. The intensity of immunoreactivity for alpha-actin varied, with the cells of the vasculature reacting more intensely than the luteal cells. A difference in intensity of immunoreactivity was also observed among the luteal cells, with the inner granulosa cells showing stronger immunoreactivity than the peripheral theca lutein cells. There was no detectable immunoreactivity in the steroidogenic cells of the atretic corpus luteum. However, in both the active and atretic corpora lutea, alpha-actin-positive vascular cells were dispersed within the tissue. (b) Total alpha-actin (luteal and non-luteal), as determined by Western blot analyses, does not change during the luteal phase and subsequent corpus luteum demise (atretic corpora lutea). (c) hCG stimulated the expression of alpha-actin and progesterone secretion by the early luteal phase (LH surge + 1-5 days) and mid-luteal phase (LH surge + 6-10 days) cells in culture, but only progesterone in the late luteal phase (LH surge + 11-15 days). The data show that alpha-actin is present in luteal cells and that its expression is regulated by hCG, thus suggesting that E-cadherin may form functional adherens junctions in the corpus luteum.

Although oxytocin has been recognized as a product of the corpus luteum in numerous species, including nonhuman primate and women, for sometime, its precise role in luteal physiology has remained obscure. However, with the recent observations that the steroidogenic activity of the large and small cells is increased in the presence of LH when these cells are in intimate contact has led to the hypothesis that cell-to-cell communication must occur between these cells. Cell-to-cell communication is possible via several mechanisms, including paracrine/autocrine and intercellular crosstalk via gap junctions. Substantial morphological and immunohistological evidence using antibodies to gap-junction specific proteins, the connexins, indicates the presence of gap junctions in corpora lutea. Our recent studies indicate that oxytocin affects the expression of the gap-junction protein connexin-43 and that the gonadotropins are intimately involved in this action. The synthesis of oxytocin and the oxytocin receptor is influenced by the gonadotropins and locally produced prostaglandins. Oxytocin stimulates estradiol synthesis, which may affect the expression of the gap-junction protein connexin-43, allowing interaction between the large cells and small cells of the corpus luteum. With the ubiquitous presence of oxytocin and its receptor, and the presence of gap junctions in the corpora lutea of numerous species, it is concluded that oxytocin is involved in not only paracrine/autocrine interaction but also may be of significant importance in intercellular crosstalk in the corpus luteum.

Baboon corpora lutea (two each from the early, mid- and late luteal phases) were individually microretrodialyzed in vitro for 48 h, 12 h initial baseline, 12 h retrodialysis with OT (9 mU/h), 12 h without OT and 12 h with cAMP (5 mmol/h). Progesterone (P) was measured by a sensitive and specific radioimmunoassay in 10-min fractions of retrodialysates and analyzed for P peaks by PC-pulsar 3.0. Neither OT nor cAMP had any effect on the characteristics of P pulses. In early and late luteal phase CL, OT inhibited P secretion within 1 h of administration followed by increased P secretion late during OT perfusion. In midluteal phase, OT did not affect P secretion. In all CL, P secretion was sustained or further increased during the 12 h after stopping OT. cAMP also sustained baseline or stimulated P secretion. In contrast, OT either increased total P output/12 h (28 to 49% above baseline) with a further increase of 21% to 296% above baseline after stopping OT, or inhibited total P output by 4% to 13% percent with a further decline of 51% to 61% after stopping OT. Thus, while overall OT is luteotropic, its dual effect (initial inhibition followed by stimulation) suggests direct and indirect effects through paracrine-autocrine mechanisms.

We have previously shown that the protein connexin-43 which forms the connexons in gap junctions is present in the human corpus luteum. Abundant expression of connexin-43 is seen in the mid-luteal phase corpora lutea. Since the formation of gap junctions in a tissue requires the presence of adherens junctions formed by the cadherins, our aim in these studies was firstly to localize immunocytochemically E-cadherin and beta-catenin (a cytoplasmic protein associated with E-cadherin) in the human corpus luteum, and secondly to determine the concentrations of these proteins in the early, mid- and late luteal phase human corpora lutea. E-cadherin was localized to the periphery of luteal cells and was not detected in non-luteal tissue. beta-catenin was observed in the cytoplasm of the luteal cells. Abundant expression of E-cadherin was observed by Western analysis in the early luteal phase and the level of expression was significantly different from that observed in the mid- and late luteal phase corpora lutea. In contrast the concentrations of beta-catenin were higher in the mid-luteal phase compared to the early luteal phase. The differential expression of the cell adhesion molecule E-cadherin suggests that it may play a significant role in cell-to-cell communication in the corpus luteum, and in the cyclic development and demise of this tissue.

We have previously shown that gap junctions and gap junction-associated protein connexin 43 are present in the human and baboon corpus luteum. Abundant expression of connexin 43 is seen in the midluteal phase corpora lutea. Since the formation of gap junctions requires the presence of adherens junctions formed by the cadherins, our aim in these studies was 1) to immunocytochemically localize E-cadherin and the associated proteins, beta-catenin and gamma-catenin (plakoglobin), in the baboon corpus luteum; 2) to determine by Western analysis the levels of these proteins in early, mid-, and late luteal phase and atretic baboon corpora lutea; and 3) to examine whether or not cell-cell contact in cells in culture is disrupted by the addition of the antibody for E-cadherin. E-cadherin was localized to the peripheral cell membranes of luteal cells at all stages examined, except atretic corpora lutea, with the strongest immunoreactivity in the early luteal phase. Both beta-catenin and plakoglobin were localized in the cytoplasm of the luteal cells. Immunoreactivity for all three peptides was not observed in nonluteal tissue. By Western analysis, abundant expression of E-cadherin was observed in the early luteal phase, and the level of expression was significantly different from that observed in the mid- and late luteal phase corpora lutea. In contrast, the levels of beta-catenin and plakoglobin were higher in the midluteal phase compared to the early luteal phase. Addition of the E-cadherin antibody to early luteal phase cells in culture disrupted the cell-cell contacts between cells. Thus, cell adhesion involving E-cadherin may play a significant role in the cyclic development and demise of this tissue.

The identification of the cell junction-forming proteins connexin-43, a gap junction protein and E-cadherin, which is a component of adherent junctions, in the corpus luteum of both humans and baboons suggests that cell-cell interactions and metabolic cooperation must occur in this tissue. Occluding junctions are a third type of junction which form a physical barrier between cells. Thus, our aims in this study were firstly to examine the presence of the tight junction-associated protein zonula occludens-1 (ZO-1) by immunohistochemistry, and secondly to determine the concentrations of this protein in the early-, mid- and late luteal phase baboon corpora lutea of the menstrual cycle by a Western analysis. ZO-1 was localized mainly at the periphery of the luteal cells, and the intensity of immunoreactivity varied through the luteal phase, with comparatively stronger immunoreactivity in the mid-luteal phase than the early and late luteal phases. Atretic corpora lutea were devoid of activity. By Western analysis, bands of immunoreactivity were observed at 225 kDa, further confirming the presence of the protein. Maximum activity, as determined by densitometry, was observed in the mid-luteal phase. These data infer the presence of tight junctions in the corpus luteum and suggest that expression of the ZO-1 protein forming these junctions may be hormonally regulated within this tissue.

In the nonhuman primate and human corpora lutea, gap junctions have been identified by means of electron microscopy. Gap junctions are formed by connexons, which consist of a multigene family of tissue-specific connexins. In the ovarian follicle, the gap junction protein connexin-43 is present and hormonally regulated. However, there is little evidence indicating the type of connexin present in the corpus luteum. Therefore, the aim of this study was to demonstrate the presence of gap junctions by electron microscopy and the presence of connexin-43 and messenger ribonucleic acid (mRNA) for this protein. Using immunocytochemical procedures, we have shown the presence of connexin-43 in baboon and human midluteal phase corpora lutea and in the atretic corpora lutea of the baboon. The intensity of immunoreactivity was lower in atretic corpora lutea than in the midluteal phase corpora lutea. Western analysis indicates the presence of two bands at 43-45 kDa, and that the levels of connexin-43 protein are abundant in the midluteal phase. The two bands suggest the presence of the protein in a phosphorylated or a nonphosphorylated form. Ribonuclease protection assay suggests that the mRNA levels of connexin-43 remain constant throughout the luteal phase. mRNA for connexin-43 was not detectable in atretic corpora lutea. Thus, connexin-43 is one of the connexin family of proteins forming the connexon of gap junctions in the baboon and human corpus luteum. The expression of the protein may be hormonally regulated by locally produced factors, such as estradiol and progesterone. We suggest that gap junctional communication between the cells of the primate and human corpus luteum may be important in hormone synthesis and secretion and may be involved in the process of luteolysis through luteal cell apoptosis.

Oxytocin has been identified in both non-human primate and human corpora lutea of the menstrual cycle by RIA, immunocytochemistry and HPLC. Evidence for the transcription of the oxytocin gene in this tissue using PCR is available. Oxytocin receptors have been characterized by biochemical procedures. However, there is some debate as to whether the oxytocin identified in these tissues is biologically active and has a role in luteal function. In this study we have demonstrated that oxytocin isolated by gel chromatography of tissue extracts from the baboon and the human corpus luteum is biologically active as determined in a rat uterine bioassay. Since both oxytocin and its receptors are present in these tissues, it is suggested that oxytocin in the human and non-human primate corpora lutea has a functional role.

Objective: To determine the presence and compare the relative abundance of messenger RNA for epidermal growth factor receptor (EGFR) in baboon corpora lutea of different luteal ages., Design: Prospective controlled nonhuman primate study., Setting: Animal facility in an academic research institution., Participants: Six adult female baboons with well-defined regular menstrual cycles. Stage of the menstrual cycle was determined by observation and scoring of perineal turgescence. The day of maximal turgescence was referred to as the day of LH surge., Intervention: Ten corpora lutea were obtained by luteectomy during early (LH + 1 to 5 days, n = 3), midluteal (LH + 6 to 10 days, n = 3), and late (LH + 11 to 15 days, n = 4) luteal phases., Main Outcome Measure: Relative levels of messenger RNA for EGFR as determined by ribonuclease protection assay using RNA probe generated from complementary DNA for human EGFR., Results: Messenger RNA for EGFR is present in baboon corpora lutea with relative levels of 0.51 +/- 0.18 (mean +/- SEM) in the early, 0.43 +/- 0.17 in the midluteal, and 0.50 +/- 0.17 arbitrary units in the late luteal phase., Conclusion: Baboon corpus luteum is a site of EGFR production; the levels of its messenger RNA did not change appreciably throughout the luteal phase.

Although oxytocin (OT) was identified in the human and primate corpus luteum (CL) over a decade ago, a definitive role for this peptide has not been elucidated. Several in vitro models have been utilized to examine the most obvious role for OT in luteal function, that of its effect on progesterone (P) production. Using dispersed cells in short term incubations, cultured cells and microdialysis procedures utilizing intact tissue, variable effects of OT on P production have been obtained. We therefore hypothesized that OT may have other role(s) in this tissue. The follicle cells remaining after a successful ovulation has occurred undergo rapid luteinization and the tissue is extensively remodeled to form the CL. At this time, in most species, two types of luteal cells have been identified based on their morphology, biochemistry and size. Both cell types produce P, however only one cell type responds to the long distance modulator LH. Since both cell types appear to be needed for the overall synthesis of P in adequate quantities required for the preparation of the endometrium "information trafficking" must occur between the two cell types. Our recent studies have shown the presence of gap junctions and connexin-43, E-cadherin, an adhesion molecule and ZO-1 protein associated with tight junctions in the baboon and human CL. The effect of OT on connexin-43 in the baboon CL has been examined.

We have demonstrated the presence of epidermal growth factor (EGF) and its receptors in human non-gestational corpora lutea. To determine further the characteristics of EGF receptor binding, we examined 30 human corpora lutea throughout the luteal phase and during pregnancy. Scatchard plots of EGF binding in 29 of the 30 corpora lutea were curvilinear, suggesting negative co-operativity. The mean +/- SE of the association constant Ka was (0.9 +/- 0.2) x 10(9) l/mol, the dissociation constant Kd was (2.2 +/- 0.3) x 10(-9) mol/l and the number of binding sites (Rt) was (15.8 +/- 2.1) x 10(-19) mol/micrograms protein for non-gestational corpora lutea. The Kd increased significantly in late pregnancies compared to early pregnancies (P = < 0.005), while Rt was significantly higher in term pregnancies than in either early pregnancy (P < 0.01) or the menstrual cycle (P < 0.001). Corpora lutea atretica (n = 2) and ovarian stroma (n = 6) did not show any EGF binding activity. Our findings demonstrate the presence of specific EGF receptors in human corpora lutea of both the menstrual cycle and pregnancy. The changes in EGF binding parameters in early pregnancy suggest that there may be a relationship between the role of EGF and ovarian steroidogenesis.

The effects of insulin-like growth factor-1 (IGF-1) and cAMP on progesterone (P) secretion by CL of streptozotocin-induced-diabetic pregnant ewes were compared with the effects on normal pregnant animals. Two types of CL were identified in the ovaries removed on Days 126.6 +/- 2 of pregnancy. They were either large, reddish in color, and vascular (type A) or small and pale yellow (type B). Both types were found in diabetic and normal sheep. Each CL was divided in two and perfused in parallel for 14 h in an in vitro microdialysis-perifusion system. One half was used to evaluate basal P secretion and the effect of cAMP. The effect of IGF-1 and cAMP infusion was studied in the other half. During microdialysis, fractions were collected every 15 min, and P was determined by RIA. IGF-1 stimulated secretion of P in the large type A, normal and diabetic sheep CL in discrete pulses. The smaller CL (type B) from normal sheep produced comparatively higher levels of P in discrete pulses in the presence of IGF-1. However, the small CL from diabetic sheep showed no response to IGF-1 or cAMP, and P secretion was lower. Thus, it is probable that the large CL may be the "active" CL producing P and that IGF-1 stimulates pulsatile P secretion in such CL from both normal and diabetic pregnant sheep.

Pulsatility of serum progesterone (P) is usually ascribed to stimulation of the corpus luteum (CL) by pulsatile release of pituitary LH. We investigated P secretion by the primate CL by performing microretrodialysis on 6 fresh CL obtained at laparotomy from baboons (Papio anubis) with well defined menstrual cycles. Individually microdialyzed for 24-26 h with Dulbecco's Modified Eagle's Medium and Ham's F-12 enriched with HEPES buffer in a perifusion chamber, the retrodialyzed fluid was collected every 10 min and measured for P, estradiol, and 17 alpha-hydroxyprogesterone by specific and sensitive RIAs. The chronodynamics of hormone secretion were analyzed for pulse detection by PC-Pulsar 3.0. All 6 CL (2 each from early, LH +1 to +5; mid, LH +6 to +10; and late luteal phases, LH +11 to +15) demonstrated pulsatile secretion of P in vitro, with distinct and detectable peaks over the 24-26 h studied. The CL secreted 23-27 pulses of P in 24 h in early luteal, 8-20 pulses in midluteal, and 6-19 pulses in late luteal phases. Peak lengths were 23.8 +/- 18.5 to 35.7 +/- 17.1 min. Four CL gave interpeak intervals of 46-55 min, whereas two gave intervals of 136-137 min. Analysis of distribution of pulses against different interpulse intervals in individual CL and all CL together revealed a bell-shaped distribution, with the largest number of pulses seen at an interpulse interval of 21-40 min. Because of the low concentrations of estradiol and 17 alpha-hydroxyprogesterone retrodialyzed, a similar analysis of these data was not possible. Histological examination of the tissue at the termination of the experiment using hematoxylin and eosin and localization of 3 beta-hydroxysteroid dehydrogenase activity indicates that the steroidogenic potential of the tissue is minimally affected, although some morphological changes do occur. Our findings suggest autonomous pulsatile P secretion by the primate CL, indicating local control by and the presence of an intraluteal oscillator or pulse generator for P secretion.

Objective: To examine hCG receptor levels and binding kinetics in corpora lutea (CL) from tubal pregnancies., Design: Corpus luteum from tubal pregnancies, term normal pregnancies, and midluteal phase were obtained and compared for hCG binding., Setting: Department of Obstetrics and Gynecology, University Hospital., Interventions: Corpus luteum were obtained at elective cesarean section at term, tubal ligation, or surgery for ectopic pregnancy (EP)., Main Outcome Measures: Human chorionic gonadotropin receptor levels, dissociation and association constants using saturation binding assay, and Scatchard analysis., Results: Corpus luteum from late tubal pregnancies (> 7 weeks) and term intrauterine pregnancies had undetectable hCG binding. Corpus luteum from early tubal pregnancies (< 6 weeks) had significantly reduced hCG receptors with higher dissociation constants than from midluteal phase CL., Conclusions: Significant reduction of hCG receptors with accelerated binding dissociation may be the basis of CL dysfunction in some EPs.

To determine whether the effect of oxytocin (OT) on progesterone production in primate luteal tissue may be receptor-mediated, we used biochemical procedures to examine the CL of the baboon (Papio anubis) for the presence of OT receptors (OT-R). CL (n = 5) were obtained by luteectomy from the early (LH + 1-5 days), mid- (LH + 6-10 days), and late (LH + 10-15 days) luteal phases. Biopsies from the uterine fundus were also obtained at the same time. Total receptors were measured by incubating the membrane fractions (120,000 x g fraction) with increasing concentrations of [3H]oxytocin (0.08-1.6 nM) for 30 min at 22 degrees C with a protein concentration of 200 micrograms per tube. Nonspecific binding was evaluated by the addition of excess nonradioactive OT. The receptor-bound tritiated ligand was separated from the nonreceptor-bound ligand through use of polyethylene glycol. For the midluteal phase CL, the affinity Kd was 1.70 +/- 0.03 nmol/L and the mean receptor number (RT) was 122.6 +/- 8.3 fmol/mg protein. In comparison, myometrial tissue examined at the same phase of the cycle had a Kd of 1.3 +/- 0.1 nmol/L and an RT of 205.0 +/- 49.7 fmol/mg protein. Receptors were not detectable in the late luteal phase CL and were significantly lower in the early luteal phase CL (50.7 +/- 6.9 fmol/mg protein), with no difference in the Kd value. Competitive studies with structurally related and nonrelated peptides indicated the presence of specific receptors for OT in baboon CL.(ABSTRACT TRUNCATED AT 250 WORDS)

Objective: To determine the effect of melatonin on baboon corpus luteum (CL) cell progesterone (P) production., Design: Five baboon CL obtained during the midluteal phase by luteectomy were dissociated using collagenase, and incubations were performed (50,000 cells per plate) in quadruplicate for 3 hours at 37 degrees C with melatonin (0.001 to 1.0 ng/mL) (basal) or with melatonin and 10 IU of human chorionic gonadotropin (hCG) (hCG-stimulated). Total P was measured by a specific radioimmunoassay., Main Outcome Measures: Progesterone concentrations measured in the presence and absence of melatonin and hCG., Results: Melatonin (0.01 to 1.0 ng/mL) inhibited basal P production in all the CL (41.8 +/- 9.9 ng P without melatonin compared with 32.2 +/- 2.0 ng P, 28.4 +/- 2.1 ng with 0.01 and 1.0 ng/mL melatonin, respectively). Human chorionic gonadotropin-stimulated P production was significantly inhibited with as little as 0.01 ng of melatonin (150.8 +/- 11.4 ng with 10 IU hCG versus 120.3 +/- 6.4 ng with 10 IU hCG and 1.0 ng melatonin). The degree of inhibition in the hCG-stimulated cells was greater than in the nonstimulated cells. Melatonin at a concentration of 0.001 ng/mL did not affect P production in both stimulated and nonstimulated cells. Serotonin in similar concentrations had no effect on luteal cell P production., Conclusion: These findings indicate that melatonin exerts a suppressive effect on baboon dispersed luteal cell P production and thus may play a role in luteal function.

Objective: To determine the presence and binding characteristics of insulin-like growth factor (IGF) receptors in corpora lutea (CL) of spontaneous and clomiphene citrate (CC)-induced cycles and to identify any relationship between IGF receptors and cytosol progesterone (P) and 17 alpha-hydroxyprogesterone (17 alpha-OHP) levels., Design: Women undergoing bilateral tubal ligation were divided into two groups. One group received no medication (controls) and the other took 50 mg of CC. Midluteal phase CL were recovered at tubal ligation for hormone and receptor analysis., Setting: Patients were recruited from a university hospital setting., Patients: Eleven fertile women 26 to 37 years of age requesting bilateral tubal ligation were studied., Interventions: Four women were given 50 mg/d of CC from days 5 through 9 of study cycle. Seven women did not take any medication. Minilaparotomy bilateral tubal ligation and luteectomy were performed 7 to 9 days after midcycle urinary (LH) surge., Main Outcome Measures: Insulin-like growth factor receptor concentrations and binding characteristics and cytosol P and 17 alpha-OHP levels in individual CL., Results: Optimal binding for 125I-IGF-I with membrane fractions of homogenized CL was obtained with incubation at 4 degrees C for 16 hours. Specific binding (mean +/- SEM) was significantly higher in CC-treated (53.6% +/- 4.8%) than in control cycles (25.9% +/- 5.5%, P less than 0.001). Receptor concentrations were also significantly higher in CL from CC-induced (145.6 +/- 21.8 pmol/mg protein) than from control cycles (74.8 +/- 15.2 pmol/mg protein, P less than 0.02). Insulin-like growth factor receptor levels correlated with neither serum nor cytosol P and 17 alpha-OHP in CL from either cycles., Conclusion: Specific IGF-I receptors are present in human CL of the menstrual cycle with higher concentrations present in CC-induced cycles. Thus IGF may express its action on luteal function through its receptors in CL.

A variety of molecular techniques were used to search human and baboon gonadal tissues for evidence of transcription of the genes for the peptide hormones oxytocin and vasopressin. Only a highly sensitive assay based on a modification of the polymerase chain reaction succeeded in detecting mRNA copies of the oxytocin gene in both human and baboon corpus luteum. Vasopressin gene transcription was not detected in human testis and corpus luteum and was found only once in four different experiments in baboon corpus luteum. Evidence for oxytocin gene transcription in the human testis was found in three of five experiments. The method employed and subsequent sequence analysis of the polymerase products verified the presence of oxytocin mRNA with normal hypothalamic-type exonic structure in primate corpus luteum. Nevertheless, the very low levels of mRNA present are unlikely to support other than local functions for the encoded nonapeptide hormones.

Midluteal phase corpora lutea (CL) obtained from women induced with 50 mg (n = 5), 100 mg (n = 5), and 150 mg (n = 5) of clomiphene citrate (CC) were measured for luteinizing hormone/human chorionic gonadotropin (LH/hCG) concentrations and cytosol progesterone (P) and 17 alpha-hydroxyprogesterone (17-OHP) and compared with midluteal phase CL from eight normal women (controls). More CL (26) that were significantly heavier (2.0 +/- 0.3 g, [mean +/- SEM]) were obtained with CC than in controls (10). Clomiphene citrate treatment increased LH/hCG receptor concentrations and the dissociation constant significantly from 69 +/- 12 fmol/mg protein and 1.1 +/- 0.2 x 10(-10) M, respectively, in controls to 112 +/- 6 fmol/mg protein and 2.1 +/- 0.1 X 10(-10) M. Cytosol P and 17-OHP levels were not significantly increased. Cumulatively these cellular effects may be responsible for increasing serum P and responsiveness to hCG and for correcting luteal dysfunction.

Cytosol progesterone (P) and 17 alpha-hydroxyprogesterone (17-OHP) levels and luteinizing hormone/human chorionic gonadotropin (LH/hCG) receptors were measured in 27 corpora lutea and four corpora albicantia. Cytosol P concentrations were highest in corpora lutea (mean +/- SEM, 3.1 +/- 0.8 micrograms/g) during the midluteal phase (days 15 to 19) rather than the early (2.2 +/- 0.8 micrograms/g, days 20 to 25) and late luteal phases (1.8 +/- 0.8 micrograms/g, days 26 to 30). Cytosol 17-OHP concentrations also were 3.3 +/- 0.5, 4.3 +/- 0.6, and 3.3 +/- 1.0 micrograms/g in early, midluteal, and late luteal phases, respectively, and was significantly inversely correlated with occupied LH/hCG receptors in midluteal phase. Corpora albicantia had the lowest P (0.3 +/- 0.05 microgram/g) and 17-OHP (0.9 +/- 0.6 micrograms/g) concentrations. Cytosol P and 17-OHP may therefore reflect the balance between the luteal cell production and secretion, whereas the amount of occupied and unoccupied LH/hCG receptors may partially explain the relationship between LH and P secretion.

The mechanisms controlling luteal function may involve factors that are produced both within the corpus luteum and outside the ovary. The process of luteal control appears to involve a series of molecular species, proteins, peptides, steroids and prostaglandins. Each of these factors may act independently or in concert modifying the actions of one another. The effect of GnRH on luteal function has not been completely examined and thus its significance is unclear. The neurohypophyseal peptides, oxytocin and arginine vasopressin, in combination with LH, prolactin, oestrogens and prostaglandins may play an important regulatory role on the corpus luteum.

Corpora lutea, corpora albicantia, and ovarian stroma from normal human premenopausal ovaries were examined for the presence of oxytocin and neurophysin by using highly specific antisera and peroxidase-antiperoxidase light-microscopic immunohistochemistry. Oxytocin and neurophysin immunoreactivity was found in some but not all cells of the corpora lutea obtained on days 19 to 24 of the menstrual cycle. Stromal tissue and corpora albicantia did not give a positive reaction for either of these peptides, and negative results were also obtained with corpora lutea of mid- and term-pregnancy and preovulatory follicles. Specificity of the immunohistochemical reaction was confirmed by immunoabsorption tests. The specific localization of immunoreactive oxytocin and neurophysin in corpora lutea of the human menstrual cycle directly demonstrates the presence of oxytocin- and neurophysin-positive cells within the human corpus luteum.

Six corpora lutea (day 17 to day 25) of the menstrual cycle and four ovarian stromal tissues from seven cycling women were examined for the presence of epidermal growth factor (EGF) by immunocytochemistry with the use of the indirect peroxidase-antiperoxidase (PAP) procedure. After tissue sections of 4 micron were mounted, endogenous peroxidases were removed with hydrogen peroxide, and the sections were incubated for 1 hour at room temperature followed by 16 hours at 4 degrees C with a highly specific antisera for mouse EGF (mEGF), nonimmunized normal rabbit serum, or antiserum preadsorbed with excess antigen. After the reaction with the second antibody (goat antirabbit IgG) for 1 hour at room temperature, the peptide was localized by use of PAP and 3.3'-diaminobenzidine as the chromogen. EGF could be localized in the luteal cells of five corpora lutea (day 17 to day 25) but not in a corpus luteum of day 22 and all ovarian stromal tissue examined. The localization of immunoreactive EGF in the human corpus luteum demonstrates directly for the first time the presence of this growth-promoting peptide in this tissue, which suggests its local production.

Ovarian tissues (n = 26) obtained at surgery were assayed for oxytocin (OT) concentrations in different parts of the ovary by a specific and sensitive RIA after homogenization and extraction with 0.4 M acetic acid. Chromatography of the extract on a Sephadex G-25 column revealed a single peak identical to synthetic OT, as measured by RIA. Corpora lutea of the menstrual cycle had 10.8-53.0 ng immunoreactive OT/g tissue (n = 7), while those of early pregnancy had a concentration of 106.0 ng/g (n = 1). Ovarian stromal tissue had either undetectable or lower concentrations of OT (0-21.0 ng/g; n = 5) than the corpus luteum from the same ovary. While a luteoma of term pregnancy (n = 1), a benign cystadenoma (n = 2), and an endometriotic cyst (n = 1) had no detectable immunoreactive OT, the concentrations of immunoreactive OT were 20.0 ng/g in a thecoma, 1.4, 20.0, and 60.0 ng/g in preovulatory follicles (n = 3), and 41.0 and 37.0 ng/g in polycystic ovaries (n = 2). In one patient with premature ovarian failure, the ovaries had 9.0 ng/g and undetectable immunoreactive OT. These findings indicate the presence of immunoreactive OT in human ovaries, with significant concentrations in the corpus luteum and preovulatory follicles. It is probable that these tissues produce OTs or an OT-like material which may function as an ovarian luteolytic agent.

Oxytocin concentrations were determined in baboon (Papio anubis) corpora lutea, and the effect of oxytocin on dispersed luteal cell progesterone production was evaluated. Oxytocin concentrations increased significantly from an early luteal phase value of 2.1 +/- 1.1 ng/gm to a peak concentration of 18.1 +/- 4.3 ng/gm wet weight in midluteal phase corpora lutea. Corpora albicantia and ovarian stroma had comparatively low oxytocin concentrations. Reverse-phase high-pressure liquid chromatography of corpora lutea extracts gave a peptide peak (retention time, 17.25 min) similar to a standard oxytocin peak. Plasma oxytocin levels, which were significantly higher in the ovarian vein draining a corpus luteum than in the contralateral side or the femoral vein, declined significantly after luteectomy. Oxytocin was localized by immunocytochemical methods in luteal cells. In the early luteal phase oxytocin (4 to 800 mU; 1 mU is equivalent to 2 ng) inhibited basal and human chorionic gonadotropin-stimulated progesterone production by dispersed luteal cells, but in the late luteal phase 200 to 800 mU oxytocin inhibited only human chorionic gonadotropin-stimulated progesterone output. Oxytocin did not affect luteal cell progesterone production in the midluteal phase. Thus oxytocin is present in corpora lutea, can be localized in the luteal cells, is probably produced locally, and may modulate luteal cell progesterone production.

Immunoreactive oxytocin is detectable in the corpora lutea of women and cynomolgus monkeys by radioimmunoassay. To localize the presence of oxytocin and neurophysin I in ovarian tissues of subhuman primates, three corpora lutea and ovarian stromal tissues and two Fallopian tubes obtained during the menstrual cycle of the baboon and decidua from two pregnant baboons were examined using highly specific antisera against either oxytocin or neurophysin I and peroxidase-antiperoxidase light microscopy immunohistochemistry. Oxytocin-like as well as neurophysin I-like immunoreactivities were found in some cells of all the corpora lutea only, but could not be demonstrated in ovarian stromal tissues, Fallopian tubes and decidua. Specificity of the immunocytochemical reaction was further confirmed by immunoabsorption of the antiserum with excess oxytocin or neurophysin, after which the immunoreactivities for both oxytocin and neurophysin in the luteal tissue were negative. Similar controls using normal rabbit serum gave no positive staining for either oxytocin or neurophysin. Counterstaining of the positive immunoreactivities for oxytocin and neurophysin I with Mayer's haematoxylin and eosin demonstrated clearly that the oxytocin and neurophysin I appeared as granular material mainly within the cytoplasm of the luteal cells. The localization of immunoreactive oxytocin and neurophysin I in the corpus luteum of the baboon demonstrates directly the presence of these two neurohypophysial peptides within primate luteal cells and suggests their local production.

DNA-RNA hybridization has been used to assess the presence of relaxin gene transcripts in human luteal tissues of pregnancy and the menstrual cycle, as well as in the human testis and prostate. The results imply a substantial capacity for hormone biosynthesis in the mid to late luteal phase of the ovary in non-pregnant women. In men the prostate has been shown also to express relaxin gene transcripts, though levels are low. The testis appears negative. The results suggest that functions for relaxin must be sought also outside pregnancy.

To characterize and determine the concentration of LH/hCG receptors in human corpora lutea of the menstrual cycle, we measured occupied and unoccupied receptors and determined the association (Ka) and dissociation (Kd) constants individually in 23 corpora lutea (CL) and 4 corpora albicantia obtained at the time of tubal ligation from 25 normal cycling women. We found no [125I]hCG binding in any of the corpora albicantia. Scatchard plot analysis for each CL revealed a linear binding plot indicative of a single set of LH/hCG receptors. The mean concentration of unoccupied receptors was 36 +/- 10 (+/- SE) fmol/mg protein in the early luteal phase (days 15-19; n = 5), 64 +/- 11 fmol/mg protein in the midluteal phase (days 20-25; n = 13), and 42 +/- 19 fmol/mg protein in the late luteal phase (days 26-30; n = 5). The concentrations of occupied receptors were 56 +/- 8, 46 +/- 6, and 54 +/- 12 fmol/mg protein in the early, mid-, and late luteal phases, respectively. Total (occupied plus unoccupied) receptor concentrations reached maximum levels of 110 +/- 11 fmol/mg protein in the midluteal phase. Ka increased progressively from 12 +/- 4 X 10(9) mol/L-1 in the early luteal phase to 19 +/- 7 X 10(9) and 21 +/- 8 X 10(9) mol/L-1 in the mid- and late luteal phases. We conclude that in normal CL, 1) total and unoccupied LH/hCG receptor levels parallel progesterone secretion; 2) changes in the binding affinity may be important in sustaining and/or rescuing the CL; and 3) loss of LH/hCG receptors is probably related to luteolysis.

Recently, using a highly specific radioimmunoassay, we have demonstrated that the concentration of oxytocin in the corpus luteum of the human and cynomolgus monkey are several fold higher than in the peripheral circulation. In this study, we have examined the corpora lutea and ovarian stroma from the ovaries of normal adult cycling baboons (Papio anubis) for the presence of oxytocin through the use of immunocytochemical procedures. Tissues obtained at laparotomy were fixed in Bouin's solution and embedded in paraffin; immunoreactive oxytocin was localized with peroxidase-antiperoxidase and 3.3' diaminobenzidine. Six corpora lutea with stroma were obtained--two each from the early (Day 14-20), mid-(Day 21-24), and late (Day 25-30) stages of the luteal phase. Immunoreactive oxytocin was localized in all corpora lutea examined but was absent from all stroma samples. Larger areas of the corpus luteum from the mid-luteal phase showed staining for oxytocin, and the intensity of staining for this peptide was maximal in this phase of the cycle.

Six human corpora lutea (day 17-25) of the menstrual cycle and 4 ovarian stromal tissues from 7 cycling women were examined for the presence of the hormone, prolactin, by immunohistochemistry using the indirect peroxidase-antiperoxidase method. After mounting tissue sections of 4 micron, endogenous peroxidases were removed with hydrogen peroxide and the sections were incubated for 1 h at room temperature followed by 16 h at 4 degrees C with a highly specific antisera for human prolactin, nonimmunized normal rabbit serum for a control reaction, or antiserum preadsorbed with excess human prolactin for specificity determination. Following the reaction with the second antibody (goat antirabbit IgG) for 1 h at room temperature, prolactin was localized using peroxidase anti-peroxidase and 3.3'-diaminobenzidine as the chromogen. Prolactin was present and could be localized in the luteal cells of all 6 corpora lutea, but not in any of the ovarian stroma studied. Human adenohypophysis served as a positive tissue control for prolactin immunopositive staining. The localization of immunoreactive prolactin in the corpus luteum demonstrates directly the presence of this hormone in the human ovary, adding further evidence for its role in luteal function.

Epidermal growth factor receptors are present in many reproductive tissues but have not been demonstrated in the human corpus luteum. To determine the presence of epidermal growth factor receptors and its binding characteristics, we carried out studies on the plasma cell membrane fraction of seven human corpora lutea (days 16 to 25) of the menstrual cycle. Specific epidermal growth factor receptors were present in human corpus luteum. Insulin, nerve growth factor, and human chorionic gonadotropin did not competitively displace epidermal growth factor binding. The optimal conditions for corpus luteum-epidermal growth factor receptor binding were found to be incubation for 2 hours at 4 degrees C with 500 micrograms plasma membrane protein and 140 femtomol 125I-epidermal growth factor per incubate. The number (mean +/- SEM) of epidermal growth factor binding sites was 12.34 +/- 2.99 X 10(-19) mol/micrograms protein; the dissociation constant was 2.26 +/- 0.56 X 10(-9) mol/L; the association constant was 0.59 +/- 0.12 X 10(9) L/mol. In two regressing corpora lutea obtained on days 2 and 3 of the menstrual cycle, there was no detectable specific epidermal growth factor receptor binding activity. Similarly no epidermal growth factor receptor binding activity could be detected in ovarian stromal tissue. Our findings demonstrate that specific receptors for epidermal growth factor are present in the human corpus luteum. The physiologic significance of epidermal growth factor receptors in human corpus luteum is unknown, but epidermal growth factor may be involved in intragonadal regulation of luteal function.

To determine and compare the direct effects of prostaglandin F2a (PGF2a) and human chorionic gonadotropin (hCG) on luteal cell progesterone production in vitro, 9 human corpora lutea obtained at tubal ligation were minced and treated with collagenase to disaggregate luteal cells. Dispersed luteal cells (80% viable) were incubated in air at 37 degrees C in a shaking water bath for 3 h and total progesterone in the media and cells was determined by radioimmunoassay. Optimum progesterone production was obtained using 25,000 or more cells per incubate and an incubation time of 2-4 h. hCG-stimulated progesterone production increased significantly with 0.01 IU to as high as 100 IU. In the early luteal phase (days 1-5 post ovulation or days 15-20 of the luteal phase), PGF2a (10-1000 ng) significantly inhibited progesterone production but significantly stimulated progesterone production in the mid-luteal phase (days 21-25). PGF2a had no effect on luteal cell progesterone production in the late luteal phase (days 26-30). This age-dependent direct effect of PGF2a on human luteal cell progesterone production in vitro indicates a role for PGF2a in the total intragonadal regulation of progesterone output, possibly through a paracrine or autocrine manner directed towards synchronizing luteal progesterone secretion and endometrial preparation for nidation.

To determine if oxytocin (OT) is present in cynomolgus monkey corpus luteum, OT was measured by a specific and sensitive RIA in 13 corpora lutea, ovarian venous plasma on the ipsilateral side and peripheral venous plasma at different stages of the luteal phase. Serial dilution of acetic acid extract of the corpus luteum showed parallelism with standard OT in the RIA. Total content of OT in corpus luteum was 1.9 +/- 0.5 ng (mean +/- SEM) with a content of 0.4-0.8 ng in early luteal phase, 1.0-6.2 ng in midluteal phase, and 0.4-0.7 ng in late luteal phase. OT concentrations in corpus luteum were 21.0-75.2 ng/g wet wt in early luteal phase, increasing to 34.4-602.5 ng/g in midluteal phase; and declining to 3.4-117.4 ng/g in late luteal phase. OT concentrations per mg protein in the corpus luteum were 0.05-19.6 ng with peak concentrations of 14.7-19.6 ng/mg protein on day 22. Sephadex G-25 column chromatography of the corpus luteum extract revealed a single peak for binding activity similar to that of synthetic OT on the RIA. Ovarian vein blood from the same side as the corpus luteum had a significantly higher OT concentrations of 161.2 +/- 29.7 pg/ml on days 15-24 than 16.8 +/- 3.6 pg/ml on days 25-28 (P less than 0.01) and peripheral plasma OT levels of 23.2 +/- 3.4 pg/ml (P less than 0.025). Our findings indicate that OT is present and probably produced by monkey corpus luteum with peak OT concentrations found in midluteal phase. Thus OT may play a role in primate corpus luteum function.

To determine whether the human corpus luteum is a source of relaxin and oxytocin, we measured the concentrations of these peptides in plasma obtained from the ovarian veins of ovaries with and without a corpus luteum and compared these to peripheral plasma levels. Peripheral and ovarian venous blood samples were obtained from 34 nonpregnant women, 13 during the luteal phase and 21 during the follicular phase of their cycles, and from a 6-week pregnant woman. Plasma relaxin, oxytocin, and progesterone concentrations were determined by sensitive and specific RIAs. Plasma relaxin levels were not detectable (less than 0.16 microgram/L) in peripheral or ovarian venous plasma not draining a corpus luteum. The mean relaxin concentration in plasma draining an ovary with a corpus luteum was 0.41 +/- 0.09 (+/- SE) microgram/L. Oxytocin levels also were significantly higher in plasma draining an ovary with a corpus luteum (6.70 +/- 1.86 pmol/L) than in that draining the ovary with no corpus luteum (1.58 +/- 0.09 pmol/L; P less than 0.01) or in peripheral plasma (1.58 +/- 0.09 pmol/L; P less than 0.025). The mean progesterone concentration also was highest in plasma draining an ovary with a corpus luteum (210.2 +/- 50.5 nmol/L) compared with those in plasma from the contralateral ovarian vein (40.3 +/- 16.5 nmol/L P less than 0.005) and peripheral plasma (30.2 +/- 5.7 nmol/L; P less than 0.005) during the luteal phase. In a woman who was 6 weeks pregnant, plasma draining the ovary with a corpus luteum had 1.9 micrograms relaxin/L, but only 0.49 pmol/L oxytocin; the latter was similar to concentrations in noncorpus luteum-bearing ovarian venous plasma. These findings indicate that the human corpus luteum secretes relaxin, oxytocin, and progesterone. Both ovarian oxytocin and relaxin may function as paracrine or autocrine modulators of luteal function.