Search

Within 2 days of birth, the mouse ovary is mainly composed of oocytes surrounded by a few pregranulosa cells forming primordial follicles that remain quiescent until they are recruited by intraovarian or other unknown factors to initiate growth of the oocyte and proliferation of the attendant granulosa cells. However, the role of the oocyte in this early development and organization of the follicle is poorly understood. The Dazl knockout (-/-) mouse in which there is total ablation of oocytes in fetal life has allowed us to address this issue. Ovaries from -/- females lack any follicular structure and have no cells positive for either Mullerian inhibiting factor or sulfated glycoprotein-1, indicating a lack of small follicles or corpora lutea. However, by immunocytochemistry, there are cells positive for 3beta-hydroxysteroid dehydrogenase, 17alpha-hydroxylase, and aromatase, indicating the presence of steroidogenically active cells capable of producing estrogen. This was confirmed by the presence of hypertrophied uterine endometrium expressing both estrogen receptor alpha (ER alpha) and ER beta together with normal levels of plasma estradiol. In addition, these steroidogenically active cells contain ER beta, inhibin alpha, and betaB-subunits, and -/- mice have low measurable plasma inhibin A and B levels. The ovarian steroids and inhibins had no significant effect on either plasma or pituitary gonadotropin levels, with significantly (P < 0.01) lower LH and FSH in intact +/+ and +/- females. However, significantly (P < 0.05) increased plasma inhibin B together with significantly (P < 0.05) lower FSH were observed in the +/- females. In conclusion, our data showed that despite oocyte loss in fetal life, the adult ovaries contained steroidogenically active cells capable of producing estradiol and inhibin. Furthermore, in the +/- mice, the enhanced plasma inhibin B implies a role for Dazl protein within the oocyte either from more small follicles or increased inhibin B production from each follicle.

An important source of androgens in the human prostate are those synthesized locally from the inactive adrenal precursor dehydroepiandrosterone (DHEA) and its sulfated derivative DHEA-S. Three beta-HSD (hydroxysteroid dehydrogenase) converts DHEA into androstenedione (4-dione), whereas type 5 17beta-HSD catalyzes the reduction of 4-dione into testosterone in the human prostate and other peripheral intracrine tissues. In the present study, we have used two complementary approaches, namely in situ hybridization and immunocytochemistry, to identify the cells that contain the type 5 17beta-HSD messenger RNA and enzyme in human benign prostatic hyperplasia (BPH). Localization of 3beta-HSD and of the androgen receptor (AR) was also investigated by immunostaining in the same tissue. To find out whether there are any differences between BPH and normal prostate tissue, the localization of type 5 17beta-HSD was reexamined by immunocytochemistry in the normal human prostate samples and also in normal prostate epithelial cell line (PrEC). The in situ hybridization results obtained with a tritiated uridine triphosphate (3H-UTP)-labeled type 5 17beta-HSD riboprobe are in agreement with the immunostaining data obtained with a specific antibody to the enzyme. The immunostaining results obtained from normal prostate tissue and BPH were found to be similar. Thus, in the glandular epithelium, basal cells highly express the messenger RNA and the enzyme, whereas luminal cells show a much lower and variable level of expression. In the stroma and walls of blood vessels, fibroblasts and the endothelial cells lining the blood vessels show positive staining. Similar results are observed when the cellular distribution of 3beta-HSD is investigated. AR immunoreactivity, however, shows a different distribution because, in the epithelium, most of the nuclei of basal cells are negative, whereas the majority of nuclei of the luminal cells show positive staining. A strong reaction for AR is also found in most stromal cell nuclei and in the nuclei of most endothelial cells, as well as in some other cells of the walls of blood vessels. In conclusion, human type 5 17beta-HSD, as well as 3beta-HSD, are highly expressed, not only in the basal epithelial cells and stromal fibroblasts but also in the endothelial cells and fibroblasts of the blood vessels. AR, on the other hand, is highly expressed in the luminal cells. The present data suggest that DHEA is transformed in the basal cells of the glandular epithelium into 4-dione by 3beta-HSD and then into testosterone by type 5 17beta-HSD, whereas dihydrotestosterone is synthesized in the luminal cells after diffusion of testosterone from the underlying layer of basal cells. The potential role of androgen formation and action in blood vessels is unknown and opens new avenues of investigation for a better understanding of the multiple roles of androgens.

In this study the luteotropic activity of mouse placental lactogen I (mPL-I) at midpregnancy was assessed using in vivo and in vitro methodologies. Ovaries from 10-day pregnant mice were enzymatically dispersed and plated on fibronectin-coated wells in serum-free medium. The percentage of ovarian cells that stained for the presence of 3 beta-hydroxysteroid dehydrogenase activity was 24.4 +/- 2.7% at the time of plating and remained constant (26.1 +/- 5.0%) after a 20-h attachment period. Two types of 3 beta-hydroxysteroid dehydrogenase-staining cells, with distinct differences in size and morphology, were present in the culture. Large luteal cells (26-45 microns) were characterized by a small round nucleus and spherical shape with abundant cytoplasm. In contrast, small luteal cells (< 20 microns) were stellate, with little cytoplasm and a large oval nucleus. Basal progesterone secretion was maintained without a change in cellular DNA content and cell number for 168 h of culture. Treatment of ovarian cells with mPL-I (0.05-10 micrograms/ml) caused a dose-dependent increase in the progesterone concentration in the medium. The magnitude and time course of mPL-I-stimulated progesterone accumulation in culture were dependent on the time after plating that mPL-I treatment was initiated. The effects of mPL-II and mouse PRL (mPRL) on progesterone production were similar to those of mPL-I. The ability of sera from 10-, 14-, and 17-day pregnant mice to maintain progesterone production in bromocryptine-treated hysterectomized mice was also examined. Mice were hysterectomized on day 9 of pregnancy, and serum progesterone, mPL-I, mPL-II, and mPRL concentrations were measured 72 h later. Twice daily injections of 0.5 ml day 10 pregnancy serum maintained the circulating progesterone concentration at values not different from those present at the time of hysterectomy. In contrast, serum progesterone concentrations were not maintained in mice treated with serum of 14- or 17-day pregnant mice or with saline. Depletion of mPL-I from day 10 pregnancy serum by affinity chromatography on an anti-mPL-I column removed all luteotropic activity, as determined by the inability of this modified serum to maintain the serum progesterone concentration in bromocryptine-treated hysterectomized mice. A similar pool of day 10 pregnancy serum chromatographed on a nonspecific IgG control column did maintain progesterone production, but at somewhat lower concentrations than those present at the time of surgery. These studies offer direct evidence that mPL-I and mPL-II are luteotropic and support progesterone production at midpregnancy in the mouse.

Leydig cells are the major source of androgens in the male, and it is likely that these cells are also targets for androgen action. The role of androgen action in regulating the development of Leydig cell function has been determined using the testicular feminized (Tfm/Y) mouse, which lacks androgen receptors. In Tfm mice, the testes fail to descend at the normal time (25 days), and testicular descent was, therefore, surgically prevented in control (+/Y) animals. The activity of 3 beta-hydroxysteroid dehydrogenase increased during development from 5-40 days in control and Tfm mice, with no significant difference between the groups. In control animals, 17 alpha-hydroxylase activity was low from 5-25 days, at which time there was a 26-fold increase up to 40 days. In the Tfm group, 17 alpha-hydroxylase activity was 4-fold higher than that in controls on day 5, but showed no change in activity after 25 days and remained at neonatal levels up to 40 days. The activity of 17-ketosteroid reductase in the control animals showed a developmental pattern similar to that of 17 alpha-hydroxylase, with a marked increase in activity after 25 days. In the Tfm group, 17-ketosteroid reductase was normal on day 5, but failed to show any significant change thereafter and remained at neonatal levels on day 40. Serum LH levels in control animals increased from 5 days to a peak at 30 days. In Tfm mice, LH levels were significantly increased on days 20 and 40, but did not differ from controls on days 5, 25, and 30. In control +/Y animals, in which normal testicular descent was allowed to proceed at 25 days, the pattern of development was similar to that in the cryptorchid +/Y animals, although the increase in 17 alpha-hydroxylase activity at 30 and 40 days was significantly greater. The results show 1) that fetal Leydig cell steroidogenesis is enhanced in the absence of androgen receptors, but 2) that adult Leydig cells require receptor-mediated androgen activity around day 25 for normal functional development. In addition, 3) the lack of testicular descent at 25 days reduces the pubertal rise in 17 alpha-hydroxylase activity.

The activities of the 5 alpha-reductase [the enzyme that converts testosterone into dihydrotestosterone (DHT)] and 3 alpha-hydroxysteroid dehydrogenase [the enzyme that converts DHT into 5 alpha-androstan-3 alpha, 17 beta-diol (3 alpha-diol)] have been evaluated in primary cultures of neurons, oligodendrocytes, and type 1 and 2 astrocytes obtained from fetal or neonatal rat brain. All cultures were used on the fifth day. The formation of DHT and 3 alpha-diol was evaluated by incubating the different cultures with [14C]testosterone or [14C]DHT as substrates. The results obtained indicate that the formation of DHT takes place preferentially in neurons; however, type 2 astrocytes and oligodendrocytes also possess considerable 5 alpha-reductase activity, while type 1 astrocytes show a much lower enzymatic activity. A completely different localization was observed for 3 alpha-hydroxysteroid dehydrogenase. The formation of 3 alpha-diol appears to be mainly, if not exclusively, present in type 1 astrocytes. 3 alpha-Diol is formed in very low yields by neurons, type 2 astrocytes and oligodendrocytes. The compartmentalization of two strictly correlated enzymes (5 alpha-reductase and 3 alpha-hydroxysteroid dehydrogenase) in separate central nervous system cell populations suggests the simultaneous participation of neurons and glial cells in the 5 alpha-reductive metabolism of testosterone and possibly other hormonal steroids (e.g. progesterone, corticoids, etc.).

The enzyme complex delta 5-3 beta-hydroxysteroid dehydrogenase/delta 5-delta 4 isomerase (3 beta HSD) is involved in the biosynthesis of all classes of steroids, namely glucocorticoids, mineralocorticoids, progesterone, and sex steroids. To obtain information on the precise localization of 3 beta HSD in rat gonads and adrenal glands, two complementary cytochemical techniques were used; immunocytochemical localization was achieved with antibodies developed against purified human placental 3 beta HSD, while 3 beta HSD mRNA localization was achieved by in situ hybridization performed with a recently cloned rat 3 beta HSD cDNA. In the testis, specific immunostaining was restricted to the cytoplasm of the interstitial cells, while by in situ hybridization, specific silver grains were also seen over the interstitial cells. In the ovary, immunostaining was found in the cytoplasm of cells of the corpus luteum and theca interna, while the granulosa cells of the follicles showed no positive reaction. By in situ hybridization, a specific hybridization signal was observed over granulosa cells of the corpus luteum, which are mainly responsible for progesterone secretion, and to a lesser extent over theca interna cells, known for their role in secreting C19 androgens. In the adrenals, the three zones of the cortex were equally immunolabeled, whereas no staining could be detected in the medulla. Similarly, by in situ hybridization, silver grains were located over the zona glomerulosa, fasciculata, and reticularis, while no specific autoradiographic reaction could be observed on the chromaffin cells of the medulla. The present study provides new information about the precise cellular localization of 3 beta HSD in the adrenal glands and gonads in the rat, thus providing useful information about the site of action of 3 beta HSD, especially in the gonads. Moreover, the approaches used for localization studies, especially quantitative in situ hybridization, should provide a useful tool for assessing the role of hormones on 3 beta HSD expression in the different compartments of the gonads and adrenal glands.

The enzyme complex 3 beta-hydroxy-5-ene-steroid dehydrogenase/delta 5-delta 4-isomerase (3 beta HSD) is involved in the biosynthesis of all classes of active steroids, namely glucocorticoids, mineralocorticoids, progesterone, and sex hormone steroids. To obtain precise information about the cellular and subcellular localization of 3 beta HSD in the gonads and adrenals of the guinea pig, we have proceeded to immunocytochemical localization of the enzyme using antibodies developed against purified human placental 3 beta HSD. In the testis, specific immunostaining was restricted to the cytoplasm of the interstitial cells. In the ovary, on the other hand, immunostaining was found in the cytoplasm of the cells of the corpus luteum and theca interna. The granulosa cells of the follicles were not immunolabeled, with the exception of one layer of cells in contact with the zona pellucida. This restricted labeling was observed in growing and mature follicles, but not in primordial and small primary follicles. In the adrenals, the three zones of the cortex were labeled, whereas no staining could be detected in the medulla. Contrary to findings in the testis and ovary, staining was found in both the cytoplasm and nuclei of adrenocortical cells. The present data suggest a functional interaction between the ovocyte and the specialized layer of 3 beta HSD-containing granulosa cells which might play a role in ovocyte maturation.

The fetal zone of the human fetal adrenal (HFA) gland is established to have decreased 3 beta-hydroxysteroid dehydrogenase/delta 4-5 isomerase (3 beta HSD) activity compared to the neocortex or definitive zone. 3 beta HSD activity, however, can be induced in primary cell culture through treatment with ACTH. Therefore, the HFA with two distinct steroidogenic zones with differences in 3 beta HSD activity as well as the capacity to increase 3 beta HSD activity in response to ACTH provides an excellent model to study the regulation of this enzyme. The presence of 3 beta HSD in the fetal and neocortex zones of the HFA was examined using a polyclonal antibody raised against purified human placental microsomal 3 beta HSD. After homogenates of the fetal and neocortical zones of the HFA were electrophoresed on a sodium dodecyl sulfate-polyacrylamide gel and immunoblotted, the presence of the 3 beta HSD protein with a molecular size of 45 kDa could be demonstrated only in the neocortical zone. ACTH treatment (greater than 2 days) of fetal and neocortical zone explant cultures produced increases in cortisol secretion associated with the respective levels of immunodetectable 3 beta HSD protein. Cortisol and dehydroepiandrosterone sulfate were the respective principal steroid products of neocortical and fetal zone explants. After ACTH treatment, immunodetectable 3 beta HSD was induced to a greater magnitude in the neocortex. These findings provide evidence that the lack of 3 beta HSD activity in the fetal zone, previously considered to be the result of the presence of an endogenous inhibitor, is due to an absence of the protein in this portion of the gland. The lack or minimal expression of 3 beta HSD in the fetal zone of HFA may be due to the action (or lack thereof) of a tissue-specific factor regulating the synthesis of 3 beta HSD.

The enzymes involved in conversion of pregnenolone to testosterone in Leydig cell tumors showed a wide distribution among smooth endoplasmic reticulum (SER), rough endoplasmic reticulum (RER), and cytosol, while these enzymatic activities in normal testes were associated primarily with smooth endoplasmic reticulum. Progesterone, used as a substrate in the presence of an NADPH-generating system, was metabolized to androstenedione and finally to testosterone by microsomes from some strains of tumor which did not form testosterone from exogenous labeled androstenedione. Treatment of microsomal membranes from normal testes with 0.1 M Ca++ and Mg++ caused a marked decrease in 17 beta-dehydrogenase activity, measured as conversion of exogenous [3H]androstenedione to [3H]-testosterone, without serious effects on activities of 3 beta-ol-dehydrogenase or 17 alpha-hydroxylase. Studies of initial velocity kinetics showed that treatment with magnesium ion resulted in a marked reduction in affinity of androstenedione for 17 beta-dehydrogenase while the maximum velocity was the same as in untreated microsomes. Also, experiments using [14C]progesterone and [3H]androstenedione simultaneously as substrates demonstrated that treatment with Mg++ ion made it more difficult for exogenous [3H]androstenedione to reach the active site of 17 beta-ol-dehydrogenase than [14C]androstenedione formed in the microsomal membrane from [14C]progesterone. Microsomal proteins were more easily solubilized and 3 beta-ol-dehydrogenase was more severely influenced by Mg++ ion in tumor membranes than in normal microsomes.

Both estradiol and progesterone may act locally to modulate ovarian function in various species. This study examined the distribution of estradiol and progesterone receptors (ER and PR, respectively) within the primate ovary throughout the menstrual cycle. Ovaries were collected from rhesus or cynomolgus monkeys during the early, mid-, and late (n = 3-6/stage) follicular and luteal phases of the cycle. The tissues were processed for indirect immunocytochemical localization of receptors with specific monoclonal antibodies against ER (H222 and D75) and PR (JZB39). Specific immunocytochemical staining, as determined by comparing adjacent tissue sections incubated with either receptor antibodies or a nonspecific antibody, was exclusively nuclear. Both ER and PR were localized in the germinal epithelium of ovaries at all stages of the cycle. ER was not detected in any other ovarian structure (i.e. stroma, follicles, interstitial tissue, or corpora lutea) regardless of the stage of development. However, ER was detected in other estrogen-responsive tissues, e.g. the oviduct of the monkey and corpora lutea of the pseudopregnant rabbit. In the monkey ovary, PR was detected in stromal and interstitial tissues as well as theca interna and externa of healthy and atretic follicles at all stages of the cycle. The granulosa cells of some primordial and primary follicles demonstrated staining for PR. However, the granulosa layer of follicles that developed beyond the primary stage were consistently negative for PR. Only the granulosa layer of large preovulatory follicles that showed signs of luteinization after the LH surge showed staining for PR equivalent to that in the theca. Monkey corpora lutea exhibited specific nuclear staining for PR. Moreover, the percentage of receptor-positive nuclei in the corpus luteum varied (P less than 0.05) between the early (28 +/- 3%), mid (48 +/- 1%)-, and late (4 +/- 2%) luteal phase of the cycle. Nonfunctional (serum progesterone less than 0.5 ng/ml) regressing corpora lutea did not exhibit for staining for PR. Luteal cells that were PR positive also contained histochemically detectable 3 beta-hydroxysteroid dehydrogenase. These data are consistent with the concept of a receptor-mediated autocrine or paracrine role for progestins, but not estrogens in the gametogenic and endocrine functions of the primate ovary throughout the menstrual cycle.

Enzymatically dispersed testis cells derived from 7-day-old male rats maintained their gonadotropin-stimulated testosterone production for 18 days in culture. Treatment with hCG or LH stimulated androgen production in a dose-dependent manner, with ED50 values of 0.030 +/- 0.007 and 1.0 +/- 0.4 ng/ml for hCG and LH, respectively. Concomitant treatment with a phosphodiesterase inhibitor further enhanced LH action. In contrast, treatment with FSH, GH, or PRL was without effect. Treatment with forskolin, cholera toxin, or 8-bromo-cAMP induced dose-dependent increases in testosterone biosynthesis; this was accompanied by stimulation of 3 beta-hydroxysteroid dehydrogenase activity after treatment with hCG, forskolin, or 8-bromo-cAMP. RIA measurement of different androgens in HPLC-fractionated medium revealed that the main androgen secreted by the neonatal testis cells was testosterone, with lower production of 5 alpha-androstane-3 alpha,17 beta-diol and negligible 5 alpha-dihydrotestosterone, androstenedione, and androsterone. Treatment with epidermal growth factor, GnRH, and arginine vasopressin (AVP) decreased hCG-induced testosterone biosynthesis. Since the inhibitory actions of GnRH and AVP were blocked by concomitant addition of specific hormone antagonists, their inhibitory actions were probably mediated by specific testis receptors. In contrast, treatment with several potent synthetic steroid hormone analogs [diethylstilbestrol (an estrogen), dexamethasone (a glucocorticoid), R5020 (a progestin; 17,21-dimethyl-19-nor-4,9-pregnadiene-3,20-dione), R1881 (an androgen; 17 beta-hydroxy-17 alpha-methyl-4,9,11-estratrien-3-one), or cyproterone acetate (an antiandrogen; 17 alpha-acetyloxy-6-chloro-1,2-dihydro-(1 beta,2 beta)3'-H-cyclopropa-(1,2) pregna-1,4,6-trien-3,20-dione)] did not affect testosterone biosynthesis in hCG-treated cells. These results demonstrate that testosterone production by neonatal testis cells is maintained by gonadotropins during prolonged culture; the ability of cAMP-generating drugs and a cAMP analog to mimic gonadotropin actions on testosterone biosynthesis and 3 beta-hydroxysteroid dehydrogenase activity suggests a mediatory role of cAMP in gonadotropin action; and AVP, epidermal growth factor, and GnRH, through their putative testis receptors, directly inhibit gonadotropin-stimulated testosterone synthesis, while various steroids (androgens, estrogens, progestins, and glucocorticoids) do not affect Leydig cell function in the neonatal testis. The present culture system offers a unique model for elucidating the hormonal control of Leydig cell androgen biosynthesis during neonatal development.