In vitro studies of preimplantation embryos from a number of mammalian species have shown that the oviduct and uterus contain growth factors that stimulate cellular proliferation and differentiation of preimplantation embryos. The mammalian preimplantation embryo was first viewed as an autonomous entity, due to the ease with which mouse embryos could be cultured from the two-cell stage on to the blastocyst. This view changed following studies in other species, notably domestic animals, which revealed the presence of 'blocks' to development, when embryos were cultured in vitro. Another line of evidence leading to the view that the maternal environment is crucial for optimal development, was the finding that embryos cultured in vitro lag developmentally behind their in vivo counterparts. This developmental retardation could be ameliorated when washings from the reproductive tract or specific growth factors were added to the media. The expression of genes for growth factors and their receptors is regulated in a tissue-specific manner as well as temporally and spatially during mammalian development. In this review, information regarding the expression and role of polypeptide growth factors of the insulin family during preimplantation mammalian development is summarized.
Leukemia inhibitory factor (LIF) is a pleiotropic cytokine essential for uterine blastocyst implantation. We investigated human LIF production and action on human chorionic gonadotropin (hCG) secretion at different stages of gestation. Decidual and chorionic tissues were obtained at the first and second trimesters of pregnancy, and at term. Decidual and trophoblast cells were isolated and cultured separately and medium LIF or hCG levels were measured by radioimmunoassay. Decidual cells derived from the first and second trimester, and term were cultured for 72 h in serum-containing medium and produced 374 +/- 162 (mean +/- SEM), 140 +/- 14 and 466 +/- 134 ngLIF/mg protein, respectively. Pregnant mare serum gonadotropin (10 U/ml) did not affect LIF production from first-trimester decidual cells, but inhibited second-trimester LIF production to 37% of that of controls (p < 0.05), and stimulated LIF production from term decidual cells to 125% of that of controls (p < 0.05). First-trimester trophoblast cells treated with 10 nmol/l hLIF for 72 h in serum-free medium increased hCG secretion by 125% (p = 0.05). Conversely, hLIF dose-dependently inhibited basal and cAMP-stimulated hCG secretion in trophoblasts derived during the second trimester and at term, as well as in cultured JEG-3 choriocarcinoma cells. LIF exhibited more potent inhibitory action in the second trimester, maximally reducing hCG production by 37% from that of control values (p < 0.05) at a dose of 10 nmol/l. These findings indicate that (1) LIF is produced by decidual cells throughout pregnancy and may play a different regulatory role of hCG production at different stages of gestation; and (2) gonadotropin synthesized in trophoblasts may, in part, regulate LIF production through a paracrine pathway.
The goal of these studies was to correlate sonographic evidence of pregnancy during the peri-implantation period with the timing of the rise in monkey chorionic gonadotropin (mCG) as measured with an enzyme-linked immunosorbent assay. Animals were time-mated at mid-cycle, and ultrasound examinations were performed on postovulation days 12-15 (n = 77). Pregnancy was sonographically identified in 48 of 77 animals (62.3%), of which 28 had correlative ultrasound/endocrine data collected. For these animals, blood samples were obtained on postovulation days 12-15 for mCG assay. Pregnancy was identified by ultrasound on postovulation days 12 (6/28; 21.4%), 13 (6/28; 21.4%), 14 (8/28; 28.6%) or 15 (8/28; 28.6%). Seven of the 28 (25.0%) were found to have mCG levels consistent with pregnancy (> or = 1 ng/ml) on the same day as ultrasound confirmation, 12 of 28 (42.9%) were sonographically detected as pregnant 1 (n = 6), 2 (n = 3) or 3 (n = 3) days earlier than by mCG, and nine of 28 (32.1%) were found to have elevated mCG levels 1 (n = 7), 2 (n = 1) or 3 (n = 1) days earlier than ultrasound confirmation of pregnancy. The results of these studies have demonstrated (1) the utility of anatomical and endocrine techniques for detecting pregnancy approximately 3 days after the onset of implantation, and (2) the variation in the timing of implantation and the rise in circulating mCG in individual animals.
During pregnancy excess corticosteroid exposure can disturb the normal pattern of growth and differentiation of the primate fetus. This is normally prevented by the action of 11 beta-hydroxysteroid dehydrogenase (11 beta-HSD), which converts cortisol to its biologically inactive 11-oxo form, thereby ensuring that little or no cortisol is transferred to the fetus. During implantation, extravillous trophoblasts breech uterine vessels that are embedded in a decidual cell matrix. Through this invasive process the embryo gains requisite access to the maternal blood supply, while risking exposure to high circulating glucocorticoid levels. Thus, the expression of 11 beta-HSD by the decidual cell layer may be essential in regulating cortisol exposure of the developing embryo prior to placentation. In order to investigate the potential contribution of decidual cells to glucocorticoid metabolism, we evaluated the expression of both known 11 beta-HSD isoforms, 11 beta-HSD1, whose catalytic activity is NADP(+)-dependent, and NAD(+)-dependent 11 beta-HSD2, during decidualization of monolayers of human endometrial stromal cells. The differential actions of ovarian steroids on human endometrium are simulated in this in vitro model. Thus, progestins induce the expression of several decidualization markers in the cultured stromal cells, and consistent with its priming action in vivo, estradiol augments this expression. The results of our studies established a link between in vitro decidualization and enhanced glucocorticoid metabolizing capacity. Accordingly, the catalytic activities of both 11 beta-HSD isoforms were enhanced by incubation of the precursor stromal cells with medroxyprogesterone acetate, and further enhanced by estradiol, despite a lack of response to estradiol alone. This differential response to estradiol and progestin was reflected in parallel changes in steady state levels of 11 beta-HSD1 messenger RNA. The role of glucocorticoid metabolizing activity of the decidual cell is discussed in terms of its implications in determining the exposure of the implanting embryo to biologically active glucocorticoids.
OBJECTIVES To compare the efficacy and the reproductive outcomes of Tamoxifen (TMX) vs. clomiphene citrate (CC) treatment in infertile women with luteal phase dysfunction. METHODS Eighty-seven consecutive infertile women with luteal phase dysfunction were randomly assigned to two treatment groups as: group 1 (TMX followed by CC) and group 2 (CC followed by TMX). The medications administered in the initial and the subsequent cross-over phase were given for 4 to 6 cycles except in those patients who achieved pregnancy within the first three cycles of treatment. The rates of pregnancy, spontaneous abortion and live-birth were calculated for comparison. Demographic profile and the frequency of other infertility factors between the two groups were comparable. RESULTS The pregnancy rates were similar between TMX and CC treatment (initial treatment phase: TMX, 50.0% vs. CC, 56.4%; subsequent treatment phase: CC, 33.3% vs. TMX, 47.1%), but the abortion rates were significantly lower (p < 0.05) in TMX treatments (initial treatment phase: 8.3% and subsequent treatment phase: 12.5%) than that of CC treatments (initial treatment phase: 22.7% and subsequent treatment phase: 62.5%). When the rates were calculated by pooling the patients according to the medication they received, the miscarriage rate was significantly lower (p < 0.05) if the patients conceived during the TMX treatment (9.4% vs. CC, 33.3%). CONCLUSIONS Tamoxifen and clomiphene citrate seem to be similarly effective in achieving pregnancy in patients with luteal phase dysfunction. However, Tamoxifen therapy is associated with a lower incidence of spontaneous abortion; thus Tamoxifen may be the better choice for the therapy of infertile women with luteal phase dysfunction.
The influence of maternal exogenous growth hormone treatment on in utero conceptus development was evaluated in the rat. The periods of response and stimulation of DNA synthesis on embryo/fetal and placental tissues were assessed by subcutaneous injections of ovine growth hormone (oGH) preparations during pregnancy days 11-15, autopsied on day 16; and during pregnancy days 11-20 and 16-20, autopsied on day 21. To determine DNA biosynthesis potential, thymidine (methyl-3H) was administered through the jugular vein 14-16 h prior to sacrifice. DNA content and uptake of radiolabeled thymidine into DNA were analyzed for whole embryos on day 16, and for fetal liver, brain and remaining body tissues on day 21 of pregnancy. Placental tissues from oGH-treated mother and controls were also quantified for DNA content and radiolabeled thymidine uptake. oGH treatment produced a significant increase (p < 0.05) in radiolabeled thymidine uptake into DNAs of different fetal organs compared to saline-treated controls matched for weight and litter number during the latter part of gestation (fetal histogenesis period; pregnancy days 16-20). The stimulatory influence of maternal growth hormone treatment on DNA contents and radiolabled thymidine uptake on placental tissues at this period of gestation was also significantly different from that of the controls. Rat conceptus tissues (embryos and placentas) during the organogenesis period of early gestation (days 11-15) appeared to be unresponsive to such treatment. Thus, these results suggest that maternal growth hormone influences conceptus growth during the latter part of gestation and activation of placental functions may be an important aspect of stimulation of cell proliferation in the rat fetus.
OBJECTIVE:To evaluate the long-term effects of pregnancy and bromocriptine treatment on prolactin-secreting pituitary tumors in women undergoing infertility treatment for prolactinomas.METHODS:The records of 17 patients with prolactinomas were reviewed. Data regarding age, prepregnancy baseline and postpartum serum prolactin levels, and radiologic studies including CT or MRI were assessed. 16 patients were treated with bromocriptine before achieving pregnancy. Bromocriptine therapy was resumed after delivery for the duration of 1 to 14 years.RESULTS:45% of pregnancies did not affect the size of prolactinomas, 27% of pregnancies showed a decrease in size of prolactinomas or radiologic evidence of resolution of the tumor and 27% of pregnancies demonstrated radiologic increase in the size of prolactinomas.CONCLUSIONS:It is safe for patients with prolactinomas to achieve pregnancy following bromocriptine treatment. Pregnancy may lead to a slight decrease in the size of prolactinomas, increase in size, no change, and in some cases, complete resolution. There were no visual field changes during the pregnancies. Pregnancy and long-term bromocriptine treatment may be helpful in reduction of the size of the tumor.
There are potential interactions between various human chorionic gonadotropin (hCG) isoforms at the level of luteinizing hormone/chorionic gonadotropin (LH/CG) receptor. The objective of this study was to characterize the receptor-binding activities of the primary peptide variants of hCG including intact hCG, free beta subunit, beta-core fragment and nicked hCG, and to test the effects of these hCG variants on the binding of intact hCG. A radio-receptor assay based on cell membranes expressing recombinant human LH/CG receptors was validated and used in this study to avoid species differences in the receptor-binding specificity. The results showed that none of the hCG variants that we studied had sufficient binding affinity to compete with binding of intact hCG, nor were they able to antagonize the binding of intact hCG. These results suggest that hCG variants with either abbreviated polypeptide structures or incomplete peptide linkage are products or metabolites which do not have the tropic biological activity of the whole hormone, the intact heterodimeric hCG.
Interleukin-10 (IL-10) is a T-helper type-2 (Th2) cytokine noted for its ability to suppress cytokine synthesis by T-helper type-1 (Th1) cells. IL-10 may play a role in pregnancy immunotolerance through the establishment of a Th2 cytokine bias at the maternal-fetal interface. This study examines the expression and production of IL-10 by normal and malignant human trophoblast. Term placental biopsies, cloned choriocarcinoma cell lines and isolated human trophoblast were utilized for the study of IL-10 expression. Choriocarcinoma cells (BeWo, JEG-3, JAR) were maintained in T-flask culture until confluence and then harvested by enzymatic dispersion. Purified term trophoblast were obtained by sequential trypsin/DNAse digests and CD9 immunoaffinity chromatography. Amplified IL-10 mRNA was detected by a reverse transcriptase polymerase chain reaction (RTPCR) technique. BeWo cells were maintained in artificial capillary culture (ACC) and conditioned media assayed for IL-10. Granulocyte macrophage-colony stimulating factor (GM-CSF; 1.0, 10.0 and 100.0 ng/ml) was added to the BeWo cultures to examine its effects on trophoblast IL-10 production. IL-10 determinations were performed using a human ELISA system. IL-10 mRNA was detected in each trophoblast cell type examined with the exception of the JEG-3 choriocarcinoma cell line. IL-10 protein was also detected (range 6-22 pg/ml) in BeWo media on days 8 to 11 of culture. When serum was reduced in the culture media, IL-10 levels fell below the sensitivity of the assay (5 pg/ml). Subsequent addition of GM-CSF stimulated BeWo IL-10 secretion in a dose-related fashion. These results support the concept IL-10 is expressed at the human maternal-fetal interface, and production of this important immunoregulatory molecule may be regulated, in part, by GM-CSF.
Low serum estradiol concentrations during pregnancy have been associated with spontaneous abortions. A large proportion of abortuses have chromosomal abnormalities. To determine whether serum estradiol during the luteal phase of conception cycles can predict the karyotype of spontaneous abortions, serum estradiol samples were drawn 11 days after embryo transfer from 52 women who subsequently had spontaneous abortions and chromosomal analysis was performed on the products of conception. The frequency of estradiol levels < 100 pg/ml was compared between normal and abnormal karyotypic spontaneous abortions. Chromosomal analyses among 52 spontaneous abortions revealed 14 (27%) to be normal and 38 (73%) abnormal. Among the 38 karyotypic abnormalities 30 were aneuploidy and eight were polyploidy. Serum estradiol levels were < 100 pg/ml in eight women (15%) and > 100 pg/ml in 44 women (85%). The frequencies of estradiol < 100 pg/ml in spontaneous abortions with normal (7%) and abnormal (18%) karyotypes were not significantly different. Thus, serum estradiol cannot predict the karyotype of spontaneous abortions.
11 beta-Hydroxysteroid dehydrogenase (HSD) activity was measured in freshly frozen granulosa cells isolated from follicles of twenty-one infertility patients undergoing in vitro fertilization-embryo transfer (IVF-ET). A total of 213 follicles were analyzed for 11 beta-HSD activity. Both nicotinamide-adenine dinucleotide (NAD) and nicotinamide-adenine dinucleotide phosphate (NADP) dependent 11 beta-dehydrogenase activities were measured in granulosa cells. The activity in reductive direction (11-oxoreductase activity) was not measurable either with NADH or NADPH as cofactors. NAD- and NADP-dependent dehydrogenase activities are in comparable levels at 100 nmol/l and 1 mumol/l corticosteroid concentrations. For comparing enzyme activities of individual follicles, significant enzyme activity was considered to be a level of > 2 nmol/l/min/mg. 41.3% of the follicles demonstrated enzyme activity, 58.7% did not. The mean 11 beta-dehydrogenase (11 beta-DH) activity was calculated for each patient. Those patients with levels > 2 nmol/l/min/mg were considered enzyme positive; those with mean levels < 2 nmol/l/min/mg were considered negative. No significant association was noted between follicle size, oocyte maturity or fertilization rates and 11 beta-DH activity. This study noted the presence of 11 beta-DH activity in granulosa cells, however, no association with oocyte maturity and fertilization was found.
This paper is a summary of three oral presentations, as well as the ensuing discussion, at the Rijeka/Opatija 3rd Alps Adria Immunology meeting by three members of the European Biomed group on vertical transmission of HIV (G. Chaouat, F. Barre-Sinoussi, G. Scarlatti). This group also involves the laboratories of D. Dormont (CEA, Fontenay aux roses, France), P. Gounon (Electron Microscopy, the Pasteur Institute, France; Irène Athanassakis, University of Crete, Greece; Eva Maria Fenyö, Karolinska Institute, Sweden; and Larry Guilbert, Canada). As such, this paper intends to be neither a review, nor an original article, but rather is an opinion paper discussing the working hypothesis of this network, as well as some of their recent results, which were presented at this meeting. The paper was issued at the request of the organizers of the meeting.
We previously showed that a significant number of failing pregnancies are associated with production of human chorionic gonadotropin (hCG) having relatively low bioactivity. The present study was designed to compare the secretion of intact, immunoreactive hCG to the secretion of bioactive hCG during trophoblast differentiation, and to test the hypothesis that the lower bioactive: immunoreactive hCG ratios in failing pregnancies are related to reduced or impaired trophoblast differentiation. Cytotrophoblast cells were isolated from term placentas and cultured under conditions that induced or did not induce syncytiotrophoblast formation. Culture media were collected at regular intervals up to 72 h and levels of immunoreactive and bioactive hCG were measured. The differentiation of cytotrophoblast cells to multinucleated syncytiotrophoblast was monitored by immunocytochemistry and electron microscopy. During the 72 h culture period, concentrations of immunoreactive and bioactive hCG increased in both differentiating and non-differentiating cells. However, the concentrations of immunoreactive and bioactive hCG were higher under culture conditions that promoted trophoblast differentiation. Furthermore, the ratio of bioactive hCG to immunoreactive hCG was higher in differentiating cultures. When differentiation was inhibited by dimethyl sulfoxide, the secretion of bioactive hCG was reduced and the bioactive: immunoreactive hCG ratio did not change. These findings are consistent with the idea that production of bioactive hCG accompanies syncytiotrophoblast formation.
The expression pattern of c-fos, c-jun, c-kit and stem cell factor (SCF) has been investigated in developing human placenta using the highly sensitive technique of in situ reverse transcriptase-polymerase chain reaction (RT-PCR). Specific transcripts of all genes under study were observed in first-trimester placenta sections. c-fos, c-jun, c-kit and SCF transcripts were localized in cells of the villous stroma; fos, jun and kit-specific mRNAs were also found in endothelial cells; fos, kit and SCF mRNAs were detected in villous trophoblast cells. In mid-trimester and term placenta specimens only SCF transcripts were observed, restricted to trophoblast cells. The lack of c-fos transcripts in placenta from the second and third trimesters is a finding that contrasts with data from the literature obtained using extractive techniques. Parallel immunocytochemistry of placenta specimens from the three pregnancy stages under study revealed the fos protein only in first-trimester placenta, in agreement with the in situ RT-PCR findings. We conclude that the in situ RT-PCR technique is most suitable for gene expression studies because of its high level of sensitivity in correctly assigning the signal to specific cell types in complex tissues.
In this study circulating human chorionic gonadotropin (hCG) levels in the peri-implantation period of natural cycles of normal women were characterized. The hypothesis that the bioactivity of hCG in abnormal pregnancies is different from that in normal pregnancies was tested daily through serum hCG measurements in two immunoenzymometric assays, a radioreceptor assay and a human luteinizing hormone/human chorionic gonadotropin--cyclic adenosine monophosphate--luciferase bioassay. In normal pregnancies, the levels of immunoreactive and bioactive hCG were higher than in abnormal pregnancies (p < 0.05). In addition, the slopes of the rise in concentration of hCG as measured in all four assays were significantly greater for normal pregnancies than for abnormal pregnancies (p < 0.01). The rate of rise of hCG bioactivity was significantly greater than the rise of immunoreactivity or receptor-binding activity in normal pregnancies, but this difference was not observed in abnormal pregnancies. These results indicate that: (1) A steep rise of bioactive hCG is a consistent feature of the peri-implantation period of normal pregnancies. (2) Abnormal pregnancies had a deficiency in the production of hCG, which was reflected in both the hCG levels and the rate of increase, even in the first week of implantation. (3) Some abnormal pregnancies produced hCG with lower biological activity than the hCG of normal pregnancies.
Current in vitro fertilization techniques (IVF) including intracytoplasmic sperm injection (ICSI), microepididymal sperm aspiration (MESA) or testicular sperm extraction (TESE) clearly prevent any spontaneous choice of ova or spermatozoa. According to the widely admitted concept of gamete selection, pregnancies following IVF, when compared to natural fertilization, could therefore present a higher risk of genetic anomalies. However, no increased fetal or newborn abnormalities are noticed with IVF, except perhaps for sex chromosome aneuploidies. Data from the literature support the view that the uterus is, indeed, the organ where selection mechanisms occur (when they do so), as suggested by Carr in 1971. This selection concerns mainly autosome imbalances; unbalanced conceptuses are aborted. Sex chromosome aneuploidies, apparently, are less prone to natural abortion, but their higher rate of occurrence, as reported in a few series of studies, does not seem to be associated with the IVF procedures.
Transport of glucose into the cell is catalyzed by glucose transporters (Glut). Glut1 and Glut3 are expressed at various levels in many human tissues, including the placenta. It has been reported that ambient glucose can affect both glucose transport activity and expression of the Glut genes, and protein. To date, very few studies concerning Glut in the placenta have been published, and studies in vivo in human diabetic pregnancy are lacking. We therefore investigated placental Glut1 and Glut3 mRNA by Northern blot analysis in ten diabetic (five insulin dependent diabetes mellitus (IDDM), two non-insulin dependent diabetes mellitus (NIDDM) and three gestational diabetes mellitus (GDM)) and nine non-diabetic women. The quantitative results of specific mRNA/beta-actin ratios were expressed as arbitrary units. The results were evaluated according to metabolic and clinical findings. Glut1 and Glut3 mRNA values in diabetic and non-diabetic pregnant women were similar. The metabolic environment seems to affect the Glut3 mRNA levels in IDDM pregnant women but not the control women. In addition, Glut3 mRNA decreased in late pregnancy in the diabetic but not in the control women. Moreover, Glut1 mRNA levels were correlated with maternal age in the diabetic as well as in the control women (significantly). Finally, an inverse correlation was found between Glut1 mRNA levels and placental weight (in both diabetic and non-diabetic women). These results, although preliminary, shed some light on the function of these glucose transporters in normal as well as in diabetic pregnancies and prompt us to carry out a further investigation to better elucidate fetomaternal metabolic correlation at the placental level.
The association of 'early pregnancy factor' ('EPF') activity in early pregnancy sera with multiple fractions of heterogeneous molecular mass was re-examined to test whether previously perplexing observations could be explained by a new multi-factorial model of the serum components required for this activity expression. Gel permeation fractionation of human pregnancy sera revealed 'EPF' activity associated with fractions containing components ranging in MW from < or = 1 kDa to > or = 500 kDa. A significant activity peak was observed eluting on the total volume (Vt) of the column, indicating the presence of active molecules of very low molecular mass. Multiple activity peaks were also observed in the macromolecular fractionation region ranging in apparent molecular weight from 12 kDa, through 25, 70 and 250 to > or = 500 kDa. Analysis of these fractions revealed that they all contained thioredoxin and active moieties of low molecular mass, with the latter probably directly associated with the former. Adsorption with specific anti-thioredoxin antibodies removed from these fractions the capacity to display 'EPF' activity. Further analyses revealed that in these fractions thioredoxin played a permissive role allowing the low molecular mass active moieties to express activity in the bioassay in the presence of otherwise counteracting substances. The results of these studies are consistent with the proposal that 'EPF' activity expression in pregnancy sera is due to the presence of a multi-factorial system in which thioredoxin plays an essential permissive role in concert with active moieties of low molecular mass.
The reverse transcription-polymerase chain reaction/Southern, Northern and Western blottings demonstrated that 19-day-old fetal rat brains contained the luteinizing hormone (LH)/human chorionic gonadotropin (hCG) receptor transcripts and receptor protein. Further investigation with Western blotting demonstrated that diencephalon, mesencephalon, rhombencephalon and telencephalon and several areas in each of these regions contained the LH/hCG receptor protein. The receptor protein levels in different regions were somewhat variable but they were not different from each other or from the whole brain. The 15-day-old fetal rat brains contained lower receptor protein levels which increased by 3-fold in 17-day-old and by 12-fold in 19-day-old fetal rat brains with no further significant increase in 21-day-old fetal, neonatal or adult rat brains. In summary, fetal rat brains contained LH/hCG receptors and these receptors were developmentally regulated. These findings suggest that the gonadotropins LH and hCG may regulate growth, development and differentiation in fetal brain.
