Extrapituitary action of gonadotropin-releasing hormone: direct inhibition ovarian steroidogenesis.

Gonadotropin-releasing hormone (GnRH) and its agonistic analogs inhibited the follicle-stimulating hormone (FSH)-induced increase of estrogen and progesterone production in vitro by rat ovarian granulosa cells. Likewise, GnRH analogs inhibited FSH-induced changes in ovarian function in hypophysectomized rats in vivo. These results indicate that GnRH, in addition to its well-known gonadotropin-releasing action in the pituitary, exerts a direct inhibition of ovarian steroidogenesis.

The role of the oocyte in folliculogenesis.

Novel regulatory proteins have been identified within oocytes that are crucially involved in folliculogenesis. One of the most exciting oocyte signaling molecules is a novel member of the transforming growth factor beta (TGF-beta) superfamily, growth differentiation factor 9 (GDF-9). Loss-of-function studies have established that GDF-9 is obligatory for proper folliculogenesis and fertility in female mice. The current challenges are to understand how oocyte morphogens regulate folliculogenesis and how their actions and interactions are integrated into the overall processes of physiology and pathophysiology. Who would have thought that oocyte morphogens would be so crucial for reproduction?

Insulin-like growth factor-I selectively stimulates cholesterol side-chain cleavage expression in ovarian theca-interstitial cells.

Evidence accumulating in the literature supports the concept that insulin-like growth factor I (IGF-I) may be an important local regulator of ovarian function. Recent studies have demonstrated that IGF-I synergistically augments LH stimulation of theca-interstitial cell (TIC) androgen biosynthesis. The purpose of the present studies was to begin to elucidate the molecular mechanisms of the interaction between IGF-I and LH. TIC were purified from ovaries of hypophysectomized immature rats by Percoll gradient centrifugation. When isolated TIC (5 x 10(6) viable cells per dish) were cultured (4 days) in serum-free medium, low amounts (less than 10 ng/ml) of androsterone were produced. Basal androsterone production was not changed by incubation with IGF-I (30 ng/ml). Treatment with LH (50 ng/ml) caused an 85-fold stimulation of androsterone synthesis that was further increased 2.1-fold by concomitant treatment with IGF-I. Immunoblot analysis demonstrated that untreated TIC contained low levels of 17 alpha-hydroxylase/C17-20 lyase enzyme (P450(17 alpha] that were unchanged by incubation with IGF-I alone. LH treatment increased P450(17 alpha) content 5.5-fold and coincubation with LH plus IGF-I increased P450(17 alpha) content 16-fold above control levels. Cholesterol side chain cleavage enzyme (P450scc) was readily detected in immunoblots from untreated TIC. P450scc content was increased 2.6-fold by LH treatment and 4.2-fold by LH plus IGF-I. Interestingly, IGF-I alone induced a 2-fold increase in P450scc. To determine if the increases in P450scc content were associated with increased enzyme activity, progesterone production was measured.(ABSTRACT TRUNCATED AT 250 WORDS)

Norepinephrine amplifies human chorionic gonadotropin-stimulated androgen biosynthesis by ovarian theca-interstitial cells.

Ovarian theca-interstitial cells, when cultured in serum-free medium, secreted androgens in response to hCG stimulation. This production was dependent on time (maximum production attained after 96 h) and dose (half-maximal effective dose of hCG, 9 ng/ml). When the sympathomimetics norepinephrine, epinephrine, and isoproterenol were added to the medium, androgen production in response to hCG was enhanced by 100-300%. The ability of the catecholamines to stimulate androgen production was dependent on the continuous presence of hCG. Treatment with catecholamines alone did not induce theca-interstitial cells to produce androgens. Catecholamine stimulation of steroid hormone metabolism was selective for intermediates in the delta 4-pathway, with greatest increases occurring in the production of androstenedione and testosterone. It was found that the effect of the catecholamines on androgen production was dependent on both beta 1-and beta 2-adrenergic receptors. The acquisition of catecholamine responsiveness was specific to hCG; if theca-interstitial cells were induced to differentiate with either prostaglandin E2 or cholera toxin, then isoproterenol did not enhance androgen synthesis. The catecholamine-induced increases in androgen production were not due to a granulosa cell contribution of steroid. The interstitial cells are the only steroid-producing cells in the ovary that are directly innervated by norepinephrine-containing fibers of the sympathetic nervous system. Our finding of catecholamine-augmented androgen production provides a direct link between the autonomic nervous system and regulation of ovarian steroid synthesis.

Purification of ovarian theca-interstitial cells by density gradient centrifugation.

The study of the biochemical mechanisms regulating the differentiation of the ovarian androgen-producing cells has been difficult because of the lack of a method for isolating the theca-interstitial cells (TIC) from the granulosa and other cell types. We report here a simple, rapid and reproducible method for obtaining large numbers of highly enriched TIC using density gradient centrifugation. When dispersed cells from ovaries of hypophysectomized immature rats were centrifuged in a Percoll gradient (20-70%), the cells were separated into five incompletely resolved bands. Of these, band V contained the androgen-producing cells. The band V cells produced 3.2- and 3.9-fold more cAMP and androgen, respectively, than the whole ovarian cells and contained 3.8- and 3.5-fold more 125I-hCG binding and 3 beta-HSDH labeled cells, respectively. These data indicated that the TIC were 65% pure. In order to improve the degree of purification, a discontinuous density procedure was used. When the ovarian cells were centrifuged in d = 1.055 g/ml Percoll, the purified TIC produced 4.7- and 5.9-fold more cAMP and androgen than the whole ovarian cells and they contained 5.3-fold more 125I-hCG binding. The TIC were cleanly separated from the granulosa cells and histochemical staining for 3 beta-HSDH activity revealed that the TIC were 93.0 +/- 1.3% pure. This method provides, for the first time, the opportunity to study TIC differentiation under defined conditions without granulosa cell contamination.

Inhibition of synthesis of luteinizing hormone (LH) receptors by a down-regulating dose of LH.

Although LH is known to cause down-regulation of the LH/human CG (hCG) receptor, the mechanisms underlying this process are not well understood. Here we have analyzed the effects of LH on the turnover of LH receptors in cultured granulosa cells. Down-regulation was stimulated in FSH-primed granulosa cells by exposure to LH for 1 h. By 24 h after LH exposure, the LH/hCG receptor content, as measured using [125I]iodo-hCG was 70% less than controls. To determine whether this loss of LH/hCG receptors was due to accelerated receptor degradation, turnover measurements were made using tunicamycin to inhibit LH/hCG receptor synthesis. Under these conditions, there was a small (16%) increase in the rate of degradation of LH/hCG receptors in LH-treated cells compared to controls. By contrast, the rate of LH/hCG receptor synthesis, estimated mathematically, was reduced by 75% in the LH-treated cells. LH treatment had no effect on the incorporation of [35S]methionine into total cellular protein. Withdrawing FSH from the culture medium had a similar effect, slightly enhancing LH/hCG receptor degradation while markedly reducing its synthesis. The role of lysosomes in the degradation of LH/hCG receptors was also investigated. The lysosomotropic amine NH4Cl failed to cause LH/hCG receptors to accumulate, contrary to what would be expected if LH/hCG receptors were degraded lysosomally. In contrast, NH4Cl inhibited the degradation of receptor-bound [125I]iodo-hCG, suggesting that the ligand and the receptor may be handled differently by the granulosa cell. Our results suggest that in the granulosa cell, LH down-regulates its own receptor by inhibiting the synthesis of LH/hCG receptors, possibly by interfering with the ability of FSH to stimulate the synthesis of LH/hCG receptors. Furthermore, they suggest that the degradation of LH/hCG receptors may occur by nonlysosomal mechanisms.

Gonadotropin-releasing hormone induces classical meiotic maturation in subpopulations of atretic preantral follicles.

GnRH has been shown to induce premature meiotic maturation in preantral follicles of the immature estrogen-primed hypophysectomized rat. As these animals are free of circulating gonadotropins and contain large numbers of full-grown oocytes in preantral follicles, we have investigated this model to determine its usefulness in studying meiotic maturation. We show that a maximum dose of the agonist D-Trp6,Pro9,Net-LRF (GnRH-a) induces approximately 25% of full grown oocytes to resume meiosis within a 12-h period. This response is dose dependent (ED50 = 0.24 microgram/rat) and specific for GnRH-a. GnRH-a stimulates germinal vesicle breakdown and first polar body formation within 2 and 8 h, respectively. More than 75% of those oocytes that initiate meiotic maturation reach metaphase II by 15 h. This effect of GnRH parallels the time course of physiological meiotic maturation triggered by LH as well as that of oocytes maturing spontaneously in vitro. Oocytes in primordial and primary follicles do not respond to GnRH. The majority of affected follicles are small tertiary follicles (200-400 micron in diameter) and show signs of atresia. This atresia is not caused by GnRH-a and does not, in itself, result in meiotic maturation, but appears to confer susceptibility to GnRH-a-induced meiotic maturation. Our studies indicate that this animal model will be useful to elucidate further the mechanisms and requirements for meiotic maturation. It will also facilitate investigation of the role of atresia in the GnRH response of tertiary follicles and the issue of follicle heterogeneity within these animals.

The effect of LH/FSH, dibutyryl cyclic AMP, and prostaglandins on the production of estrogens by rabbit granulosa cells in vitro.

The purpose of this study was to establish the effects of trophic hormones on the production of estrogens by rabbit granulosa cells. A pure population of these cells was isolated from preovulatory follicles (1-15 mm in diameter) of estrous rabbits, and cultured for 6 days with either one or a combination of the following hormones: androstenedione, Pergonal (LH/FSH), dibutyryl cyclic AMP (Bu2cAMP), prostaglandin F2alpha (PGF2alpha), or prostaglandin E2 (PGE2). The medium was collected every 2 days and progesterone (P), estrone (E1) and estradiol-17 beta (E2beta) were measured by radioimmunoassay. Granulosa cells cultured as controls (i.e., without exogenous trophic hormones) secreted P spontaneously and its secretion was stimulated 100 to 1,000 fold with LH/FSH and Bu2cAMP, but not with PGF2alpha or PGE2. Androstenedione, either alone or with trophic hormones had no apparent effect on the cytology of the granulosa cells or their ability to secrete P. In the absence of exogenous androstenedione, the cultures produced very small amounts of E1 or E2beta (smaller than 100 pg/ml), either spontaneously or in response to LH/FSH, B12cAMP, PGF2alpha, or PGE2. Incubating granulosa cells with exogenous androstenedione (1 mug/ml) resulted in a 30- to 150-fold increase in E2beta production, which was stimulated an additional 3- to 5-fold with LH/FSH and Bu2cAMP, but not with PGF2alpha or PGE2. In most cultures, E2beta production was restricted to the first 2 days in vitro. Bu2cAMP, however, maintained E2beta production at relatively high levels throughout the duration of the experiment, but there was a progressive decrease in its production. The production of E1 was only 5 percent of E2beta, but the pattern of secretion was similar for both estrogens. These results suggest that cyclic AMP could have a role in regulating the synthesis of estrogens by rabbit granulosa cells.

Stimulation of testosterone production in isolated rabbit thecal tissue by LH/FSH, dibutyryl cyclic AMP, PGE2alpha, and PGE2.

The capacities of isolated rabbit theca and granulosa cells to secrete testosterone were studied in vitro. Large Graafian follicles (1-1.5 mm in diameter) were dissected intact from the ovaries of adult estrous rabbits. Granulosa cells from 4 follicles (50,000 cells) and theca tissue (16 pieces per dish, equivalent to 4 follicles) were cultured separately for 6 days either as controls (without exogenous hormones) or with one of the following agents: 1 lU/ml LH/FSH (Pergonal), 10-3M dibutyryl cyclic AMP (Bu2cAMP), 1 mug/ml prostaglandin F2alpha (PGF2alpha), or 1 mug/ml prostaglandin E2 (PGE2). The media were collected every 2 days, and the testosterone (T) was measured by radioimmunoassay. The control cultures of granulosa cells secreted small amounts of T (700 +/- 317 pg/culture: mean +/-SE) during the first 2 days in vitro, and the addition of LH/FSH, Bu2cAMP, PGF2alpha, or PGE2 did not significantly stimulate T production. After 2 days in vitro, very little T (greater than 200 pg/culture) was produced by control and prostaglandin-treated granulosa cells, whereas those incubated with LH/FSH and Bu2cAMP maintained their initial T production rates. Theca control cultures produced 3 +/- 0.4 ng of T (mean +/- SE) during the first 2 days in 13.6-fold by LH/FSH, 3.6-fold by Bu2cAMP, and 3-fold by PGF2alpha and PGE2- T was not detected in theca cultures after 2 days except in those treated with LH/FSH or Bu2cAMP, which produced 1.5 +/- 0.5 and 1.6 +/- 0.3 ng of T, respectively, at 4 days (mean +/- SE). These results suggest that under the present conditions, pieces of rabbit thecal tissue have a greater capacity to produce T de novo than do isolated granulosa cells, and indicate that T production is transiently stimulated by LH/FSH, Bu2cAMP, PGE2alpha, and PGE2.

Secretion of "inhibin" by rat granulosa cells in vitro.

Rat granulosa cells secrete a substance in vitro that acts directly upon cultured rat pituitary cells to preferentially suppress FSH production. It is inferred from these results that the granulosa cell is a source of ovarian "inhibin" and that the pituitary gonadotroph is a site of "inhibin" action.

Prolactin inhibition of luteinizing hormone-stimulated androgen synthesis in ovarian interstitial cells cultured in defined medium: mechanism of action.

The mechanism by which PRL acts on ovarian interstitial cells to inhibit androgen synthesis was examined using primary cultures of ovarian cells from hypophysectomized immature rats grown in serum-free medium. In the presence of LH, the cultured interstitial cells showed a 200-fold increase in androgen production, of which androsterone was the principal metabolite. The addition of highly purified PRL (100 ng/ml) markedly inhibited (98%) the LH-stimulated androsterone accumulation. The ED50 of PRL action was calculated to be 1.3 +/- 0.4 ng/ml. The inhibition of androsterone production by PRL was rapid (t 1/2 = 75 min), not readily reversible, and could be evoked at any time during the culture period. The effects of PRL on LH-stimulated androgen production were not due to changes in [125I]iodo-hCG binding. LH-stimulated adenylate cyclase, cell number, or cell viability. As with LH, prostaglandin E2, cholera toxin, or 8-bromo cAMP also caused marked increases in androsterone synthesis, and these effects were blocked (98-99%) by PRL. Studies on the metabolism of steroid hormones revealed that PRL decreased LH-stimulated androsterone and 5 alpha-androstane-3 alpha, 17 beta-diol accumulation by 99%, androstenedione by 94%, testosterone by 90%, dehydroepiandrosterone by more than 80%, pregnenolone and 17 alpha-hydroxyprogesterone by 80%, 17 alpha-hydroxypregnenolone by 84%, and progesterone by 71%. Binding experiments demonstrated the presence of a single class of high affinity (Kd = 2.42 x 10(10) M), low capacity (2.37 fmol/10(6) cells) [125I]iodo-PRL-binding sites in the interstitial cells, suggesting that such receptors mediate the inhibitory action of PRL. It is inferred from these results that PRL antagonizes the stimulatory effects of LH on ovarian androgen biosynthesis by inhibiting a step distal to cAMP formation and before or at the cholesterol side-chain cleavage step.

Mechanism by which 17 beta-estradiol inhibits ovarian androgen production in the rat.

The mechanism by which 17 beta-estradiol inhibits ovarian androgen biosynthesis was investigated. Immature 25-day-old rats were treated for 2 days with estradiol, after which whole ovaries were dispersed, and gonadotropin binding and cAMP and steroid hormone production were examined. When dispersed cells from untreated control ovaries were incubated with hCG (100 ng/ml), there were marked increases (10-fold) in steroid production, with the major steroids being progesterone and androstenedione. The stimulation of steroidogenesis by hCG was dose-related (ED50 = 100 pg/ml hCG). After 2 days of estradiol treatment, the maximum hCG stimulation of androstenedione, testosterone, 17 alpha-hydroxypregnenolone, and 17 alpha-hydroxyprogesterone production by ovarian cells was inhibited by 90%; pregnenolone production was unchanged, while progesterone production was increased by 30%. Time course studies showed that the stimulatory effect of hCG on androgen production was maximally inhibited (90%) after 12 h of estradiol treatment. Implanting miniestradiol capsules unilaterally under the ovarian bursa caused a 77% decrease in the hCG stimulation of androgen production by the estradiol-treated cells, while progesterone production was unchanged. hCG-stimulated steroidogenesis in the contralateral ovary was not altered. Estradiol treatment did not affect the binding capacity, the hCG stimulation of cAMP production, or the number of steroid-producing cells in the ovaries. It is concluded from these experiments that exogenous estradiol acts directly on the rat ovary to abolish the hCG stimulation of androgen production by rapidly inhibiting 17 alpha-hydroxylation.

Transforming growth factor-beta promotes differentiation of ovarian thecal-interstitial cells but inhibits androgen production.

Evidence that transforming growth factor-beta (TGF beta) is produced by thecal-interstitial cells (TIC) has suggested the hypothesis that TGF beta may be an autocrine regulator of TIC function. The purpose of these studies is to begin to test this hypothesis. In the present experiments we tested the effects of TGF beta on steroid production by TIC isolated from the ovaries of hypophysectomized immature rats by Percoll gradient centrifugation. When TIC (10(4) viable cells/well) were cultured in serum-free medium (0.2 ml in 96-well plates), low amounts of androsterone (less than 4 ng/ml) were produced at 2, 4, and 6 days. TGF beta (0.01-100 ng/ml) did not change basal androsterone production. Treatment with LH (50 ng/ml) stimulated a 100-fold increase in androsterone at 2 days and 60-fold increases at 4 and 6 days. Concomitant treatment with TGF beta (10 ng/ml) caused a 65% inhibition (ED50 = 2.3 +/- 0.7 ng/ml) of androsterone production at each time period. Analysis of key steroid metabolites demonstrated that androsterone and androstenedione were inhibited equally, while progesterone was significantly increased (ED50 = 1.2 +/- 0.2 ng/ml). Time-course studies revealed that TGF beta alone did not alter progesterone production at 2 days, but markedly increased progesterone (10-fold) above control levels at 4 and 6 days. Dose-response experiments showed that TGF beta did not alter the sensitivity of the TIC to LH stimulation, indicating that LH activation of the intracellular signaling pathway was not blocked by TGF beta. Treatment with insulin-like growth factor-I (IGF-I) together with LH caused a synergistic increase in androsterone production. The synergistic stimulation of LH action by IGF-I could be blocked by TGF beta. Interestingly, TIC were more sensitive to TGF beta in the presence of IGF-I (ED50 = 0.18 +/- 0.04 ng/ml). In contrast, TGF beta enhanced progesterone production only at the highest dose of TGF beta (10 ng/ml). To further elucidate the mechanism of TGF beta action, the effects of TGF beta on the TIC content of 17 alpha-hydroxylase/C17-20 lyase (P450(17)alpha) and cholesterol side-chain cleavage (P450scc) were analyzed by immunoblotting. TGF beta alone or in combination with LH stimulated an increase in P450scc content, but did not alter P450(17 alpha content. These results lead us to conclude that 1) the TIC are targets for TGF beta; 2) IGF-I increases the sensitivity of TIC to TGF beta action; and 3) TGF beta acts directly on TIC to stimulate progesterone while inhibiting androgen production.

A specific FSH receptor in rat granulosa cells: properties of binding in vitro.

The specific binding of [125I]iodoFSH to granulosa cells collected from immature, hypophysectomized, DES-treated rats, was studied in vitro. Specific binding occurred after 5 min and reached maximum after 3 h of incubation at 37 C. Non specific binding was very low (less than 10% of the total binding). The [25I]iodoFSH remained tightly associated with the receptor at pH 7.5, but was rapidly dissociated at pH 5. Unlabeled hFSH competitively inhibited [125I]iodoFSH binding. Kinetic analyses of equilibrium binding experiments gave an apparent association constant (Ka) of 1.34 (+/- 0.31) x 10(10)M-1 [mean (+/- SE)] and a number of binding sites per cell of (NB) 1, 130 +/- 70 (mean +/- SE). Rat prolactin, wheat germ agglutinin, and concanavalin-A did not compete with [125I]iodoFSH, but hLH, hCG, and rTSH competed at doses 300- to 900-fold higher than those of hFSH. Granulosa cells isolated from adult DES-treated rats, as well as cells collected from medium and preovulatory follicles of proestrous rats, gave Ka and NB values similar to those described above. A comparative study of rabbit granulosa cells indicated a much lower binding affinity compared with those from the rat.

Activin causes premature superovulation.

In the rat ovary, follicle cells produce and respond to activin, but as yet, the functional significance of the autocrine/paracrine effects of ovarian activin remains equivocal. To assess the effects of activin on folliculogenesis, normal cycling female rats were injected once every 8 h over a 40-h period with recombinant human activin A (120 micrograms/kg) beginning at 1300 h on estrus (day 1 of treatment). A total of 10 rats were injected with activin in 2 separate experiments. On days 3 and 4 of treatment, blood was obtained for hormone measurements, and the ovaries were removed for histology. Follicle counts were performed in 1 ovary from 3 representative animals in each treatment group. All antral (Graafian) follicles 300 microns or more in diameter were measured and classified as healthy or atretic based on the number of pyknotic nuclei in the largest cross-section. On day 3, all rats were in diestrus (diestrous day 2). After 3 days of activin administration, serum levels of estradiol were increased 200%, progesterone levels were decreased 67%, and FSH levels were unchanged compared with those in matched controls. By day 4 (i.e. 1 day after the last injection), no changes in the levels of these hormones were observed. Injection of activin for 3 days did not change the total number of antral follicles per ovary (control, 41.3 +/- 4.9; activin, 43.7 +/- 3.9); however, activin significantly increased the total number of atretic follicles (control, 69%; activin, 92%). Morphometric analysis of the ovaries removed on day 3 showed a marked increase (2-fold) in the number of large follicles, but most (89%) were atretic. Follicle counts suggested that the additional large follicles may have come from the pool of healthy small follicles. Histological studies showed that some of the day 3 activin-treated follicles had initiated ovulation. On day 4, control and activin-treated animals were at proestrus and estrus, respectively. Therefore, activin shortened the estrous cycle by 1 day. Little or n change in the follicle populations was observed in day 4 control ovaries. Interestingly, however, in 2 of the 3 activin-treated animals, 1 set of large follicles had ovulated (12 +/- 1 expanded egg cumulus complexes/oviduct), and another set (13 +/- 2) was just about to rupture. The activin-exposed oocytes (tubal and follicular) appeared arrested in metaphase I. After ovulation, activin-treated follicles developed into typical corpora lutea. No ovulations were found in control animals.(ABSTRACT TRUNCATED AT 400 WORDS)

Evidence for widespread atresia in the hypophysectomized estrogen-treated rat.

The question considered in this study concerns to what extent the ovaries of the immature hypophysectomized diethylestilbestrol-treated rat are populated by atretic follicles. Histological studies of ovaries on postoperative days 4-7 reveal the presence of numerous follicles showing signs of ill health. The granulosa cells constituting these follicles are heterogeneous and exist in three distinct zones. The cells in zone 1 are affixed to the basal lamina and form a tightly packed sheet of cuboidal epithelium one to two cells thick. The cells constituting zone 2 are located just medial to zone 1 and appear as a mass of loosely aggregated cells which show intense necrosis. Zone 3 is composed of the prospective cumulus-oocyte complex which appears completely normal. Results of histochemical studies show heterogeneity in the granulosa cells with respect to the presence and location of markers specific for atresia. In atretic follicles, the granulosa cells populating zones 1 and 2 contain lipid droplets and esterase activity, but are negative for the lysosomal enzyme acid phosphatase. The exact reverse is true for zone 3, where the cumulus express intense acid phosphatase activity but are negative for lipid and esterase activities. The granulosa cells in healthy follicles are totally negative for these histochemical markers. As a healthy preantral follicle grows, there is a dramatic increase in the mitotic index of the granulosa cells; however, in all atretic follicles, granulosa cell division is decreased by more than 90%. Further experiments using monoclonal antibodies as probes confirm the presence of atretic follicles and demonstrate that the structure-function changes in zones 1 and 2 are associated with striking changes in the antigenic properties of the granulosa cells. Morphometric studies indicate there are a total of 235 +/- 29 preantral follicles/ovary in the hypophysectomized diethylstilbestrol-treated rat. On postoperative day 4, 37 +/- 1.2% of the preantral follicles are atretic; this number increases linearly with time posthypophysectomy, reaching nearly 50% on postoperative day 7. Examination of follicles at different stages of development reveals that atresia is rare in the very early stages, e.g. follicles less than 200 microns in diameter; however, in the population of larger preantral follicles (those 200-400 microns in diameter), atresia is common, with nearly 80% being atretic on postoperative day 7. Collectively, these findings provide convincing evidence of widespread atresia in the ovaries of the hypophysectomized diethylstilbestrol-treated rat.(ABSTRACT TRUNCATED AT 400 WORDS)

Localization of insulin-like growth factor-binding protein-5 messenger ribonucleic acid in rat ovaries during the estrous cycle.

To investigate the potential role of insulin-like growth factor-binding protein-5 (IGFBP-5) in ovarian physiology, we employed in situ hybridization and Northern analysis to localize IGFBP-5 mRNA in rat ovaries during the estrous cycle. By Northern analysis, the mRNA was abundant at all stages of the cycle. Two species of mRNAs were detected, with sizes of 6.0 and 1.8 kilobases, respectively. The relative amounts of the two transcripts changed throughout the cycle. By in situ hybridization, IGFBP-5 mRNA was expressed in only a few cell types: 1) granulosa cells of some atretic follicles, 2) some secondary interstitial cells, 3) some corpora lutea, and 4) the surface epithelium. The levels of message in both the granulosa and secondary interstitial cells changed over the cycle. At 1000 h on proestrus (before the LH/FSH surge), the message was expressed in only a few follicles. Interestingly, all were small atretic preantral (200-250 microns) follicles. At 2000 h on proestrus (after the LH/FSH surge), the IGFBP-5 mRNA was more abundant; now almost every atretic preantral follicle showed a strong hybridization signal. At 0200 and 1000 h on estrus, the mRNA appeared for the first time in granulosa cells of some atretic antral follicles and in secondary interstitial cells. Hence, virtually all atretic follicles, preantral and antral, now showed IGFBP-5 gene expression. In contrast to that on proestrus and estrus, the hybridization signal on diestrous days 1 and 2 was much less prominent and was found in only a few atretic preantral follicles. Throughout the cycle, IGFBP-5 mRNA was evident in some corpora lutea, but it was not particularly prominent. Abundant IGFBP-5 mRNA was evident in the surface epithelium, and no change was detected over the cycle. Dominant follicles were devoid of IGFBP-5 mRNA. In conclusion, this paper presents the first evidence that the IGFBP-5 gene is expressed in the adult rat ovary. The IGFBP gene is expressed in a cell-specific manner, e.g. in atretic granulosa, secondary interstitial cells, corpora lutea, and the surface epithelium, and the stage of the cycle significantly affected message levels, especially in atretic granulosa and secondary interstitial cells around estrous morning. These findings suggest that IGBP-5 may be an autocrine/paracrine regulator of ovarian physiology, particularly in relation to preantral follicle atresia.

Tissue-specific expression of four insulin-like growth factor-binding proteins (1, 2, 3, and 4) in the rat ovary.

We investigated the possible presence of the mRNAs encoding four insulin-like growth factor-binding proteins (IGFBP-1, -2, -3, and -4) in the rat ovary. It has been demonstrated previously by Northern analysis that adult rat ovaries contain mRNA transcripts for IGFBP-2 and -3. Here we show by Northern analysis that adult rat ovaries contain mRNA for IGFBP-4, but the mRNA for IGFBP-1 was below the limit of detection. Using in situ hybridization, IGFBP-1 mRNA was not detected in any of the ovaries tested. The IGFBP-2 mRNA was localized specifically in thecal interstitial cells (TIC) and secondary interstitial cells of all ovaries. The IGFBP-2 signal was very strong in TIC of all Graafian follicles (healthy and atretic) and very strong in all secondary interstitial cells located in different regions of the ovary, but weak in TIC of preantral follicles. The IGFBP-3 hybridization signal was localized specifically to some corpora lutea. Here the hybridization single for IGFBP-3 was strong and distributed throughout the corpus luteum, strong but localized to only some luteal cells, or not detectable. The IGFBP-4 mRNA was localized almost exclusively to granulosa cells of atretic follicles. The intensity of the IGFBP-4 signals appeared to increase as the level of atresia increased. In some cases, IGFBP-4 signals were detected in the TIC, but they were weak and variable. These results show that the mRNAs for IGFBP-2, -3, and -4 are localized to the interstitial cells, corpora lutea, and atretic granulosa cells, respectively. The tissue-specific synthesis of IGFBP subtypes in specialized ovarian cells provides an excellent system to study the manner in which IGFBP synthesis is controlled and their potential role as an autocrine and paracrine factors.

Cyclic changes in insulin-like growth factor-binding protein-4 messenger ribonucleic acid in the rat ovary.

We have previously shown that the mRNA for insulin-like growth factor-binding protein-4 (IGFBP-4) is present in adult rat ovaries, being localized predominantly to granulosa cells of atretic follicles. Now we have considered the following questions. What class of atretic follicles expresses IGFBP-4 mRNA? How does IGFBP-4 mRNA expression change during the estrous cycle? In keeping with our earlier work, a strong hybridization signal for IGFBP-4 mRNA was present in subpopulations of follicles throughout the estrous cycle. In all cases, the hybridization signal was localized to granulosa cells. Among the various types of follicles, IGFBP-4 mRNA was present almost exclusively in atretic graafian (antral) follicles. Morphologically, the outer layer of granulosa cells was positive, while cells in the cumulus oophorous were negative. By Northern analysis and in situ hybridization, the levels of IGFBP-4 mRNA were found to change over the estrous cycle. At 1000 h on proestrus (before the LH/FSH surge), the hybridization signal was relatively weak, being restricted in some (but not all) atretic Graafian follicles. At 2000 h on proestrus, (after the LH/FSH surge), essentially all atretic Graafian follicles were strongly positive for the message. The pattern of hybridization was similar at 0200 h on estrus, but the signal was less intense. At 1000 h on estrus, the hybridization signal was variable, ranging from very strong to weak or undetectable in atretic follicles. At this stage, however, the highest levels of IGFBP-4 mRNA were measured by Northern analysis; interestingly, a strong signal became apparent in the stromal cells. On diestrous day 1, the message levels decreased, and the signal was restricted to some atretic follicles. On diestrous day 2, the hybridization signal was very weak. There was virtually no detectable IGFBP-4 mRNA in any healthy follicle. In summary, we found that IGFBP-4 mRNA is 1) not detected in healthy dominant follicles; 2) localized almost exclusively to atretic Graafian follicles, except on estrus when it also appears in stromal cells; 3) localized predominantly to the mural granulosa cells in atretic follicles; and 4) undergoes changes during the cycle, being most prominent around estrous morning. The possibility that IGFBP-4 plays a role in the cyclic destruction of cohort Graafian follicles at estrus, perhaps by mechanisms involving hormones, is discussed.

Direct inhibitory effect of GnRH on androgen secretion by ovarian interstitial cells.

We report here 1) that GnRH and a potent GnRH agonist inhibit basal and LH stimulated ovarian androgen biosynthesis in vitro, 2) that the inhibitory effects of GnRH are dose-dependent and completely inhibited by concomitant treatment with a GnRH antagonist and 3) that the GnRH action is very rapid (t1/2 = 10 min) and persists after thorough washing of the cells. These data suggest that GnRH acts at a stereospecific binding site on ovarian theca and/or interstitial cells (probably the theca cells) to rapidly and perhaps irreversibly inhibit androgen biosynthesis.