Search

Background Gonadotrophins exert their functions by binding follicle-stimulating hormone receptor (FSHR) or luteinizing hormone and human chorionic gonadotropin receptor (LHCGR) present on endometrium. Within ovaries, FSH induces autophagy and apoptosis of granulosa cells leading to atresia of non-growing follicles, whereas hCG and LH have anti-apoptotic functions. Endometrial cells express functioning gonadotrophin receptors. The objective of this study was to analyze the effect of gonadotrophins on physiology and endometrial cells survival. Materials and methods Collected endometria were incubated for 48 or 72 h with 100 ng/mL of recombinant human FSH (rhFSH), recombinant human LH (rhLH) or highly purified hCG (HPhCG) alone or combined. Controls omitted gonadotrophins. The effect of gonadotrophins on cytochrome P450 family 11 subfamily A polypeptide 1 (CYP11A1), hypoxia inducible factor 1α (HIF1A), and cell-death-related genes expression was evaluated by reverse transcription quantitative polymerase chain reaction (RT-qPCR). Immunohistochemistry for microtubule-associated proteins 1A/1B light chain 3B (MAP1LC3B) and apoptotic protease activating factor 1 (APAF-1) was performed. Results Gonadotrophins are able to modulate the endometrial cells survival. FSH induced autophagy and apoptosis by increasing the relative expression of MAP1LC3B and FAS receptor. In FSH-treated samples, expression of apoptosis marker APAF-1 was detected and co-localized on autophagic cells. hCG and LH does not modulate the expression of cell-death-related genes while the up-regulation of pro-proliferative epiregulin gene was observed. When combined with FSH, hCG and LH prevent autophagy and apoptosis FSH-induced. Conclusions Different gonadotrophins specifically affect endometrial cells viability differently: FSH promotes autophagy and apoptosis while LH and hCG alone or combined with rhFSH does not.

This review article centers upon family of gonadotropin hormones which consists of two pituitary hormones - follicle-stimulating hormone (FSH) and luteinizing hormone (LH) as well as one non-pituitary hormone - human chorionic gonadotropin (hCG) secreted by placenta, and their receptors. Gonadotropins play an essential role in proper sexual development, puberty, gametogenesis, maintenance of pregnancy and male sexual differentiation during the fetal development. They belong to the family of glycoprotein hormones thus they constitute heterodimeric proteins built of common α subunit and hormone-specific β-subunit. Hitherto, several mutations in genes encoding both gonadotropins and their receptors have been identified in humans. Their occurrence resulted in a number of different phenotypes including delayed puberty, primary amenorrhea, hermaphroditism, infertility and hypogonadism. In order to understand the effects of mutations on the phenotype observed in affected patients, detailed molecular studies are required to map the relationship between the structure and function of gonadotropins and their receptors. Nonetheless, in vitro assays are often insufficient to understand physiology. Therefore, several animal models have been developed to unravel the physiological roles of gonadotropins and their receptors.

Studies suggest that oocyte cumulus expansion is regulated by both cumulus expansion-enabling factor (CEEF) and cumulus expansion-inhibiting factors (CEIF). Many reports on CEEF have appeared, but CEIF has rarely been studied. By cumulus expansion assays using mouse cumulus-oocyte complexes (COCs) and oocytectomized complexes, the present study demonstrated that whereas follicular fluid (FF) from medium (diameter, 2-4 mm) goat follicles contained both CEEF and CEIF activities, FF from large (diameter, 5-6 mm) abattoir or large (diameter, 5-7 mm) follicle-stimulating hormone (FSH)-stimulated follicles contained neither. FF from (diameter, 5-7 mm) human chorionic gonadotropin-stimulated follicles showed CEEF but not CEIF activity. Whereas medium conditioned with cumulus or mural granulosa cells from medium goat follicles contained only CEEF activity, theca cell-conditioned medium (CM) showed both CEEF and CEIF activities. Whereas 0.01 mg/ml of heparin efficiently inhibited cumulus expansion of mouse COCs in vitro, FF from large follicles that showed no CEIF activity contained much higher concentrations (0.23-0.25 mg/ml) of heparin. None of the glycosaminoglycans (GAGs) tested inhibited cumulus expansion of goat COCs. Among the follicles observed, only FF from medium goat follicles contained a linoleic acid (LA) level sufficient to inhibit cumulus expansion of both mouse and goat COCs in vitro. CM contained some amount of GAGs but no LA. Taken together, the results suggest that 1) the FSH and luteinizing hormone (LH) surges before ovulation promote cumulus expansion by down-regulating CEIF and up-regulating CEEF activity, respectively; 2) GAGs are not the CEIF in goat follicles; and 3) LA has CEIF activity but additional factors must be involved, because CM that showed high CEIF activity contained no LA.

This paper analyzes, in the toad Bufo arenarum, the effect on spermiation and androgen secretion of two human recombinant gonadotropins, human recombinant LH (hrLH) and human recombinant FSH (hrFSH) as well as the well-known spermiation-inducing hormone, human chorionic gonadotropin (hCG). For this purpose, testes were incubated with different concentrations of hrLH (0.01-2.5 microg/ml) and hrFSH (0.05-5 microg/ml), and results were compared with those obtained with 2.5 microg/ml hCG. Spermiation was most efficiently stimulated by hrFSH, which elicited a higher response than either hrLH or hCG. Both hrFSH and hrLH produced a bell-shaped dose-response curve, with a 50% inhibition on spermiation at a concentration twice higher than that necessary to get the highest response. However, none of the gonadotropins yielded a biphasic response on androgen secretion, hrLH producing the highest response at a concentration that evoked a 70% inhibition in the spermiation test. Regarding steroidogenesis, hrLH and hrFSH were more active than hCG. Taken together, the results described in this paper suggest that, in B. arenarum, spermiation and androgen secretion are mediated by different receptors. After comparing the effects of recombinant hormones, we conclude that hrFSH has a greater effect on spermiation than hCG or hrLH.

Non-invasive methods for monitoring reproductive status of chimpanzee based on the measurement of urinary steroids and gonadotropins were examined. A typical pre-ovulatory urinary estrone conjugate (E1C) surge and post-ovulatory increase in pregnandiol glucuronide (PdG) were seen during the menstrual cycle. Urinary follicle stimulating hormone (FSH) showed two peaks over the infertile menstrual cycle. The earliest changes indicating pregnancy were a coincident rise in E1C and chorionic gonadotropin (CG) levels and a concomitant fall in FSH levels. Urinary PdG levels showed a prolonged rise. Urinary E1C in the pregnant chimpanzee was higher than during the menstrual cycle and increased with advancing gestation, with maximum levels occurring near term. In the case of stillbirth, E1C and CG levels from mid- through late-pregnancy were low and the prepartum progressive increase in E1C was not shown. The data presented here are of great practical value in captive breeding management of chimpanzees.

In the gonads, LH and hCG act via the same receptor to stimulate the production of progesterone in the luteal phase of the menstrual cycle and in early pregnancy. There are numerous reports that these two hormones can have direct actions on the uterus in addition to their indirect actions via stimulation of ovarian steroid hormones. However, unlike the situation in the gonads, various uterine tissues have been shown to respond to the related hormones FSH and TSH or the alpha-subunit common to these hormones. These additional actions cannot be mediated by the gonadal LH/hCG receptor. There have also been a series of reports that the uterus contains LH/hCG receptors. Attempts to characterize the molecular structure of these receptors have been difficult; thus, the possibility of a variant receptor cannot be excluded. The possibility also exists of a nonhomologous receptor, which would explain the differences in ligand specificity in uterine tissues. I will review the evidence regarding gonadotropin action in nongonadal tissues, primarily the uterus. In addition, the data regarding receptors will be reviewed. Finally, the clinical areas informed by this information will be explored.

Background: Human chorionic gonadotropin (hCG), lutropin, follitropin, and thyrotropin act as alpha beta heterodimers to control reproduction and thyroid function. The alpha and beta subunits of these proteins are divided into three loops (alpha 1,alpha 2,alpha 3; beta 1,beta 2,beta 3) by cysteine knots and the heterodimer is stabilized by 20 beta-subunit residues wrapped around alpha 2 like a seatbelt. Understanding how these hormones interact with their receptors, a matter of considerable dispute, would facilitate design of pro- and anti-fertility agents., Results: By swapping alpha 2 for beta 2 and vice versa and, in some cases, adding an amino-terminal coiled-coil dimerization domain, we prepared homodimeric analogs that have the conformation found in each 'half' of hCG. Homodimers containing loops beta 1,alpha 2,beta 3 and none, part, or all of the seatbelt stimulated signal transduction to the same extent as hCG, albeit with lower potency. Those containing alpha 1,beta 2,alpha 3 were inactive., Conclusions: The activities of homodimers containing the beta 1,alpha 2,beta 3 groove exceed those of other minimized analogs more than 100-1000-fold, suggesting this portion of the hormone forms the major receptor contact. The discovery that glycoprotein hormone heterodimers can be converted to functional homodimers supports the proposal that this protein family evolved from an active homodimeric ancestor by gene duplication and acquisition of mutations to loop 2 that prevent homodimerization. This approach to protein minimization should be applicable to other proteins composed of architecturally related subunits, including those that might have arisen by gene duplication.

Purpose: The results obtained with a protocol consisting of ovarian stimulation with low doses of highly purified FSH (FSH HP), administration of a GnRH analogue to induce an endogenous surge of gonadotropins, and IUI were evaluated. These results were compared with those seen with similar FSH stimulation and hCG administration followed by IUI., Methods: Three hundred sixty-four patients scheduled for IUI, after inclusion in a total of 345 FSH HP/GnRH-stimulated cycles and 416 FSH HP/HCG-stimulated cycles, were studied. The stimulation protocol consisted of daily subcutaneous injection of 75 IU of FSH HP from day 3 or 5 of the cycle, depending on the duration of the spontaneous cycle. hCG was administered on days 0, +2, and +5 to support the luteal phase. Monitoring was conducted using circulating estradiol levels and vaginal ultrasonography. Administration of two s.c. doses of leuprolide acetate (LA) or 7500 IU of i.m. hCG when at least one 18-mm-diameter follicle was seen and estradiol levels reached 120 pg/ml per follicle with a diameter > or = 16 mm. Intrauterine insemination was with semen capacitated by swim-up, thawed at room temperature if previously frozen., Results: The ovulation rate was 99.28 after hCG and 99.23 with LA. No significant differences were seen between the estradiol and progesterone levels of both groups or in the estradiol/progesterone ratio. The duration of the luteal phase was similar in both groups. Pregnancy rates per cycle were 17.31% (hCG) and 27.25% (LA), respectively (P = 0.0007), and abortion rates 22.22% (hCG) and 24.47% (LA), respectively. No cases of ovarian hyperstimulation were seen., Conclusions: After FSH HP administration according to a low-dose protocol, the use of LA to trigger a gonadotropin surge as a means of inducing ovulation in FSH-stimulated women could be a good alternative to improve the results and prevent ovarian hyperstimulation in IUI cycles.

The glycoprotein hormones (TSH, FSH, LH, CG) are structurally related proteins with diverse physiologic functions. This family of hormones offers an opportunity to address fundamental questions concerning how gene expression is regulated in a cell-specific manner and in response to different hormones. For example, the alpha-subunit gene is expressed in several different pituitary cell types (gonadotropes and thyrotropes) as well as in the placenta. Because it must be coordinantly expressed with the different beta-subunit genes, the alpha-gene provides an interesting model for multihormonal control which varies in a cell-type specific manner. Many of the promoter regulatory DNA sequences and cognate transcription factors in the alpha-gene have been identified. These studies reveal a remarkable series of composite regulatory elements that interact with families of transcription factors that are still being characterized. In contrast, the beta-subunit genes are notable for restricted cell-type expression and more limited hormonal regulation that reflects their individual physiologic roles. The TSH beta gene is expressed only in thyrotropes where, in conjunction with the alpha-gene, it is subject to transcriptional repression by thyroid hormone. The FSH beta gene is expressed in gonadotropes where its expression is controlled primarily by activin and inhibin, with additional regulation by GnRH. The LH beta gene is also expressed in gonadotropes, but it is more dependent upon GnRH input and its expression is unaffected by the activin/inhibin system. The CG beta gene evolved recently from the LH beta gene and in the process, the CG beta promoter acquired new regulatory elements that favor its expression in the placenta rather than the pituitary gland. Less is known about the regulatory elements in the beta genes, in part because highly differentiated cells are required for their normal regulation. This chapter reviews the regulation of this family of genes with an emphasis on recent studies from our laboratory involving the gonadotropins (LH, FSH, CG). Concomitant with our advancing understanding of how the gonadotropin genes are regulated, we are also learning about genetic causes of gonadotropin deficiency syndromes.

Human choriogonadotropin (hCG) and follitropin (FSH) belong to the glycoprotein hormone family. These hormones are heterodimers and composed of a common alpha subunit and a distinct beta subunit which confers receptor-binding specificities. In addition to this structural similarity, they share a similar signal pathway involving G protein, adenylyl-cyclase and induction of cAMP synthesis. Therefore, a presumptive relationship of these common structure and function has been the subject of extensive past investigations, but a definitive clue has been elusive. As a step to address this important issue, a series of recombinant mutants of hCG and human FSH were generated in which the COOH-terminal amino acids of the alpha subunit were successively removed or substituted. Furthermore, a set of peptides were synthesized with sequences corresponding to different regions of the alpha subunit. Deletion of the alpha COOH-terminal Ser92 had no effect on receptor-binding or cAMP induction by FSH and hCG. Truncation of alpha Lys91-Ser92 or alpha His90-Lys91-Ser92 abolished the ability of both hormones to induce cAMP synthesis. It significantly reduced receptor binding of FSH but not hCG. The different functions of the alpha COOH-terminal region are further noticed with a peptide corresponding to the last 10 amino acids of alpha. It failed to bind to the FSH receptor but was capable of binding to the LH/CG receptor and stimulating cAMP synthesis. These results are the first conclusive evidence that alpha His90-Lys91 play an essential role in cAMP induction of both hormones. In contrast to this common role, they are necessary for FSH binding to the FSH receptor but not for hCG binding to the LH/CG receptor. The hCG alpha COOH-terminal region makes direct contact with the LH/CG receptor, and this low affinity contact is necessary and sufficient to activate the receptor for signal generation. This conclusion is supported by the study using mutant hCGs in which either alpha His90 or Lys91 was substituted.

Plasma prolactin levels rise in stimulated cycles. To clarify the effects of gonadotrophin on the lactotrophs, three studies were performed. First, plasma concentrations of prolactin during clomiphene citrate (CC)-human menopausal gonadotrophin (HMG)-human chorionic gonadotrophin (HCG) treatment of women enrolled for in-vitro fertilization (IVF) were compared with those during HMG-HCG administration while under pituitary suppression with a gonadotrophin releasing hormone (GnRH) analogue (buserelin). Women suppressed with buserelin had higher basal levels of PRL in plasma (14.4 +/- 4.3 ng/ml versus 6.9 +/- 1.4 ng/ml, P less than 0.001). Only buserelin-suppressed women showed a significant rise in plasma prolactin before HCG administration, while both patient groups had marked prolactin peaks after HCG injection. This peak was higher in the buserelin group (71.9 +/- 50.7 ng/ml versus 52.6 +/- 29.7 ng/ml). The second study showed that plasma levels of prolactin of 6 post-menopausal women were significantly increased 48 h after an injection of 5000 IU HCG, i.m. (24.9 +/- 17.4 ng/ml versus 12.4 +/- 6.2 ng/ml P less than 0.05). Third, plasma prolactin was studied in 5 women over 30 days after surgical castration. An upward trend was observed similar to that of endogenous gonadotrophin, with the change in prolactin values closely correlating with the change in concentrations of follicle stimulating hormone (P less than 0.005). All these findings suggest that human gonadotrophins stimulate lactotrophs.

Factors regulating LH/hCG responsiveness in primate granulosa cells were examined in the marmoset monkey (Callithrix jacchus). Granulosa cells were isolated and pooled from small antral (0.5-1.0 mm) and large preovulatory (greater than or equal to 2 mm) follicles from mid- to late follicular phase ovaries of cyclic marmosets. The cells from small and large follicles were cultured in serum-free medium for 48 h in the absence or presence of increasing concentrations of hCG (0.1-100 ng/ml) with or without 0.1 microM androgen [testosterone or 5 alpha-dihydrotestosterone (DHT]). Granulosa cells from small follicles were also cultured in the absence or presence of a constant concentration of human FSH (30 ng/ml) with or without androgen for 48 h before exposure to hCG for an additional 48 h. Steroidogenic responsiveness was assessed by measuring progesterone accumulation in culture medium and aromatase activity in washed monolayers. Granulosa cells from large follicles showed dose-dependent increases in both progesterone accumulation and aromatase activity in response to treatment with hCG. In contrast, granulosa cells from small follicles were unresponsive to hCG. However, pretreatment of granulosa cells from small follicles for 48 h with FSH stimulated hCG responsiveness. The effects of both testosterone and DHT on hCG-stimulated aromatase activity and progesterone accumulation by granulosa cells from large preovulatory follicles were inhibitory. Testosterone and DHT also suppressed basal (no hCG) progesterone accumulation in these cells, but had no effect on basal aromatase activity. The effects of androgens on FSH-induced hCG responsiveness in immature granulosa cells were variable. The results show a development-related increase in marmoset granulosa cell responsiveness to LH/hCG and provide evidence that FSH and androgens interact to regulate the onset and expression of this critical event during preovulatory follicular development in the primate ovary.

Recently we have demonstrated that administration of a "pure" follicle-stimulating hormone (FSH) preparation (Urofollitropin, Serono Laboratories, Inc., Randolph, MA) to normally cycling monkeys induces multiple follicular development. In these earlier studies, a spontaneous luteinizing hormone (LH) surge was uncommon; no attempt was made to induce ovulation with exogenous human chorionic gonadotropin (hCG). In this study, multiple follicular development and ovulation were induced in normally cycling monkeys by daily follicular phase administration of "pure" FSH followed by hCG. Short-term administration of "pure" FSH during the early or late follicular phase also induced multiple follicular development; however, multiple ovulations subsequent to a spontaneous LH surge never occurred. One monkey treated in the late follicular phase did demonstrate a spontaneous LH surge and single ovulation following late follicular phase FSH treatment. These findings suggest that administration of "pure" FSH alone, to enhance the natural ovarian cycle, may be useful for inducing multiple follicular development, but that ovulatory competence is usually dependent on exogenous LH/hCG.

The objective of the present studies was to assess whether hormone induction of oocyte maturation in isolated intact follicles may be linked to desensitization of follicle-stimulating hormone (FSH) in the oocyte-cumulus complex (OCC). Incubation of follicles with chorionic gonadotropin (hCG), FSH or epidermal growth factor (EGF) produced a marked inhibition of FSH-dependent cyclic AMP accumulation in OCC with a time-course coincident with the onset of germinal vesicle breakdown (GVBD). These effects were evident within 3 h for both hCG and FSH, but with EGF a reduced response to FSH was seen within 1 h of treatment followed by an increase in GVBD. In contrast, no inhibition of cyclic AMP accumulation was seen in response to cholera toxin, forskolin or LH in OCC derived from follicles incubated with hCG for 3 h. The time-course for induction of oocyte maturation by incubation of the intact follicle with hCG was also coincident with production of prostaglandin (PG) F2 alpha, an indirect marker of cyclooxygenase induction. No effect on metabolic coupling between the oocyte and cumulus cells was seen until 9 h after hCG treatment. Retinoic acid caused a marked decrease in metabolic coupling between the oocyte and cumulus cells but inhibited oocyte maturation both in denuded oocytes and OCC. Since FSH desensitization in OCC, the resumption of meiosis, and production of arachidonic acid-derived products were coincident, it is suggested that abrogation of FSH action in cumulus cells by the ovulatory surge of gonadotropins may initiate oocyte maturation.

Relationships between the sequences of thyrotropin (thyroid-stimulating hormone, TSH), lutropin (luteinizing hormone, LH), human choriogonadotropin (chorionic gonadotropin, hCG) and follitropin (follicle-stimulating hormone, FSH) are now well established. Each beta-subunit contains six disulphide bonds and considerable homology is seen when all four linear sequences are aligned with half-cystine residues in juxtaposition. Major questions about the tertiary structures of the subunits and their interactions to form active hormone remain. Determination of the disulphide bridges in both alpha- and beta-subunits has not yielded to usual methods and conflicting data about the alpha-subunit have been reported. Partial reduction of the beta-subunits of LH and TSH with subsequent labelling of the cysteines formed has shown that a single bond is first reduced. This bond is between positions 93 and 100 in LH-beta and the corresponding positions 88-95 in TSH-beta. Thus, as would be expected from the fact that interhormone hybrids can be made with the common alpha-subunits, the chemical data, though still limited, indicate similar tertiary structures for the different beta-subunits. To investigate whether other useful intermediates can be obtained after partial reduction, we have studied reduction and derivative formation in various conditions. Intact LH is more resistant to reduction than either its alpha- or beta-subunit but no intermediates have been observed which are not present after partial reduction of individual subunits. Preliminary experiments on the reoxidation of fully reduced alpha-subunit show that the reoxidized material will recombine with native beta-subunits to yield biologically active TSH or LH. Studies from this and other laboratories on chemical modifications of several amino acid residues of glycoprotein hormones and their subunits are also summarized.

The effect of synthetic LHRH on serum levels of FSH, LH, and hCG was determined during early and midgestation. Eight healthy volunteers were studied during the first (8--9 weeks) and second (15--20 weeks) trimesters of pregnancy with 4 patients in each group. Serum samples, obtained before and 15, 30, 45, 90, 120, 150, and 180 minutes after an intravenous bolus of 100 microgram LHRH, were assayed for LH, FSH, and hCG by specific radioimmunoassays. Serum levels of estradiol, progesterone, and 17 alphahydroxyprogesterone were also measured in samples obtained prior to and 3 hours after LHRH injection. The serum levels of LH and FSH were undetectable in all samples, and no increase was observed after administration of LHRH. Levels of chorionic gonadadotropin, estradiol, progesterone, and 17alpha-hydroxyprogesterone did show slight fluctuations, but there was no significant effect from LHRH administration. These results support the concept that pituitary gonadotropic function is markedly suppressed during early human pregnancy.

Using an affinity-purified antibody against cholesterol side-chain cleavage P450 (P450scc) and a human P450scc cDNA probe, 11 rat P450scc cDNA clones were identified and isolated from our rat granulosa cell lambda gt11 cDNA expression library. Two clones were plaque purified and subcloned into pBR322. One of these P450scc cDNA clones, approximately 1.2 kilobases (kb) in size, was used as a probe for Northern and filter hybridization assays to analyze the tissue distribution and hormonal regulation of P450scc mRNA in rat ovarian follicles and corpora lutea. Northern transfers revealed a single P450scc mRNA species about 2.0 kb in size. Filter hybridization assays showed that P450scc mRNA was low in granulosa cells and thecal cells of small antral follicles, was increased in both tissues of preovulatory follicles, and was rapidly (within 7 h) and maximally increased (30-fold) during hCG-induced luteinization. P450scc enzyme and mRNA were also elevated in corpora lutea isolated from pregnant rats (days 4-22 of gestation) and rats 1 day after parturition (day 23). The elevation of P450scc enzyme and mRNA was maintained despite the marked decline in serum progesterone concentrations between days 19-22, suggesting that once P450scc mRNA is induced in luteal tissue it may be constitutively expressed. Administering hormones to granulosa cells in culture and to hypophysectomized immature rats in vivo demonstrated that the induction of P450scc mRNA by FSH in granulosa cells was time, dose, and estradiol dependent. High doses of FSH acting on estradiol-primed cells gave the greatest response. The increase in P450scc mRNA in cultured granulosa cells was also stimulated by forskolin and was directly associated with increased synthesis of cAMP and progesterone accumulation. Thus, whereas the induction of P450scc mRNA in granulosa cells was dependent on hormones and cAMP, the maintenance of P450scc mRNA and P450scc protein in corpora lutea appears to involve constitutive expression of P450scc mRNA.

We performed a series of studies in normal men to determine the relative roles of FSH and LH in the regulation of human spermatogenesis. Experimental gonadotropin deficiency was induced in normal men by administration of high-dosage testosterone (T) enanthate, resulting in suppression of spermatogenesis to severely oligospermic or azoospermic levels. In these gonadotropin deficient normal men, we found that selective replacement of either LH activity, with human chorionic gonadotropin (hCG) or hLH, or FSH activity with hFSH increased sperm counts into the 20-50 million/ml range, but quantitatively normal spermatogenesis was not achieved. To determine whether both LH and FSH together are required for quantitatively normal spermatogenesis, selective FSH deficiency was induced in normal men by administration of hCG. In the setting of high LH-like activity and markedly suppressed FSH levels induced by hCG, sperm production was partially suppressed again into the 20-50 million/ml range. Replacement of FSH activity, with either hFSH or human menopausal gonadotropin in hCG-treated normal men stimulated sperm production into the control range in all subjects. These studies demonstrate that neither normal serum levels of FSH or LH are required for qualitatively normal spermatogenesis in man, since sperm counts in the range of 20-50 million/ml are found in states of selective FSH or LH deficiency. However, normal levels of both FSH and LH are required to achieve quantitatively normal sperm production in man.

Human granulosa cells cultured with calf serum actively proliferated for 18-20 generation and secreted progesterone into the medium; progesterone levels appeared to decline with increase in generation number. Cells cultured under serum-free conditions secreted significant amounts of progesterone and insulin-like growth factor II (IGF-II). The progesterone secretion was enhanced by the addition of human follitropin, lutropin, and chorionic gonadotropin but not by growth hormone. These cells, when challenged to varying concentrations of human growth hormone, human chorionic somatomammotropin, human prolactin, chorionic gonadotropin, follitropin, and lutropin, secreted IGF-II into the medium as measured by specific IGF-II RIA. Among these human hormones, chorionic gonadotropin, follitropin, and lutropin were most effective in inducing IGF-II secretion from these cells. When synthetic lutropin-releasing hormone and alpha-inhibin-92 were tested, only lutropin-releasing hormone was effective in releasing IGF-II. The results described suggest that cultured human granulosa cells can proliferate and actively secrete progesterone and IGF-II into the medium. IGF-II production in human granulosa cells was influenced by a multi-hormonal complex including human growth hormone, human chorionic somatomammotropin, and prolactin.

The extent of FSH-mediated LH/hCG receptor induction and of basal and FSH-stimulated progesterone production by porcine granulosa cells in vitro, in serum-containing medium, is directly related to the plating density. Relative to pre-culture levels, low- and high-density cultures of cells routinely exhibited 1-2- and 10-11-fold increases in [125I]iodo-hCG binding, respectively. Monolayer growth, i.e. cell division, as measured by increases in cell protein or DNA content, was inversely related to plating density. This density-directed inverse relationship between growth and differentiation did not appear to be coupled under the conditions utilized. Whereas monolayer growth was dependent upon the cell surface density, i.e. the number of cells per unit surface area, differentiation was dependent upon cell concentration, i.e. cells per unit volume of medium. Cells plated at low density in medium containing 10% serum exhibited 50% less [125I]iodo-hCG binding than cells in 5% serum (P less than 0.025). Conversely, cells plated at high density exhibited a 14% increase (P less than 0.025) in binding at the higher serum level. Thus, it appears that the extent of differentiation depends upon the capacity of cells to neutralize serum inhibition which in turn is dependent upon the cell concentration. Serum neutralization by granulosa cells is an FSH-dependent process. Conditioned medium derived from insulin-treated, high-density cultures did not facilitate optimum LH/hCG receptor induction in low-density cultures. Conditioned medium from cultures treated with insulin plus FSH, however, facilitated LH/hCG receptor induction in low-density cultures to the same extent as obtained in high-density cultures. The enhancement by FSH-conditioned medium cannot be attributed to residual FSH or to dilution of serum components during the preparation of the conditioned medium. The phenomena of serum-attenuated granulosa cell differentiation in vitro, and of a density-dependent reversal of this process, may have regulatory implications in vivo since follicular fluid contains many serum components and since the granulosa cell complement is an important determinant of follicle maturation.