Search

Context: Human gonads arise as a pair of epithelial ridges on the surface of intermediate mesoderm (IM)-derived mesonephros. Toxic environmental factors and mutations in various genes are known to disturb normal gonadal development, but because of a lack of suitable in vitro models, detailed studies characterizing the molecular basis of the observed defects have not been performed., Objective: To establish an in vitro method for studying differentiation of bipotential gonadal progenitors by using human embryonic stem cells (hESCs) and to investigate the role of bone morphogenetic protein (BMP) in gonadal differentiation., Design: We tested 17 protocols using activin A, CHIR-99021, and varying durations of BMP-7 and the BMP inhibitor dorsomorphin. Activation of activin A, WNT, and BMP pathways was optimized to induce differentiation., Setting: Academic research laboratory., Main Outcomes Measures: Cell differentiation, gene expression, and flow cytometry., Results: The two most efficient protocols consistently upregulated IM markers LHX1, PAX2, and OSR1 at days 2 to 4 and bipotential gonadal markers EMX2, GATA4, WT1, and LHX9 at day 8 of culture. The outcome depended on the combination of the duration, concentration, and type of BMP activation and the length of WNT signaling. Adjusting any of the parameters substantially affected the requirements for other parameters., Conclusions: We have established a reproducible protocol for directed differentiation of hESCs into bipotential gonadal cells. The protocol can be used to model early gonadal development in humans and allows further differentiation to mature gonadal somatic cells., (Copyright © 2017 Endocrine Society)

The process of germ cell depletion in patients with Klinefelter syndrome (KS) is incompletely characterized. In the current work, we evaluated the presence of germ cells in adolescent boys with KS for possible future use in assisted reproduction techniques. Fourteen nonmosaic 47,XXY boys (aged 10-14 yr) were enrolled. Every fourth month their puberty was staged, and serum was obtained for hormone analyses. Each boy underwent a single testicular biopsy. Biopsy specimens of seven peripubertal boys (testicular volume < 2.0 ml) had spermatogonia of adult type, whereas older boys with larger testes (> 2.0 ml) exhibited no germ cells. No meiotic germ cells were detectable in any of these subjects. Depletion of germ cells was associated with an increase in testicular volume but was not immediately reflected in levels of serum gonadotropin, inhibin B, or anti-Müllerian hormone. In contrast, hypergonadotropism and suppression of serum inhibin B and anti-Müllerian hormone developed later, during midpuberty, after an unequivocal increase in serum testosterone (>2.5 nmol/liter) levels and degeneration of Sertoli cells. In conclusion, these prepubertal and early pubertal boys with KS had diploid germ cells that vanished in early puberty when testicular volume increased, whereas serum gonadotropin and inhibin B levels displayed pathological changes later during midpuberty.

Infertility caused by ovarian failure is a characteristic feature in Turner's syndrome. Spontaneous pregnancies are seen in 2-5% of these women, and up to 30% have at least some pubertal development, indicating the presence of follicles in their ovaries in adolescence. It has not been clear at which age the follicles disappear. We analyzed the numbers and densities of follicles in ovarian cortical tissue from nine adolescent girls with Turner's syndrome who came to our clinics after having been informed about the study, with an aim to preserve ovarian tissue for possible infertility treatment later in life. A quarter to one whole ovary was laparoscopically removed for the procedure. Follicles were seen in the biopsy tissue in eight of nine subjects from whom ovarian tissue was laparoscopically obtained, the highest numbers being seen in the youngest girls and in those with mosaicism. In one 17-yr-old girl, no ovarian tissue was found. Follicle density was correlated with serum levels of FSH; individuals with the lowest FSH levels had the highest follicle density. One to 190 follicles were counted in the approximately 0.1-2.0 mm(3) of tissue analyzed, giving a density of 1.5-499 follicles/mm(3) of ovarian cortical tissue. Girls up to the age of 17 had primordial follicles in their ovaries. Three girls, two aged 15 yr and one aged 19, had only secondary follicles, with many being atretic. Our finding that adolescent girls with Turner's syndrome still have follicles in their ovarian cortical tissue raises the possibility of future fertility through cryopreservation of ovarian tissue. However, before such procedures can be recommended for clinical management, it is essential that future studies be performed to determine whether the oocytes retrieved from girls with Turner's syndrome have a normal chromosomal complement.

Growth differentiation factor 9 (GDF-9) is a transforming growth factor-beta family member that is required for normal folliculogenesis in female mice, but its role as a regulator of human fertility is still unclear. We determined here by in situ hybridization and immunohistochemical analyses the localization of the GDF-9 messenger ribonucleic acid (mRNA) and protein during human folliculogenesis. The GDF-9 transcripts were not detected in primordial follicles, but they are abundantly expressed in primary follicles in frozen sections of ovarian cortical tissue material obtained at laparoscopic surgery. We raised antipeptide antibodies against GDF-9 and showed by immunohistochemical studies on paraffin sections of whole human ovaries that the GDF-9 protein is most abundantly expressed in primary follicles. We recently demonstrated that a novel GDF-9-related factor, GDF-9B, is coexpressed with GDF-9 during murine folliculogenesis. We now isolated human GDF-9B complementary DNA and genomic clones and report the unusually restricted expression pattern of human GDF-9B. The human GDF-9B transcript can be detected only in the gonads by RT-PCR analysis, and in situ hybridization studies indicate that it is not expressed in small primary follicles but, rather, in the oocytes of late primary follicles. Functional studies using the Xenopus laeuis embryo model indicate that unlike the transforming growth factor-beta family members activin and bone morphogenetic protein-4, neither GDF-9 nor GDF-9B affects mesoderm induction, suggesting that they may use signaling pathways distinct from those well defined for activin and bone morphogenetic protein-4. We conclude that 1) both GDF-9 mRNA and protein are abundantly expressed in oocytes of primary follicles in human ovary, suggesting that the GDF-9 transcript is translated at this early stage of folliculogenesis; 2) human GDF-9B is specifically expressed in gonads at low levels; and 3) the expression of GDF-9 mRNA begins slightly earlier than that of GDF-9B in the human oocytes during follicular development. Our results are consistent with the suggestion that GDF-9 and GDF-9B may regulate human folliculogenesis in a manner specific to the ovary.

Bone morphogenetic proteins (BMP) belong structurally to the transforming growth factor-beta superfamily comprising several growth and differentiation factors such as inhibin, activin, and Müllerian inhibitory factor that regulate ovarian function. We studied here the potential expression of BMP-2, -3, and -4 messenger RNAs (mRNAs) in isolated human granulosa cells obtained at oocyte retrieval for in vitro fertilization. Freshly isolated granulosa cells were found to express BMP-3 (also known as osteogenin) mRNAs but not those of BMP-2 or -4. The BMP-3 transcripts were detected with RT-PCR amplification followed by Southern blot hybridization, as well as by Northern and dot blot hybridization analyses. To investigate whether BMP-3 mRNAs are hormonally regulated, cultures of human granulosa-luteal (GL) cells were treated with different concentrations of purified human chorionic gonadotropin (hCG) at varying stages of culture. hCG decreased BMP-3 mRNA levels from the first day of the culture up to day 5. Time-dependence studies showed that a clear decrease in BMP-3 mRNA levels was evident at 24 h after hCG treatment, and that the effect of hCG was concentration dependent with 3 ng/mL hCG decreasing significantly (P < 0.05) BMP-3 mRNA levels. Furthermore, the cAMP analog, 8-bromo-cAMP (8-Br-cAMP), which activates protein kinase-A, and 12-0-tetradecanoylphorbol 13-acetate, an activator of protein kinase-C, both markedly decreased BMP-3 mRNA levels in an 8-h treatment. We conclude that: 1) BMP-3 mRNAs are expressed in human granulosa cells; 2) their steady state levels are hormonally regulated in cultured human GL cells as evidenced by the ability of hCG to markedly decrease BMP-3 transcript levels; and (3) that activation of both protein kinase-A-and protein kinase-C-mediated signaling pathways also results in a decrease in BMP-3 mRNA levels in GL cells. We suggest that BMP-3, like several other members of the transforming growth factor-beta superfamily, is a potential local regulator of female gonadal function.

We studied the effects of recombinant human FSH (rhFSH) and purified hCG on the steady state messenger ribonucleic acid (mRNA) levels of inhibin alpha- and beta A-subunits in cultured granulosa-luteal cells of preovulatory ovarian follicles obtained from women undergoing in vitro fertilization. Specific mRNA transcripts for the alpha- and beta A-subunits were detected in Northern and dot blot filter hybridization analyses, and the levels of these mRNAs were induced by rhFSH and hCG in a distinct concentration- and time-dependent manner. The basal and hCG-stimulated alpha-subunit mRNA levels were first determined at 2- to 3-day intervals over a 3- to 10-day culture period after the initiation of the cultures. Both the basal and hCG-stimulated alpha-subunit mRNA levels declined steadily during culture, but the maximal relative stimulatory effect of hCG was observed on day 7 of culture. All subsequent experiments, therefore, were performed on days 6-8 of culture. Both gonadotropins induced alpha-subunit mRNA levels with slower kinetics than those of the beta A-subunit. Varying between experiments, rhFSH and hCG increased the expression of the alpha-subunit with a maximal effect of 2.5- to 5.7-fold and 1.7- to 7.2-fold, respectively, above basal levels 24-48 h after stimulation. rhFSH and hCG induced beta A-subunit mRNA levels with 3.0- to 5.8-fold and 2.3- to 8.6-fold increases above basal levels, respectively, at 2 h; thereafter, only moderate or no stimulation of the beta A-subunit mRNA levels could be detected at 7-48 h. Treatment of the cells with the RNA synthesis inhibitor actinomycin-D prevented the induction of alpha-subunit mRNA levels by hCG, and no significant differences were detected in the stability of alpha-subunit mRNA transcripts in hCG-treated cells vs. untreated cultures. This indicates that hCG induces transcription of the alpha-subunit gene rather than maintains the levels of preexisting transcripts. As the kinetics of induction of alpha- and beta A-subunit mRNAs by gonadotropins were different, we examined how the inhibition of protein synthesis affects the induction of alpha- and beta A-subunit mRNAs by hCG. Cycloheximide had no effect on basal alpha-subunit mRNA levels at 2 or 24 h. However, it inhibited at 24 h the induction of the alpha-subunit by hCG.(ABSTRACT TRUNCATED AT 400 WORDS)

Activins have potent effects on early morphogenetic events during amphibian embryogenesis but no evidence for their role during human development other than their expression in steroidogenic tissues has been reported. We previously showed the expression of the activin type II and IIB receptor mRNAs in several tissues of the mid-gestational human fetus with highest expression levels in developing neural, muscular and exocrine glandular organs. We now report that the mRNA transcripts for activin beta A- and beta B-subunits and for the activin-binding protein follistatin are found co-expressed in several of these extragonadal tissues. Their mRNAs were detected by Northern analyses using specific single-stranded 32P-labeled cDNA probes. In the nervous system, both activin beta A- and beta B-subunit transcripts were expressed in the cerebrum and spinal cord. Follistatin was abundantly expressed in the spinal cord whereas weaker signals where observed in the cerebrum and cerebellum. In the muscular system, beta A-subunit was abundantly expressed in the heart but to a lesser extent in the skeletal muscle while the opposite was observed for follistatin. Follistatin, and activin beta A- and beta B-subunit mRNAs were also detected in developing kidney, salivary gland, liver, and adrenal. The predominance of beta A-subunit mRNAs in the bone marrow and beta B-subunit mRNAs in the salivary gland suggests specific roles for activin A and B, respectively, in these tissues. No hybridization signal was detected for the inhibin alpha-subunit in non-steroidogenic tissues indicating that, in contrast to activins and follistatin, the effects of inhibins may be restricted to the gonads and adrenals which are known to express high levels of the alpha-subunit transcript. Taken together, our results suggest that the activin-follistatin system regulates the development of several organ systems in the mid-gestational human fetus.