Effect of recombinant activin on androgen synthesis in cultured human thecal cells

The effect of activin-A on ovarian androgen synthesis was tested in vitro using serum-free monolayer cultures of human thecal cells. Maximal rates of androgen (androstenedione and dehydroepiandrosterone) production were induced by treating the cells for 4 days with LH (10 ng/mL) in the presence of insulin-like growth factor-I (greater than or equal to 30 ng/mL). The additional presence of recombinant activin-A (1-100 ng/mL) in culture medium caused dose-dependent suppression of thecal cell androgen production, with 50% maximal inhibition occurring at an activin-A concentration of about 10 ng/mL. Progesterone production was only suppressed by high dose (100 ng/mL) activin-A, and inhibition of steroid production occurred without inhibition of DNA synthesis (tritiated thymidine uptake). These results reveal a potent and selective inhibitory action of activin-A on thecal cell androgen synthesis, consistent with a paracrine function for activin(s) in modulating follicular androgen biosynthesis in the human ovary.     

Ovarian nerve extracts influence androgen production by cultured ovarian thecal cells

There is growing evidence that ovarian steroidogenesis is controlled not only by pituitary gonadotropins but also by ovarian nerves. Nerves reach the ovary via the plexus nerve and via the superior ovarian nerve (SON), which runs in the suspensory ligament, and innervate theca cells of all sizes of follicles. To investigate the role of ovarian nerves in steroidogenesis we have examined the effects of adding an extract of SON from adult rats on androgen production by cultured porcine theca cells. Addition of SON extract to cultured theca interna from 3 to 6 mm diameter follicles of prepubertal gilts significantly inhibited (p less than 0.05) LH-stimulated androstenedione production in a dose-dependent manner; significant inhibition (10.8%) occurred with the addition of the extract of 2 mg of SON/ml culture medium, and near maximal inhibition (83%) resulted when the SON extract was increased to 60 mg SON/ml. Extracts of sciatic nerves, used as non-ovarian nerve controls, failed to inhibit, and in fact significantly increased (p less than 0.05) androstenedione production over the same concentration range of neural tissue extract. The inhibitory effect of the SON extract was unaffected by chymotrypsin digestion or by the presence of the beta-adrenergic antagonist propranolol (10(-6) M), but was removed by charcoal treatment. These results suggest that the nervous system has the potential for modulation of follicular steroid biosynthesis via direct innervation of the ovaries, in addition to the well-established indirect mechanism of neural control exerted via the hypothalamic-pituitary system.     

Effect of androgen on androgen receptors in cultured human epididymis

Optimal assay conditions were developed to determine androgen receptor concentrations in samples of human epididymis. Pretreatment of cytosol with mersalyl, as well as the inclusion of molybdate in the homogenization buffers, resulted in a substantial increase in the number of soluble sites detected. A high yield of nuclear and microsomal binding sites was obtained by prolonged incubation (12 h) in 0.6 mol NaCl/l. Organ culture for 6 days resulted in a major loss of androgen-binding sites. In the absence of androgen in the culture media, only 20% of the original sites were found in cultured tissue. Inclusion of dihydrotestosterone (0.1 mumol/l) in the media resulted in samples containing twice as many sites as controls. It is concluded that androgens influence the number of androgen-binding sites in cultured human epididymis in a manner analogous to that described for rat epididymis in vivo.     

Effect of recombinant human inhibin on gonadotropin secretion by the male rat

We have examined the effect of recombinant human inhibin-A on basal and GnRH-induced gonadotropin secretion by male rats or cultured anterior pituitary cells. Inhibin, administered sc 6 h before the experiment, induced dose- and time-related decreases in plasma FSH, but not LH, levels in both intact and castrated male rats. Inhibin also significantly interfered with the in vivo stimulatory effect of 20-500 ng GnRH on FSH release, but had inconsistent and usually modest effects on the LH response. While exposure of cultured pituitary cells to inhibin for 72 h has been reported to interfere with GnRH-induced gonadotropin release, we examined here the effects of shorter exposure periods relevant to in vivo experiments. Exposure of the cells to inhibin (31.3-312.5 pM) for 2-6 h measurably (P less than or equal to 0.01) decreased the ability of 10 nM GnRH to stimulate both FSH and LH released by cultured cells. In contrast, lower (3.1 and 9.4 pM) doses of inhibin had little or no effect. Longer exposures to inhibin (10, 24, and 72 h) increased the inhibitory effect of 31.3-312.5 pM inhibin, while 3.1 and 9.4 pM remained ineffective at all times. These results indicate that exposure of the male rat to inhibin for 6 h decreases FSH secretion, and that this effect is at least partially mediated through blunting of the pituitary response to GnRH. In contrast, the ability of inhibin to interfere with LH release, which is readily apparent in cultured pituitary cells, appears to be of lesser importance in the intact male rat.     

Effect of danazol on gonadotropin and cAMP stimulated progesterone synthesis in cultured human granulosa cells

The effect of danazol on progesterone formation was studied in cultured human granulosa cells obtained from 20 follicles in the mid- to late follicular phase of normally menstruating women. The cells were cultured for 2 to 6 days in the presence of danazol (0.01 to 5 mg/l) alone and in combination with one of several progesterone stimulatory agents. The medium was changed every other day and analysed for progesterone with radioimmunoassay. In all cases progesterone synthesis was markedly stimulated by human FSH (10 micrograms/l), human LH (10-100 micrograms/l), hCG (1000 IU/l), forskolin (1 mumol/l) or 8-Bromo-cAMP (1 mmol/l). In the presence of danazol (1 mg/l) the FSH-, LH- and hCG-stimulated progesterone synthesis was partially inhibited by 55 to 60%. The response to forskolin or 8-Bromo-cAMP was also inhibited by danazol although to a somewhat lower extent (35 to 50%). Basal progesterone synthesis was inconsistently inhibited in some cases. It was concluded that not only gonadotropin-stimulated steroid synthesis, as previously demonstrated for hCG, but also forskolin and 8-Bromo-cAMP-stimulated steroidogenesis is sensitive to danazol inhibition. This suggests that the effect of danazol in human granulosa cells is due, at least partly, to interference with steps distal to cAMP generation.     

Effects in vivo of recombinant human inhibin on myelopoiesis in mice.

Inhibin, a protein dimer, has been implicated in negative regulation of human erythropoiesis in vitro. In this study, purified recombinant human (rhu) inhibin was assessed for its effect in vivo on the proliferation of hematopoietic progenitor cells in C3H/HeJ mice. Administration of single doses of inhibin i.v. to mice resulted 24 hrs later in significant decreases in cell cycling status of marrow and splenic granulocyte-macrophage (CFU-GM), erythroid (BFU-E) and multipotential (CFU-GEMM) progenitors. While no apparent effect was observed on marrow cellularity or on absolute numbers of marrow CFU-GM and BFU-E, inhibin significantly reduced absolute numbers of marrow CFU-GEMM, spleen nucleated cellularity and also absolute numbers of CFU-GM, BFU-E and CFU-GEMM in the spleen. The results demonstrate in vivo myelosuppressive effects for inhibin and demonstrate that effects in vivo are not restricted to erythropoiesis.     

Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium

The direct inhibitory effect of estrogen on ovarian androgen synthesis was investigated. When primary cultures of rat ovarian theca-interstitial cells were grown in defined medium with LH there was a marked increase in androgen synthesis of which 98% was androsterone (control = 11 ± 2 ng; LH = 1219±217 ng/ml/10⁶ cells). Diethylstilbestrol (DES), estrone (E₁), estradiol (E₂), and estriol (E₃) inhibited LH-stimulated androsterone synthesis by 81%, 81%, 81%, and 47%, respectively. The ED₅₀'s of the estrogens were: DES = 4.2±2.l × 10^?9M; E₁ = E₂ = 9.5±2.4 × 10^?8 M; and E₃ = 3.8±2.6 × 10^?7 M. The estrogen effect was very rapid ($\text{t}{}_{\mathrm{<mtext>1</mtext><mtext>2</mtext>}}$ = 10 min) and long-lasting. Metabolic studies revealed that estrogen inhibited androsterone, androstenedione, 5α-androstane-3α, 17β-diol, and testosterone accumulation by 80%, dehydroepiandrosterone and 17α-hydroxypregnenolone by 40%, 17α-hydroxyprogesterone by 30%, while pregnenolone and progesterone were unchanged. These results prove, for the first time, that estrogen can directly inhibit LH-stimulated androgen production in ovarian theca-interstitial cells and suggest that mechanism involves, at least in part, a rapid selective inhibition of the 17α-hydroxylase/C_17–20 desmolase activities.     

Effect of lipopolysaccharides on cultured human endothelial cells

Summary Exposure to lipopolysaccharides (LPS; 10 g/ml derived from either S. enteritidis or E. coli or to their lipid A moiety alone induced procoagulant activity in cultured human endothelial cells. This exclusively cell-associated activity was identified as tissue factor activity by two criteria: Firstly, the presence of Factor VII was required for its expression and, secondly, clotting was abolished by the addition of the IgG fraction of anti-human tissue factor antibodies. Concomitant analysis of prostacyclin (PGl₂) formation by the cells showed a substantial increase in the production of this potent platelet inhibiting substance during exposure to endotoxin. LPS-induced release of PGl₂ did not result in refractoriness of the cells to generate new PGl₂ as indicated by the retained response to stimulation with 20 M arachidonic acid. While the release of PGl₂ could be inhibited by pretreatment of the cells with 100 M acetylsalicylic acid (ASA), the induction of tissue factor activity remained unaffected by ASA. In contrast to LPS-free control cultures, ASA did not completely prevent PGl₂ formation by human endothelial cells after exposure to LPS suggesting the induction of a cyclooxygenase-independent pathway by LPS.     

Effect of low density lipoprotein on DNA synthesis of cultured human arterial smooth muscle cells

Increased concentration of plasma low-density lipoprotein (LDL) is one of the risk factors in atherosclerosis. We studied how DNA synthesis of human arterial smooth muscle cells (SMC) was influenced in the culture added with or without small dose (25 micrograms of protein/ml) of LDL. LDL were ultracentrifugally obtained from normal subjects and diabetics. DNA synthesis was investigated at 6-h intervals during 36 h in the culture with or without LDL using [methyl-3H]thymidine. We found that DNA synthesis reached a maximum value at 24 h after addition of LDL. On the other hand sequential changes were not detectable in the culture without LDL addition. This effect of diabetic LDL was significantly (P less than 0.001) greater than that of normal LDL. These results suggested that LDL induces synchronization of the cultured SMC to synthesize DNA and diabetic LDL may play an atherogenic role more strongly than normal LDL in the arterial wall even in the normal range of LDL concentration.     

Hormonal regulation of inhibin production by cultured Sertoli cells

The hormonal regulation of inhibin production by cultured rat Sertoli cells was examined using a specific radioimmunoassay (RIA) which detects the N-terminal portion of the porcine inhibin alpha chain. FSH, but not hCG or prolactin caused a dose-dependent increase in inhibin production (EC50 for FSH = 2.4 ng/ml); both secreted and intracellular levels of inhibin were increased, but the secreted form represented one-half to two-thirds of the total. The FSH-stimulated production of inhibin was augmented by addition of a phosphodiesterase inhibitor, and could be mimicked by cholera toxin, forskolin, or dibutyryl cAMP, all of which are known to increase intracellular cAMP levels. Inclusion of either dihydrotestosterone or estradiol in the cultures had no effect on inhibin production, both in the presence and absence of FSH. Examination of the conditioned media from forskolin-treated Sertoli cells by gel filtration chromatography revealed a single peak of bioactive and immunoreactive inhibin, at a molecular weight of approximately 32,000, similar to that observed for the porcine and bovine follicular fluid inhibins. Thus, FSH activated the cAMP pathway to stimulate Sertoli cell production of inhibin which in turn suppresses pituitary FSH release to form a closed-loop feedback system.     

The specificity of inhibin bioassay using cultured pituitary cells

It has been established that the inhibition of the cellular content of FSH in cultured pituitary cells can be used as a sensitive and precise bioassay for inhibin. During studies on the inhibin content of human plasma, FSH suppression was noted to occur together with LH suppression. Further studies revealed that where combined FSH and LH suppression occurred, cytological changes in the pituitary cells suggestive of toxicity were found, indicating non-specific effects of these substances. Charcoal treatment or gel filtration of seminal plasma removed or separated the toxic substances from the inhibin activity, the latter characterized by FSH suppression parallel to the standard preparation, no LH suppression and no light-microscopic changes in pituitary cells. It is recommended that careful evaluation of all inhibin bioassay systems should be undertaken to detect substances producing non-specific effects and additional guidelines for the assessment of specificity are suggested.     

Steroidogenic enzyme activities in cultured human definitive zone adrenocortical cells: comparison with bovine adrenocortical cells and resultant differences in adrenal androgen synthesis

The activities of 3 beta-hydroxysteroid dehydrogenase, 17-hydroxylase, 21-hydroxylase, 11 beta-hydroxylase, C17,20-lyase, and dehydroepiandrosterone sulfotransferase were measured in cultured human fetal definitive zone adrenocortical cells with and without prior exposure to 1 microM ACTH for 48 h. Enzyme induction and measurements of activity were performed using serum- and lipoprotein-free conditions. ACTH induced increases of 5- to 100-fold in the activity of all of these enzymes. Although 3 beta-hydroxysteroid dehydrogenase activity was increased 15-fold, its activity was still an order of magnitude less than that of the hydroxylases. In contrast, when similar experiments were performed using bovine adrenocortical cells, 3 beta-hydroxysteroid dehydrogenase activity was similar to that of the hydroxylases after induction with ACTH. The lower activity of 3 beta-hydroxysteroid dehydrogenase in human cells compared to that in bovine cells resulted in different sequences of transformation of [3H]pregnenolone. The initial product in human cells, before or after induction with ACTH, was 17-hydroxypregnenolone, which was then converted about equally to cortisol (via 17-hydroxyprogesterone and 11-deoxycortisol) and dehydroepiandrosterone sulfate (via dehydroepiandrosterone). In contrast, bovine cells converted pregnenolone to progesterone, with or without prior exposure to ACTH, which was then converted to 17-hydroxyprogesterone, with minimal formation of dehydroepiandrosterone. Adrenal androgen synthesis by human adrenocortical cells thus results from low 3 beta-hydroxysteroid dehydrogenase, which is an intrinsic cell property. Since these experiments were performed using serum-free conditions, cells were not exposed to hormones other than ACTH. The results support the hypothesis that human adrenal androgen synthesis does not require a special hormone.     

Presence and synthesis of inhibin subunits in human decidua

A growing number of studies provided the evidence that human decidua is a pregnancy-related tissue capable of hormone production and metabolism. The aim of the present study was to evaluate the possible presence of inhibin subunits in human decidua. Tissue samples were collected in pregnant women during the first (8 weeks) and second trimester (18 weeks) of gestation and at term (40 weeks). Immunohistochemical data were obtained using affinity purified polyclonal antisera raised in rabbit against porcine alpha, beta A, or beta B subunits. Levels of the respective inhibin subunits were evaluated by Northern blot analysis using cDNA probes encoding sequences corresponding to each subunit. The present results indicated that human decidua contains and synthesizes inhibin alpha, beta A, and beta B subunits. The immunohistochemical data showed that decidual cells were stained with both inhibin alpha and beta B antisera, showing a similar localization. On the other hand, cells stained with inhibin beta A antisera were sparse and followed a distribution pattern different from that of cells stained with alpha or beta B antisera. The first inhibin alpha and beta B subunit mRNAs were both expressed in first trimester of pregnancy, and those mRNA levels showed a gestational related increase. The beta A subunit mRNA was expressed at very low levels at term and could not be detected earlier during pregnancy. The present data showed that human decidua actively produces inhibin subunits with a gestational-related profile. The results suggest that decidua may be a further source of inhibin-related proteins during pregnancy and emphasize the endocrine competence of human decidua.     

17 beta-Estradiol biosynthesis in cultured granulosa and thecal cells of human ovarian follicles: stimulation by follicle-stimulating hormone.

Cellular sites and gonadotropic control of human follicular estrogen secretion have been assessed by culturing the theca and granulosa components separately under different hormonal conditions. Granulosa cells from human follicles were grown in chemically defined media containing gonadotropins and/or testosterone (T) for 24 h. The production of 17 beta-estradiol (E2) by cells cultivated in T-free media with or without FSH was very low during the culture period. There was a highly significant increase (P less than 0.001) in E2 production when T alone was added and a more marked increase was consistently noted in the presence of FSH and T. In all cases, hCG failed to exert any significant effect on E2 production by granulosa cells in the presence or absence of T. No treatments examined altered the E2 production of thecal cells during a 24-h culture period and the amounts of E2 released into media were negligible when compared with levels produced by granulosa cells from the same follicles. It is concluded that granulosa cells but not thecal cells are the prime site of follicular estrogen production and that FSH regulates estrogen secretion by nonluteinized granulosa cells of the human follicle.     

Effect of human growth hormone on proteoglycan synthesis in cultured rat chondrocytes

We have previously demonstrated that hGH stimulates DNA synthesis in cultured chondrocytes in the absence of serum. The present study is concerned with the effects of hGH on proteoglycan synthesis by cultured chondrocytes. Chondrocytes were isolated from rat rib growth cartilage by collagenase digestion, plated in plastic dishes, transferred to serum-free MCDB 104 medium, and incubated for 24 h to establish growth arrest. The cultures were then preincubated for 0-24 h with various concentrations of hGH and ovine prolactin (oPrl) and finally pulse-labelled for 30 min with radioactive sulphate in the presence of hormone. hGH, but not oPrl, stimulated sulphate incorporation with an apparent maximum at 50 ng/ml (approximately 170%). The stimulatory effect was apparent after 2 h and maximal after 3h preincubation. After 12 h the stimulatory effect had decreased to insignificant levels. Qualitative analysis of isolated proteoglycans indicated that the stimulation of sulphate incorporation by hGH is exerted at the level of protein synthesis with little effect on glycosylation and sulphation. Further experiments are required to demonstrate whether the stimulatory effect on proteoglycan synthesis is a specific phenomenon or represents one aspect of a general stimulation on cell metabolism in preparation for DNA synthesis.     

Effect of various plasminogen activators on prostacyclin synthesis in cultured vascular cells

In this study, we examined the effects of various plasminogen activators on arachidonic acid release and prostacyclin biosynthesis in cultured rat aortic smooth muscle cells and bovine pulmonary artery endothelial cells. Prostacyclin was the major product formed from arachidonic acid in aortic smooth muscle cells and endothelial cells. When intact cells were incubated with streptokinase, one of the plasminogen activators, a significant stimulatory effect on prostacyclin biosynthetic activity in cells was evident without any cellular damage at all concentrations used (1-5,000 units/ml). Streptokinase also caused a marked release of arachidonic acid. However, when it was incubated with cell-free homogenates and [3H]arachidonic acid, it did not show any effects on prostacyclin biosynthesis. The addition of urokinase and tissue-type plasminogen activator had no effect on prostacyclin biosynthesis. Urokinase stimulated the release of arachidonic acid from cells, but it did not show any effect on prostacyclin release at any concentration of urokinase (1-5,000 units/ml). The release of arachidonic acid and the increased prostacyclin synthesis were not observed when tissue-type plasminogen activator was added. These results indicate that, among various plasminogen activators investigated, only streptokinase causes the release of arachidonic acid and prostacyclin. This could be a beneficial effect in thrombolytic therapy.     

Regulation of steroid production in cultured porcine thecal cells by transforming growth factor-beta.

Evidence that transforming growth factor-beta (TGF beta) is produced by porcine thecal cells and acts upon porcine granulosa cells suggests that this peptide may be a local regulator of follicular function in this species. The objective of the present study was to investigate the effects of TGF beta on steroidogenesis in thecal cells from 4-6 mm follicles of prepubertal gilts. In this culture system, cells undergo functional luteinization such that production of androstenedione, the major steroid product in 24 h incubations, declines, and in the presence of luteinizing hormone (LH) (250 ng/ml) and insulin (1 micrograms/ml), progesterone production increases over a 3-day culture period. TGF beta (0.1-10 ng/ml) had no effect on production of androstenedione from endogenous precursors in the presence or absence of LH, although there was a slight inhibition of androstenedione production in the presence of exogenous progesterone (up to 23%). As the cells luteinized in culture, the increase in progesterone production in response to LH increased (day 1, 4.4-fold; day 3, 13-fold). TGF beta at concentrations as low as 0.1 ng/ml caused marked (up to 90%) inhibition of LH-stimulated progesterone production in day 3 cultures. In the presence of TGF beta (10 ng/ml), the response to LH was completely abolished, and the response to dibutyryl cAMP was considerably attenuated (25% of controls). Since the primary site of action of TGF beta appeared to be distal to cAMP formation, the effect of TGF beta on conversion of exogenous 22-hydroxy-cholesterol and pregnenolone to progesterone was determined in day 3 cultures. 22-Hydroxycholesterol and pregnenolone restored progesterone production to at least 80% and 89% of controls, respectively. While the primary inhibitory action of TGF beta appears to be exerted distal to cAMP formation, neither cholesterol sidechain cleavage nor the 3 beta-hydroxysteroid dehydrogenase: delta 5-delta 4 isomerase reactions are primary targets of this factor. Together with evidence of thecal production of TGF beta, the results of this study indicate that this peptide may be an autocrine regulator of thecal steroidogenesis.     

Production and purification of recombinant human inhibin and activin.

Inhibin and activin are protein hormones with diverse physiological roles including the regulation of pituitary FSH secretion. Like other members of the transforming growth factor-beta gene family, they undergo processing from larger precursor molecules as well as assembly into functional dimers. Isolation of inhibin and activin from natural sources can only produce limited quantities of bioactive protein. To purify large-scale quantities of recombinant human inhibin and activin, we have utilized stably transfected cell lines in self-contained bioreactors to produce protein. These cells produce approximately 200 microg/ml per day total recombinant human inhibin. Conditioned cell media can be purified through column chromatography resulting in dimeric mature 32-34 kDa inhibin A and 28 kDa activin A. The purified recombinant proteins maintain their biological activity as measured by traditional in vitro assays including the regulation of FSH in rat anterior pituitary cultures and the regulation of promoter activity of the activin-responsive promoter p3TP-luc in tissue culture cells. These proteins will be valuable for future analysis of inhibin and activin function and have been distributed to the US National Hormone and Peptide Program.