| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Biological Chemistry, The Alexander Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem 91904, Israel
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
follicle-stimulating hormone, gene regulation, granulosa cells, insulin-like growth factor receptor
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cytochrome P450scc is an inner mitochondria membrane protein that catalyzes the conversion of cholesterol to the first steroid, pregnenolone [1–4]. For this purpose, free cholesterol needs to be translocated across the outer mitochondrial membrane to become available as a substrate at the site of P450scc [5]. Transfer of cholesterol into the mitochondria is facilitated by a novel protein recently identified by Stocco and colleagues and named steroidogenic acute regulatory (StAR) protein [6–8]. Loss-of-function mutations in the StAR gene are lethal and cause congenital lipoid adrenal hyperplasia that is characterized by an almost complete inability of the newborn infant to synthesize steroids [9–11]. Once produced, pregnenolone diffuses from the mitochondria to the cytoplasm, where 3ß-HSD converts it to the first biologically active hormone, progesterone [12]. In most of the steroidogenic cell types, 3ß-HSD is localized in the endoplasmic reticulum. An exception to this rule is found in cells of the adrenal cortex, where as much as 40% of the enzyme is also localized inside the mitochondria [13].
What is currently known about IGF-1 and ovarian steroidogenesis? Earlier studies have suggested that the granulosa cells serve as the predominant site of ovarian IGF-1 production, reception, and action [14–17]. In addition to its growth-promoting activity [18–20], studies of cultured granulosa cells have suggested that the most important role of IGF-1 appears to be contingent on its ability to synergize with gonadotropins and to amplify their steroidogenic output (i.e., to promote progesterone and estradiol synthesis) [21–25]. Less is known about the molecular mechanism of IGF-1 action. Effects of IGF-1 on the level of P450scc, StAR, and 3ß-HSD expression were studied in pig granulosa cells [24, 25], but information about the regulation of these genes in ovarian cells of other species is rather limited. For example, IGF-1 can enhance StAR expression in human granulosa-lutein cells [26] but not in cells of human and bovine corpora lutea [27, 28]. In prepubertal rat granulosa cells, we have shown that StAR promoter is trans-activated by CCAAT enhancer-binding protein ß (C/EBPß), the level of which is FSH-inducible but does not require IGF-1 [29]. Other studies of immature rat granulosa cells have recently suggested that IGF-1 augments steroidogenesis by a concerted attenuation of steroid-catabolizing enzymes in addition to a limited upregulation of P450scc and 3ß-HSD [30].
In the present study, we examined the process of steroidogenic differentiation in granulosa cells obtained from estradiol-primed, prepubertal rat ovaries. Taken from tertiary follicles, these granulosa cells cannot produce steroids because of a total lack of P450scc and StAR proteins. Yet, the latter two genes are readily expressed on hormone treatment of the granulosa cells grown in defined culture medium. This report describes new insights related to the regulation of StAR and 3ß-HSD expression in this cell model. We also show that, among the three genes examined, P450scc is the only one that requires IGF-1 to potentiate the trophic effect of FSH on its expression.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ovine FSH (NIDDK-oFSH-20; FSH) was kindly provided by the National Institutes of Health (NIH) NIAMD (Bethesda, MD). The IGF-1 (I-3769), indomethacin (I-7378), protease-inhibitor cocktail (P-8340), sodium orthovanadate (S-6508), acetyl-coenzyme A (A-2056), and Protein-A Sepharose (P-9424) were obtained from Sigma (St. Louis, MO). The AG18 was kindly provided by Dr. Alexander Levitzki (The Hebrew University of Jerusalem) and synthesized as previously described [31]. Dulbecco modified Eagle medium (DMEM) and Ham F-12 were obtained from Gibco BRL, Life Technologies (Paisley, Scotland). Metabolic labeling starvation medium without cysteine and methionine (Ham F-12, 06-1095-08-1A) was supplied by Biological Industries (Kibbutz Beit-Haemek, Israel). Radiolabeled reagents, including [
-32P]dCTP, [-32P]dATP, [35S]methionine, and [14C]chloramphenicol were obtained from Amersham International (Little Chalfont, U.K.).
Immunoreagents
Polyclonal rabbit antiserum to StAR (provided by Drs. D.B. Hales and K.H. Hales, University of Illinois at Chicago) was raised against a mouse recombinant StAR as described previously [32, 33]. Polyclonal rabbit antiserum to rat P450scc was previously characterized [34], and rabbit antiserum was raised against human placental 3ß-HSD (provided by Dr. I. Mason, The Royal Infirmary of Edinburgh, U.K.). Peroxidase-conjugated AffiniPure goat anti-rabbit serum was obtained from Jackson ImmunoResearch, Inc. (West Grove, PA).
Animals
Intact, immature, female Sprague-Dawley rats (21 days old) were obtained from Harlan (Jerusalem, Israel) and maintained under a 16L:8D photoperiod with food and water ad libitum. Animals were treated in accordance with the NIH Guide for the Care and Use of Laboratory Animals. All protocols had the approval of the Institutional Committee for Animal Care and Use, The Alexander Silverman Institute of Life Sciences, The Hebrew University of Jerusalem.
Granulosa Cell Culture
Naive granulosa cells were isolated from estradiol-primed rats by a slight modification of a previously described procedure [29, 35]. The basic serum-free (SF) medium used for isolation of granulosa cells consisted of a 1:1 (v:v) mixture of DMEM:F-12 [35]. After incubation of the ovaries in hypertonic sucrose/EGTA medium, indomethacin (10 µM) was present in all steps of the cell preparation. This inhibitor of prostaglandin synthase prevents increase of intracellular cAMP and elevation of A-kinase activity resulting from stress-induced synthesis of prostaglandin E2 during needle-pricking of the ovary. Granulosa cells were plated onto serum-coated wells (equivalents of two ovaries per six wells) in 24-multiwell plates (16 mm; Nunc, Copenhagen, Denmark) containing 0.5 ml of SF medium without indomethacin [35]. The IGF-1 (100 ng/ml) was added immediately after inoculation. Treatments with FSH (100 ng/ml) and/or AG18 (80 µM) were initiated 3–4 h after seeding. A fresh dose of AG18 had to be added after 24 h to allow it to exert its maximal effect. This drug is degraded during incubation at 37°C as observed by a progressive loss of its orange color with or without cells. Cultures were incubated at 37°C in a 95% air and 5% CO2 humidified incubator for the indicated times.
Reverse Transcription-Polymerase Chain Reaction Analysis
Total RNA from a single or duplicate culture wells was extracted in 0.15 ml of RNAzol B (Tel-Test, Inc., Friendwood, TX) according to the manufacturer's instructions. Semiquantitative radioactive reverse transcription-polymerase chain reaction (RT-PCR) analysis of total RNA extracts was performed as previously described [32, 36]. Primer sequences for RT-PCR determination of StAR, P450scc, rat type I 3ß-HSD, and ribosomal protein L19 mRNAs have been previously described [36]. The size of the expected PCR products of the rat cDNAs is 246 base pairs (bp) for StAR, 536 bp for P450scc, 405 bp for 3ß-HSD, and 194 bp for L19. The latter cDNA product served as an internal control. The reverse primer for StAR was added during the RT reaction. The PCR reaction was carried for 20 cycles. The dried gels were quantified using a Fuji Bio-Imaging Analyzer (BAS-1000; Fuji Photo Film, Tokyo, Japan). The radioactivity in each PCR band was normalized to the radioactivity of the L19 band. Gels were also exposed to RX medical x-ray film (Fuji Photo Film) for 2–16 h at -80°C and developed by a Curix 60 film processor (Agfa, Munchen, Germany).
Western Blot Analysis
Cultured granulosa cells were extracted by lysis buffer (RIPA buffer containing protease-inhibitor cocktail and 10 µM Na-vanadate) and analyzed by SDS-PAGE (5–10 µg protein/lane) and electroblotting procedures as previously described [32]. Specific signals were detected by chemiluminescence using the LumiGlo substrate (New England BioLabs, Beverly, MA). Enhanced chemiluminescence (ECL) signals recorded on x-ray film were scanned and analyzed by a relative semiquantitative approach using NIH Image Program (available on the Internet at http://rsb.info.nih.gov/nih-image/).
Acid-Wash Removal of Bound Hormones
Granulosa cells were primed for 40 h with FSH and IGF-1 or with FSH alone. Thereafter, free and receptor-bound hormones were removed using a mild acid wash as previously described by Ascoli [37]. Briefly, cells were incubated with cold acid-wash medium (50 mM glycine, 100 mM NaCl; pH 3.0) for 2 min at 4°C, followed by two washes with fresh SF medium. Cells were further incubated in fresh SF medium with or without hormones, and at the indicated time points, RNA extracts were prepared for RT-PCR.
CAT Assay
For each promoter construct, estradiol-primed granulosa cells were obtained from two to three ovaries, electroporated in the presence of 40 µg of plasmid DNA [35], and seeded onto four wells of a 24-well plate (an equivalent of 
0.7 ovary/well). After a 3-h recovery period, hormonal treatments were initiated for a 6-h incubation. Thereafter, cells were lysed, their protein concentration determined by a modification of the Bradford method [38], and their CAT activity assayed [35]. Quantitation of the CAT assay was performed using the Fuji Bio-Imaging Analyzer. It should be noted that the P450scc- and P450 aromatase (P450arom)-promoter activities were measured 6 h after onset of hormonal treatments, because the dynamic range of this assay did not extend beyond a 12-h incubation with FSH (not shown).
Electron Microscopy
Rat ovaries were trimmed free of fat, and small fragments were fixed and prepared for immunogold electron microscopy as previously described [32]. Thin sections were incubated with 1:20 (v:v) dilution of anti-3ß-HSD serum, followed by incubation with 1:10 (v:v) dilution of 9 nm of gold-labeled goat anti-rabbit immunoglobulin G (Jackson ImmunoResearch). Finally, the sections were counterstained with 1% (v/v) aqueous uranyl acetate and lead citrate. Sections were observed and photographed using a Technai 12 electron microscope (FEI, Eindhoven, The Netherlands) equipped with CCD camera. Cytochemical controls included omission of the first antibody (data not shown) or incubation of the ovarian sections with preimmune serum (not shown).
Metabolic Labeling and Immunoprecipitation
Three hours after seeding, the indicated hormones were added for a 17-h priming period, after which the cells were washed and incubated in 0.2 ml of a cysteine-methionine-free medium (starvation medium) supplemented with the corresponding hormones. One hour later, pulse labeling was conducted by addition of 22 µCi of [35S]methionine for a 3-h incubation. Then, cells were thoroughly washed with a complete culture medium and harvested in RIPA buffer (250 µl/well) containing protease inhibitor, vanadate, and BSA (0.5 mg/ml). Cellular debris was removed by a short centrifugation (1 min, 13 000 x g), and extracts were stored at -70°C until immunoprecipitation was performed. For pulse-chase experiments, following removal of the labeling medium and cell washing, onset of chase was initiated by further incubation of the cells in 0.5 ml of complete culture medium enriched with unlabeled methionine and cysteine (0.6 and 0.8 mM, respectively). At the indicated time points, the cell content in each well was extracted in 0.25 ml of RIPA buffer. Immunoprecipitation was initiated by adding 3ß-HSD and StAR antisera (1:1000 v:v) for a 2-h incubation on ice. Then, 25 µl of Protein-A Sepharose beads (20% v/v) were added to each sample, followed by an hour of incubation in a revolving carousel at 4°C. Samples were centrifuged at low speed (200 x g) for 2 min, and the pelleted beads were washed twice in PBS-Tween 20 (0.1% v/v). The immune complexes were eluted with 15 µl of SDS-PAGE sample buffer, resolved on 10% SDS-PAGE, and transferred onto nitrocellulose membranes as described for Western blot analysis. 35S-labeled proteins were quantified by the Fuji Bio-Imaging Analyzer and documented by exposure of the filters to Kodak BioMax MS-1 film (5–16 h at -70°C) using BioMax TranScreen low-energy intensifying screen (Eastman Kodak, Rochester, NY).
Statistical Analysis
Data are presented as mean ± SEM of at least three independent experiments performed by use of identical protocols. Comparisons between different treatment groups were analyzed using unpaired two-tailed Student t-test. A level of P < 0.05 was considered to be statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To study the effect of IGF-1 and FSH on the two genes controlling de novo steroid synthesis, we performed a preliminary study comparing the responses of P450scc and StAR genes in undifferentiated granulosa cell cultures treated for 44 h with either FSH, IGF-1, or both hormones added together. As shown in Figure 1A, expression of P450scc mRNA was totally dependent on the dual presence of both FSH and IGF-1. Alone, IGF-1 had no effect on P450scc message, whereas addition of FSH generated a barely noticeable response. A different pattern was observed with respect to StAR mRNA. Alone, FSH was sufficient for induction of the StAR transcript, whereas IGF-1 only modestly improved the effect of the gonadotropin. These responses were fully corroborated at the protein level (i.e., IGF-1 strongly amplified FSH-dependent induction of P450scc), whereas this growth factor was less required for induction of StAR protein (Fig. 1B). Also, IGF-1 alone did not generate any expression of StAR or P450scc proteins.
|
The limited improvement of StAR expression observed when FSH was added together with IGF-1 apparently did not represent a direct effect by the latter growth factor on gene expression. This conclusion could be drawn when the levels of the gene products were examined at shorter time intervals after hormone addition. Figure 2A1 shows that StAR mRNA rose in a typical acute manner shortly after addition of FSH alone, and that IGF-1 did not improve this response. After a peak rise of StAR mRNA, it declined to half the maximal peak value and, thereafter, maintained a steady-state level for the rest of the incubation period. At incubation intervals longer than 30 h, the cell content of StAR mRNA was twice as high if IGF-1 was maintained in the medium. The reason for this observation is not clear. Possibly, the long-term effect of IGF-1 on StAR levels may have to do with a general pleiotrophic improvement of cell functions. In contrast to StAR expression, the content of P450scc mRNA was always dependent on a concerted addition of IGF-1 and FSH (Fig. 2B1). Furthermore, unlike the acute rise of StAR transcript, P450scc mRNA was not detected before 17 h of induction. The P450scc mRNA could not be detected during the initial lag period even when 35 PCR cycles were employed to increase the sensitivity of the RT-PCR assay (not shown).
|
At the protein level, the accumulation of P450scc protein correlated well with the chronic pattern of its mRNA levels. Figure 2B2 shows an IGF-1-dependent expression of P450scc protein accumulating linearly with time after a lag period of 17–20 h. In contrast, the acute accumulation of StAR protein was independent of IGF-1 (Fig. 2A2). Alone, FSH set in motion StAR accumulation observed as early as 2 h after hormone addition. At longer incubation times (5–30 h), StAR protein kept increasing, whereas its mRNA did not change much. Finally, in concordance with the cellular content of StAR mRNA, a reproducible fall of StAR protein occurred if the cells were kept without IGF-1 for 2 days in culture.
Effects of Tyrphostin AG18 on Gene Expression
The action of IGF-1 on cells is mediated by a well-characterized tyrosine kinase receptor [39]. In support of this notion, we have previously shown that a reversible inhibitor of protein tyrosine kinases, tyrphostin AG18 [31, 40], can exert a profound inhibitory effect on FSH-induced P450scc expression [35, 36, 41]. The present study questioned if AG18 can attenuate StAR expression. Figure 3A corroborates our previous observations showing that AG18 abolished production of P450scc mRNA and protein in long-term cultures treated with FSH and IGF-1 (compare lanes 2 and 4). In contrast, levels of the StAR gene products remained unaffected by the presence of the inhibitor (lanes 2 and 4). To further test if AG18 affects the acute response of StAR gene, a time-dependent expression of StAR was examined in the presence of AG18 and FSH. Figure 3B shows that AG18 had no effect on the accumulation rates of StAR mRNA or protein. On the contrary, and for unclear reasons, cells incubated with AG18 alone responded with a slight increase of StAR mRNA and protein content (Fig. 3A, lane 3). These findings further supported the notion that IGF-1 is not involved in the mechanism of StAR induction.
|
Searching for the Mechanism of IGF-1/FSH Synergy
We hypothesized that IGF-1 potentiates FSH-induced P450scc expression by either affecting the P450scc rate of transcription or by conferring message stabilization. First, the potential effect of IGF-1 on transcription was examined by transiently transfecting granulosa cells with a promoter construct of P450scc (-379/+6) ligated upstream of the CAT reporter. Figure 4A shows that a 6-h incubation with FSH alone was sufficient to generate a 10-fold increase of CAT activity. In other words, unlike the synergistic effect of IGF-1 on P450scc mRNA and protein, the growth factor did not affect any of the promoter activities (Fig. 4A), even when tested in the presence of submaximal concentrations of FSH (Fig. 4B). Similar results were obtained when we tested the promoter activity of cytochrome P450 aromatase, which was the first gene shown to be dramatically affected by FSH and IGF-1 synergy [21]. Figure 4C shows that FSH induced more than a 40-fold activation of the aromatase promoter (-534/+6), but IGF-1 was ineffective whether added alone or with FSH. The reason for IGF-1 ineffectiveness in these promoter assays is not clear. Yet, one of the possibilities is that endogenous IGF-1 is sufficient to obtain maximal values of promoter activity in response to FSH. In support of this notion, the present study shows that AG18 inhibited 70–80% of the P450scc promoter activity measured in the presence of FSH alone (Fig. 4A, FSH+AG18). Similar inhibitory effects of the tyrphostin were previously demonstrated at the level of the aromatase promoter activity [35].
|
To test if the marked synergy of IGF-1 and FSH on P450scc mRNA results from stabilization of the message, we examined the effect of IGF-1 on the turnover rate of the P450scc transcript. To this end, P450scc mRNA was allowed to accumulate in the presence of FSH and IGF-1 (Fig. 5A) or FSH alone (Fig. 5B). At the end of this priming period, hormone action was terminated by a mild acid wash of the cells [37, 42], and the rate of mRNA decay was determined in fresh medium with or without IGF-1. As expected, priming with both hormones resulted in higher P450scc mRNA accumulation (compare lanes 1 and 17). Figure 5, A1 and B1, shows that 2 h after removal of FSH, the P450scc mRNA levels declined with a half-life of 2.5 h. Addition of IGF-1 did not alter the turnover rate of the P450scc transcript whether IGF-1 was or was not present during the priming period (Fig. 5, A1 vs. B1, respectively). For comparison, this experiment also monitored the turnover rate of StAR mRNA (Fig. 5, A2 and B2). The IGF-1 had no impact on the degradation rate of this message as well (t
= 2–3 h). Control cultures receiving FSH replenishment after removal of the priming hormones retained P450scc and StAR transcripts at high level (lanes 16 and 25). This result negated the possibility that degradation of the mRNAs reflected a general perturbation of the cells caused by the acid-wash treatment.
|
As shown in Figure 5A1, a transient rise of P450scc transcript was observed following the wash procedure, just before mRNA degradation was set in motion. The IGF-1 clearly augmented this response (Fig. 5A1, IGF-1). We did not observe this pattern in P450scc-expressing cells primed without IGF-1 (Fig. 5B1) or at the level of StAR expression (Fig. 5, A2 and B2). The mechanism underlying this selective "flare" of P450scc mRNA occurring next to removal of the priming hormones is unclear at present. However, a similar phenomenon was previously observed when the turnover rate of the LH-receptor mRNA was examined in rat granulosa cells [42].
Expression of 3ß-HSD
The third gene whose expression is required for active steroid hormone synthesis is 3ß-HSD. Very little is known about the regulation of 3ß-HSD in rodent ovaries. Figure 6 depicts RT-PCR and Western blot analyses of cells examined immediately after isolation or 17 and 44 h after hormonal treatment. The results suggest that 3ß-HSD represents a third mode of gene expression: First, 3ß-HSD mRNA and protein were observed in freshly isolated granulosa cells (Fig. 6, lanes 1 and 10). These observations were rather unexpected considering that the granulosa cells of prepubertal rat ovaries are not functionally differentiated and lack P450scc protein [43]. Further support for the presence of 3ß-HSD in granulosa cells of prepubertal follicles was achieved by immunoelectron microscopy of a tertiary follicle. Figure 7 shows a substantial immunogold decoration of cytoplasmic 3ß-HSD (Fig. 7, A and b) expressed in a mural granulosa cell. As expected, higher labeling of 3ß-HSD was observed in steroidogenically active theca cells [43], which were easily identified by their typical cholesterol-ester lipid droplets in the cytoplasm (Fig. 7, A and c). In agreement with the ultrastructural findings, a Western blot analysis of freshly isolated cells shows that compared to the granulosa cells, the theca-interstitial cells contain higher level (2.8-fold) of the 3ß-HSD protein (Fig. 7D).
|
|
Figures 6A shows that the basal level of 3ß-HSD mRNA remained almost unchanged during 44 h in culture without hormonal treatment (lanes 6 and 15). On hormonal induction, FSH induced a 2.5-fold (lane 4) to 4.1-fold (lane 8) increase of 3ß-HSD mRNA. Unlike the inability of IGF-1 to promote expression of P450scc and StAR when added on its own, IGF-1 alone generated a 4.1-fold (lane 3) and 3.6-fold (lane 7) increase of 3ß-HSD mRNA. Addition of FSH and IGF-1 together did not generate a statistically different response when compared to IGF-1 alone measured at 17 h (lane 5). A weak synergistic effect of IGF-1 and FSH was observed after a longer incubation time of 44 h (lane 9). At the protein level, Figure 6B reveals an aspect of 3ß-HSD expression that to our knowledge has never before been addressed. After 44 h with IGF-1 and FSH, the level of 3ß-HSD protein was elevated no more than twice relative to the nontreated cells (compare lanes 18 and 15), whereas the two hormones generated a 15-fold increase of the message content (compare lanes 9 and 6).
We next attempted to understand the mechanism accounting for the unusual disproportion between the marked induction of 3ß-HSD mRNA and the apparently weak changes at the level of the protein. To this end, we hypothesized that 3ß-HSD protein is either translated at a slow rate or quickly degraded. To address these possibilities, we measured the rate of de novo [35S]3ß-HSD synthesis while priming the cells with FSH, IGF-1, or the two hormones together. Immunoprecipitation of [35S]3ß-HSD did not reveal any new insights (Fig. 8A) when compared to the Western blot analyses previously performed (Fig. 6B). In other words, the hormone-induced rise of de novo [35S]3ß-HSD synthesis fell behind the marked rise of the corresponding mRNA levels (Fig. 6A). In addition, a pulse-chase experiment showed that the turnover rate of [35S]3ß-HSD (t = 20 h) is markedly slower than that of StAR (t = 3 h), which served as a short-lived reference protein (Fig. 8B). Altogether, these findings suggest that the limited effect of hormones on the cellular content of 3ß-HSD protein resulted from a combination of several parameters, including high basal expression of this gene in the nondifferentiated cells, a marked stability of the 3ß-HSD protein, and possibly, a relatively slower rate of message translation.
|
Finally, considering that IGF-1 alone was enough to confer induction of 3ß-HSD, we hypothesized that tyrphostin AG18 should inhibit expression of this gene. Indeed, Figure 9 shows that AG18 completely blocked IGF-1-dependent expression of 3ß-HSD mRNA and protein (lanes 7, 9 and 16, 18, respectively). Furthermore, AG18 caused a 70% reduction of the basal mRNA level maintained without hormones (compare lanes 6 and 1). The latter findings suggest that the inhibitor also blocked the typical constitutive expression of 3ß-HSD. It is tempting to speculate that the basal expression of 3ß-HSD in the rat granulosa cells is perpetuated by endogenously secreted IGF-1. In this respect, it should be noted that the constitutive level of 3ß-HSD protein was not affected much during the 17-h incubation with AG18, presumably because of the previously mentioned slow turnover of the 3ß-HSD protein.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This report describes an integrated study of three gene products controlling progesterone synthesis in all steroidogenic cells, namely, P450scc, StAR, and 3ß-HSD. We chose to examine these genes in granulosa cells from prepubertal rat ovaries because they can be obtained in an undifferentiated state, grow in defined medium, and readily acquire steroidogenic functions on treatment with their trophic hormones [35, 44–46].
P450scc In the rat cells, P450scc mRNA and protein are the only gene products subject to synergy between IGF-1 and FSH. Whereas IGF-1 was totally ineffective when added alone, its presence dramatically potentiated FSH action (>20-fold) on P450scc expression. The cardinal role of IGF-1 was further demonstrated by use of tyrphostin AG18, known as a potent inhibitor of epidermal growth factor-receptor protein tyrosine kinase [31, 40]. Addition of AG18 abolished the marked induction of P450scc by the concerted action of FSH and IGF-1. Therefore, AG18 was further used as a valuable biochemical probe to detect potential involvement of IGF-1 in expression of the other genes under study.
The marked synergy between FSH and IGF-1 on P450scc expression can account for the long-known augmentation of progesterone synthesis that occurs when IGF-1 is added to FSH-treated cultured granulosa cells from rat and pig ovary [23, 24]. Although our findings confirm and extend those of previous studies regarding FSH/IGF-1 action, significant differences do exist between the present cell model and others. For example, a recent study employing somewhat different culture conditions to maintain the same rat granulosa cells showed that IGF-1 did not generate a marked effect on the P450scc mRNA level when added with FSH [30]. We showed that IGF-1 alone did not generate any P450scc response in the rat granulosa cells, whereas IGF-1 was enough to promote P450scc expression in the counterpart pig cells [22, 24]. A similar response to solely added IGF-1 was also observed in rat theca-interstitial ovarian cells [47]. Collectively, these observations suggest that qualitative as well as quantitative aspects of FSH/IGF-1 relationship can be species dependent, be cell specific, or result from different experimental conditions.
StAR Unlike the typical chronic pattern of P450scc expression, this study shows that the acute response of StAR to FSH is measurable within minutes. As such, the pattern of StAR expression in the granulosa cells is reminiscent of similar responses to trophic hormones in other steroidogenic cell types, including mouse MA-10 testicular Leydig cell lines [48], mouse Y-1 adrenocortical tumor cells [49], and primary bovine adrenocortical cells treated with ACTH [50]. These nonovarian cell types maintain substantial levels of P450scc expression at all times so that an acute production of StAR can serve its physiological purpose and boost steroidogenesis in a matter of minutes. In undifferentiated granulosa cells shown in the present study to express P450scc much later than StAR, the physiological relevance of this response might seem to be questionable. However, no such premature expression of StAR occurs in the granulosa cell compartment of developing follicles in vivo [32]. The reasons for the intriguing discrepancy between the in vivo and in vitro temporal expression of StAR are as yet unclear, but they may involve alterations in factors controlling upregulation of StAR at the level of transcription [29].
The acute induction of StAR was solely dependent on FSH. Although IGF-1 somewhat supported higher levels of StAR protein at prolonged hormone treatments, this effect was probably caused by the pleiotrophic influence of IGF-1 in its capacity as a growth factor [18, 19]. This view is supported by the fact that tyrphostin AG18 did not affect the gonadotropin-dependent acute response of StAR, suggesting that no receptor tyrosine kinase pathway is involved in upregulation of StAR gene products. Nevertheless, in other cell models, such as in porcine granulosa cells, IGF-1 has had a marked synergistic effect on StAR responses to FSH [25] or LH [51].
3ß-HSD The third pattern of ovarian gene expression is represented by the constitutive presence of 3ß-HSD. Unlike P450scc and StAR, which are not expressed in the absence of FSH, the freshly isolated granulosa cells express a relatively high level of 3ß-HSD mRNA and protein as shown by RT-PCR, Western blot analyses, and immunoelectron microscopy of follicular thin sections. Similar observations were reported by deMoura et al. [30]. Our studies also show that FSH did not greatly affect the level of 3ß-HSD protein. Moreover, out of the three genes examined, 3ß-HSD is the only one that is inducible by IGF-1 alone. Accordingly, AG18 readily inhibited IGF-1-induced 3ß-HSD. The tyrphostin also attenuated the hormone-independent expression of 3ß-HSD, suggesting that the basal levels of 3ß-HSD mRNA and protein are probably maintained by the well-characterized autocrine loop of IGF-1 secretion and reception [14].
By use of metabolic labeling and Western blot approaches, our studies suggest that the hormone-independent basal level of 3ß-HSD protein reflects a steady-state balance between a constitutive translation from existing transcript and a slow rate of 3ß-HSD protein degradation. Furthermore, a limited (2-fold) increase in the rate of de novo [35S]3ß-HSD synthesis was measured in hormone-responsive cells that generated a 5-fold increase of the basal transcript content. These observations may suggest that the rate of 3ß-HSD translation is substantially slower than expected, and that the mRNA levels evoked under hormonal treatments do not serve as the rate-limiting step in 3ß-HSD production.
The selective impairment of P450scc and 3ß-HSD expression by AG18 confirmed that this tyrphostin targeted IGF-1-induced genes only. These findings add to the growing list of genes that are subjected to dual effects of cAMP-dependent and growth factor-related pathways. For example, cytochrome P450arom depends on FSH/cAMP action [52, 53] that can be potentiated in the presence of IGF-1 [21, 54, 55]. In many other endocrine models, emerging evidence advocates the presence of distinct signaling networks that cross-talk with one another [56, 57], like a direct coupling of cAMP and Ras signaling pathways [58–61], or a protein kinase B (Akt) that was recently found to be phosphorylated and activated by FSH in rat granulosa cells [57, 62]. Akt is also known as a downstream target of insulin/IGF-1 pathways [63, 64], so that potential convergence of cAMP and IGF-1 pathways could have occurred via this protein in the present cell model.
Elusive Mechanism of IGF-1 Action
Aiming to resolve the mechanism of IGF-1 action, the present study explored a few possibilities that to our knowledge have never before been tested in the rat granulosa model. First, we examined whether IGF-1 is involved in posttranscriptional stabilization of P450scc mRNA. The degradation rate of P450scc message was not affected by IGF-1, whether added during synthesis of the mRNA or later, after cessation of transcription following the removal of the trophic hormones. Therefore, the tremendous increase of P450scc mRNA generated in the presence of IGF-1 and FSH could not result from improved message stability.
Next, we attempted to reveal a direct effect of IGF-1 on transcriptional activation of the P450scc promoter. In this regard, Urban et al. [65] have recently identified a GC-rich domain positioned at -130/-100 of the porcine P450scc promoter that imparts IGF-1 regulation through a distinct trans-acting protein complex. To study the rat model, we used the proximal -379/+6 fragment of the rat P450scc promoter. This region includes steroidogenic factor-1-binding elements (-51/-43 and -79/-71) previously shown to be critical for activation of this gene in the ovary and adrenal cells [35, 66] and was long enough to include potentially homologous GC-rich region(s) that might be involved in IGF-1 regulation of P450scc, as suggested by Urban et al. [65]. Surprisingly, FSH alone was enough to generate a 10-fold increase of the promoter activity. This response was observed within a 6-h incubation with hormone, too short a time to allow for accumulation of the authentic mRNA and protein of P450scc. Addition of IGF-1 did not increase the promoter activity at any dose of FSH. Similarly, the growth factor was ineffective when added to granulosa cells transfected with the proximal region of the P450arom promoter (-534/+6). These results agreed, therefore, with other models in which added IGF-1 readily potentiates FSH-induced genes but fails to affect their cognate promoter activity [26, 67]. Collectively, it is tempting to speculate that compared with production of the authentic P450scc transcript, the promoter activity assay is more sensitive for probing the presence of endogenous IGF-1 secretion and does not require exogenously added IGF-1 to affect transcription of CAT. This possibility is strongly supported by a marked inhibition of the FSH-induced promoter activity observed in the presence of AG18, suggesting that the tyrphostin may have intercepted possible signaling pathways of endogenous IGF-1. Further support for this notion was provided by classical series of experiments showing that IGF-binding proteins and neutralizing IGF-1 antibodies can severely diminish FSH-induced synthesis of progesterone in rat and porcine granulosa cell cultures [68, 69].
The apparent lack of IGF-1 effect at the level of promoter activity could also suggest that an IGF-1-dependent regulatory element exists outside of the region tested. This possibility is currently under investigation. Another mechanism worth considering is that IGF-1 may have affected P450scc expression indirectly. Such a mechanism may involve IGF-1 effect on other gene products that could enhance P450scc expression. For example, it has been proposed that IGF-1 can stimulate a ligand-independent activity of the estrogen receptor [70, 71]. Because estrogen is required for proper follicular performance, it is tempting to speculate that addition of IGF-1 to granulosa cell cultures may result in a marked synergistic effect exerted by a rapid crosstalk between IGF-1 and estrogen receptor-ß pathways.
Collectively, the present study has explored a few novel aspects of IGF-1 involvement in the expression of pivotal genes required for ovarian steroid hormone synthesis. However, the molecular mechanism underlying the unique ovarian synergy of IGF-1 with FSH remains unclear and is, as yet, a challenging goal for future studies.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by the United States-Israel Binational Sciences Foundation Grant 1999315. 
2 Correspondence. FAX: 972 2 658 6448; orly{at}vms.huji.ac.il
Accepted: April 9, 2002.
Received: December 3, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
regulation of messenger ribonucleic acid encoding steroidogenic acute regulatory protein in human corpora lutea. Endocrine 1998 8::153-160[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  M. Orisaka, J.-Y. Jiang, S. Orisaka, F. Kotsuji, and B. K. Tsang Growth Differentiation Factor 9 Promotes Rat Preantral Follicle Growth by Up-Regulating Follicular Androgen Biosynthesis Endocrinology, June 1, 2009; 150(6): 2740 - 2748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Yivgi-Ohana, N. Sher, N. Melamed-Book, S. Eimerl, M. Koler, P. R. Manna, D. M. Stocco, and J. Orly Transcription of Steroidogenic Acute Regulatory Protein in the Rodent Ovary and Placenta: Alternative Modes of Cyclic Adenosine 3', 5'-Monophosphate Dependent and Independent Regulation Endocrinology, February 1, 2009; 150(2): 977 - 989. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Alam, J. Weck, E. Maizels, Y. Park, E. J. Lee, M. Ashcroft, and M. Hunzicker-Dunn Role of the Phosphatidylinositol-3-Kinase and Extracellular Regulated Kinase Pathways in the Induction of Hypoxia-Inducible Factor (HIF)-1 Activity and the HIF-1 Target Vascular Endothelial Growth Factor in Ovarian Granulosa Cells in Response to Follicle-Stimulating Hormone Endocrinology, February 1, 2009; 150(2): 915 - 928. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Granot, O. Kobiler, N. Melamed-Book, S. Eimerl, A. Bahat, B. Lu, S. Braun, M. R. Maurizi, C. K. Suzuki, A. B. Oppenheim, et al. Turnover of Mitochondrial Steroidogenic Acute Regulatory (StAR) Protein by Lon Protease: The Unexpected Effect of Proteasome Inhibitors Mol. Endocrinol., September 1, 2007; 21(9): 2164 - 2177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Seto-Young, D. Avtanski, M. Strizhevsky, G. Parikh, P. Patel, J. Kaplun, K. Holcomb, Z. Rosenwaks, and L. Poretsky Interactions among Peroxisome Proliferator Activated Receptor-{gamma}, Insulin Signaling Pathways, and Steroidogenic Acute Regulatory Protein in Human Ovarian Cells J. Clin. Endocrinol. Metab., June 1, 2007; 92(6): 2232 - 2239. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sher, N. Yivgi-Ohana, and J. Orly Transcriptional Regulation of the Cholesterol Side Chain Cleavage Cytochrome P450 Gene (CYP11A1) Revisited: Binding of GATA, Cyclic Adenosine 3',5'-Monophosphate Response Element-Binding Protein and Activating Protein (AP)-1 Proteins to a Distal Novel Cluster of cis-Regulatory Elements Potentiates AP-2 and Steroidogenic Factor-1-Dependent Gene Expression in the Rodent Placenta and Ovary Mol. Endocrinol., April 1, 2007; 21(4): 948 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-J. Chen, P.-W. Hsiao, M.-T. Lee, J I. Mason, F.-C. Ke, and J.-J. Hwang Interplay of PI3K and cAMP/PKA signaling, and rapamycin-hypersensitivity in TGF{beta}1 enhancement of FSH-stimulated steroidogenesis in rat ovarian granulosa cells J. Endocrinol., February 1, 2007; 192(2): 405 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Bastida, A. Cremades, M. T. Castells, A. J. Lopez-Contreras, C. Lopez-Garcia, F. Tejada, and R. Penafiel Influence of Ovarian Ornithine Decarboxylase in Folliculogenesis and Luteinization Endocrinology, February 1, 2005; 146(2): 666 - 674. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Pierre, C Pisselet, J Dupont, B Mandon-Pepin, D Monniaux, P Monget, and S Fabre Molecular basis of bone morphogenetic protein-4 inhibitory action on progesterone secretion by ovine granulosa cells J. Mol. Endocrinol., December 1, 2004; 33(3): 805 - 817. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Payne and D. B. Hales Overview of Steroidogenic Enzymes in the Pathway from Cholesterol to Active Steroid Hormones Endocr. Rev., December 1, 2004; 25(6): 947 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. LaVoie, D. Singh, and Y. Y. Hui Concerted Regulation of the Porcine Steroidogenic Acute Regulatory Protein Gene Promoter Activity by Follicle-Stimulating Hormone and Insulin-Like Growth Factor I in Granulosa Cells Involves GATA-4 and CCAAT/Enhancer Binding Protein {beta} Endocrinology, July 1, 2004; 145(7): 3122 - 3134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Demeestere, C. Gervy, J. Centner, F. Devreker, Y. Englert, and A. Delbaere Effect of Insulin-Like Growth Factor-I During Preantral Follicular Culture on Steroidogenesis, In Vitro Oocyte Maturation, and Embryo Development in Mice Biol Reprod, June 1, 2004; 70(6): 1664 - 1669. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Alam, E. T. Maizels, Y. Park, S. Ghaey, Z. J. Feiger, N. S. Chandel, and M. Hunzicker-Dunn Follicle-stimulating Hormone Activation of Hypoxia-inducible Factor-1 by the Phosphatidylinositol 3-Kinase/AKT/Ras Homolog Enriched in Brain (Rheb)/Mammalian Target of Rapamycin (mTOR) Pathway Is Necessary for Induction of Select Protein Markers of Follicular Differentiation J. Biol. Chem., May 7, 2004; 279(19): 19431 - 19440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kajitani, T. Mizutani, K. Yamada, T. Yazawa, T. Sekiguchi, M. Yoshino, H. Kawata, and K. Miyamoto Cloning and Characterization of Granulosa Cell High-Mobility Group (HMG)-Box Protein-1, a Novel HMG-Box Transcriptional Regulator Strongly Expressed in Rat Ovarian Granulosa Cells Endocrinology, May 1, 2004; 145(5): 2307 - 2318. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.-M. Huang, K.-Y. Hsiao, P.-C. Chuang, M.-H. Wu, H.-A. Pan, and S.-J. Tsai Upregulation of Steroidogenic Enzymes and Ovarian 17{beta}-Estradiol in Human Granulosa-Lutein Cells by Cordyceps sinensis Mycelium Biol Reprod, May 1, 2004; 70(5): 1358 - 1364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Thompson, S. M. Ross, and K. W. Gaido Di(n-Butyl) Phthalate Impairs Cholesterol Transport and Steroidogenesis in the Fetal Rat Testis through a Rapid and Reversible Mechanism Endocrinology, March 1, 2004; 145(3): 1227 - 1237. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Zeleznik, D. Saxena, and L. Little-Ihrig Protein Kinase B Is Obligatory for Follicle-Stimulating Hormone-Induced Granulosa Cell Differentiation Endocrinology, September 1, 2003; 144(9): 3985 - 3994. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
