| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Cell Biology and Biochemistry, Texas Tech University Health Sciences Center, Lubbock, Texas 79430
b Department of Physiology, Institute of Biomedicine, University of Turku, FIN-20520 Turku, Finland
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.53 nmol/L) in these cells. It was also determined that the arachidonic acid (especially lipoxygenase metabolites) and mitogen-activated protein kinase pathways were also involved in the mEGF-induced steroidogenic response. However, involvement of the latter pathway was further assessed in nonsteroidogenic COS-1 cells transfected with the Elk1 trans-reporting plasmids and resulted in a significant increase in luciferase activity in response to mEGF. Furthermore, deletion and mutational analyses demonstrated a predominant involvement of activator protein-1 in addition to the multiple mEGF responsive elements found within the 5'-flanking region (-151/-1 base pairs) of the mouse StAR gene. These findings provide novel insights into the mEGF-induced regulatory cascades associated with steroid synthesis and StAR protein expression in mouse Leydig cells.
growth factors, Leydig cells, mechanisms of hormone action, signal transduction, steroid hormones
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Epidermal growth factor (EGF), a mitogenic polypeptide first isolated from mouse salivary glands, has numerous actions on different cell types both in vitro and in vivo [7–11]. The EGF family, including transforming growth factor 
and amphiregulin, mediates its action through transmembrane glycoprotein receptors [7, 12, 13] and promotes proliferation, survival, and differentiation of a wide variety of mammalian cells. Biochemical, immunocytochemical, and ligand-binding studies have identified the presence of EGF receptors in human [14], porcine [15], rat [16], and mouse [17, 18] Leydig cells. Treatment with EGF has been shown to directly modulate steroid production and to potentiate the action of LH/hCG in porcine and in cultured mouse Leydig cells, respectively [6, 15, 19]. Conversely, EGF has also been demonstrated to inhibit the stimulatory effects of gonadotropins on testosterone production in isolated rat and mouse Leydig cells [20] and in interstitial cells from hypophysectomized rats [21] because of a decrease in the steroidogenic potency of the cells [20]. Also, other studies have demonstrated that EGF is a negative modulator of Leydig cell LH-receptor (LHR) expression [22, 23].
Evidence is increasing that EGF modulates steroid biosynthesis in a cAMP-independent manner [19, 20], an observation in contrast to LH/hCG-stimulated steroidogenesis. Also, EGF has been demonstrated to activate MAPKs (extracellular signal-regulated kinases, ERK1 and ERK2) in gonadal cells [3]; however, a link between activation of the MAPK pathway and steroid biosynthesis has yet to be demonstrated. Majercik and Puett [24] have reported that EGF rapidly increases the level of intracellular arachidonic acid (AA) in MA-10 mouse Leydig tumor cells. Also, the AA pathway has been shown to be involved in the regulation of steroid biosynthesis and steroidogenic acute regulatory (StAR) protein expression in cultured Leydig cells [24–26]. Hoelscher and Ascoli [27] identified a 21-kDa phosphoprotein named stathmin following EGF/hCG stimulation in MA-10 cells, the phosphorylation of which is associated with multiple second-messenger pathways. In addition, EGF has been implicated in the morphological and functional maturation of the fetal primate adrenal gland [28]. Further underscoring its importance in testicular function, EGF plays a crucial role in regulating spermatogenesis [10, 29], an observation strengthened by EGF-null mice displaying sperm flagellar axonemal disruption and being sterile [9]. Despite the physiological importance of EGF in testicular function and, particularly, in Leydig cell steroidogenesis, its mechanism of action remains poorly understood.
The rate-limiting step in steroid hormone biosynthesis is the transport of cholesterol from the outer to the inner mitochondrial membrane, the site of P450scc, a process that is dependent on de novo synthesis of the StAR protein [30–32]. The role of StAR protein in steroid biosynthesis has been clearly demonstrated by studies of patients with congenital lipoid adrenal hyperplasia and by StAR-knockout mice; both of which are characterized by severe defects in the synthesis of adrenal and gonadal steroids [31, 33–35]. In a recent report [36], we demonstrated that hCG enhanced StAR protein and StAR mRNA levels in a time frame consistent with progesterone (P) production in mLTC-1 mouse Leydig tumor cells [37]. Utilizing these cells as an in vitro model, the present studies were designed to explore, to our knowledge for the first time, mechanisms involved in mEGF-induced regulation of steroid biosynthesis and StAR expression in mouse Leydig cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The 5'-flanking regions of the mouse StAR gene (-966, -426, -254, -151, -110, and -68 base pairs [bp]) were cloned upstream of the luciferase reporter gene into the pGL2 basic vector (Promega, Madison, WI) as described previously [38, 39]. The constructs made were in relation to the translation initiation codon (-1 bp).
Mutational analyses were carried out using the -151/-1-bp (XhoI and HindIII) segment of the StAR promoter following subcloning into the pGL3 basic vector [39]. Plasmids containing mutations in the putative recognition sites (see below) were generated using the Quikchange site-directed mutagenesis kit (Stratagene, La Jolla, CA) employing the -151/-1 StAR-pGL3 as the template. The sense strands of the oligonucleotide sequences used were as follows (mutated bases are shown in bold lowercase and primer positions in parentheses):
Specific mutations were tested by restriction endonuclease digestion using EcoRI (Sp1 mut), EcoRI (SF-1/1 mut), SalI (C/EBP mut), PvuII (SF-1/3 mut), BglII (AP-1 mut), BspEI (GATA mut), XhoI (SREBP mut), and PvuII (SF-1/2 mut). Finally, a XhoI and HindIII fragment containing the mutations was religated into the pGL3 vector. All plasmids were confirmed by sequencing on a PE biosystem 310 Genetic Analyzer (Perkin-Elmer, Foster City, CA) at the Texas Tech University Biotechnology Core Facility.
RNA Extraction and Quantitative Reverse Transcription-Polymerase Chain Reaction
Total RNA from the different treatment groups was extracted using Trizol reagent (Gibco-BRL, Grand Island, NY). The primers used for isolation and amplification of the mouse (mLTC-1) StAR cDNA were the sense primer, 5'-GACCTTGAAAGGCTCAGGAAGAAC-3', and the antisense primer, 5'-TAGCTGAAGATGGACAGACTTGC-3', which spanned bases -51 to -27 and 931 to 908, respectively [40]. The variation in reverse transcription-polymerase chain reaction (RT-PCR) efficiency was assessed with the L19 ribosomal protein gene as an internal control using the sense primer 5'-GAAATCGCCAATGCCAACTC-3' and the antisense primer 5'-TCTTAGACCTGCGAGCCTCA-3' [41].
The RT and PCR analyses were run sequentially in the same assay tube using 2 µg of total RNA. The cDNAs generated were further amplified by PCR under optimized conditions [42, 43] using the primer pairs listed above. The molecular sizes of the StAR and L19 amplicons were determined on a 1.2% (w/v) agarose gel, after which the gels were vacuum-dried and then exposed to Hyperfilm (Amersham Pharmacia Biotech, Buckinghamshire, U.K.) for 1–3 h. The levels of the StAR and L19 mRNA signals were quantitated using a computer-assisted image analyzer (Visage 2000; BioImage, Ann Arbor, MI) and the results expressed relative to the L19 gene expression.
Isolation of Mitochondria and Western Blot Analysis
Isolation of mitochondria and immunoblotting of StAR protein were carried out as described previously [36, 43, 44]. Briefly, mitochondrial protein (20–25 µg/lane) was solubilized in SDS sample buffer and loaded onto a 12% (w/v) SDS-polyacrylamide gel (Mini Protean II System; Bio-Rad, Hercules, CA). Electrophoresis, transfer of proteins to polyvinylidene fluoride membranes, and immunodetection of StAR were performed under optimized conditions [36, 43, 44]. The immunoblots were quantitated using the Visage 2000 image analyzer as described above.
Northern Hybridization Analysis
Total RNA (15–20 µg) from different treatment groups was separated on 1.2% (w/v) formaldehyde denaturing agarose gels and transferred onto Hybond-N+ nylon membranes (Amersham). Antisense cRNA probes, a NotI fragment of the mouse StAR gene (960 bp), a BglII fragment of the extracellular domain of the rat LHR (410 bp), were produced by in vitro transcription (Promega) with T7 RNA polymerase, dNTPs, and [-32P]UTP (800 Ci/mmol; Amersham). Prehybridization, hybridization, and washing of the membranes were carried out at 66°C under stringent conditions [36, 45]. To normalize for the variation in StAR and LHR mRNA levels, the membranes were rehybridized with a cDNA probe for the glyceraldehyde-3-phosphate dehydrogenase.
Receptor-Binding Studies
Mouse EGF (mEGF; Sigma Chemical Co., St. Louis, MO) and hCG (CR-127; NIDDK, NIH, Bethesda, MD) were radio-iodinated with Na[125I]iodide (Amersham) as described previously [45, 46]. The specific activity of the labeled hormones was between 22 and 40 µCi/µg. Briefly, 2.5–3 x 105 cells were incubated with labeled hormones (150 000 cpm) either in the absence (total) or presence (nonspecific) of 10 µg/L of unlabeled mEGF or 50 IU of unlabeled hCG (Pregnyl; NV Organon, Oss, The Netherlands). Following overnight incubation, the reaction was terminated by the addition of 3 ml of ice-cold Dulbecco-PBS containing 0.1% (w/v) BSA. After centrifugation, the supernatant was discarded, and radioactivity in the pellet was determined on washing using a 
-counter (1277 Gammamaster; LKB Wallac Oy, Turku, Finland). The affinity (Kd) of mEGF binding was determined by converting the binding-inhibition data into Scatchard plots [45].
Effect of mEGF on the Pathdetect Elk1 Trans-Reporting Assay
Nonsteroidogenic COS-1 cells were transiently transfected with the pathdetect trans-reporting system (Stratagene), which includes a pathwayspecific fusion trans-activator plasmid (pFA2-Elk1) that expresses a fusion protein as described previously [47]. The fusion trans-activator protein consists of the activation domain of Elk1 (307–427 bp) fused with the yeast GAL4 DNA-binding domain (dbd; 1–147 bp), which is driven by a powerful CMV promoter. The pFR-Luc is a reporter plasmid that controls expression of the luciferase gene. The pFC2-dbd is a negative control of pFA2 plasmid that lacks the activation domain of Elk1. Transfection studies were carried out using FuGENE 6 transfection reagent (Roche Diagnostics Corp., Indianapolis, IN), and the effects of mEGF were assessed by determining MAPK-mediated luciferase activity. Following 36 h of transfection, cells were stimulated for 6 h with increasing doses of mEGF (0.1–100 µg/L), and luciferase activity in the cell lysates was determined using a TD 20/20 luminometer (Turner Designs, Sunnyvale, CA).
Cell Culture, Plasmids, Transfections, and Luciferase Assays
The mouse Leydig tumor cells (mLTC-1) and COS-1 (African green monkey kidney) cells were obtained from American Type Culture Collection (ATCC, Manassas, VA) and maintained in the culture media as described earlier [37, 39]. The media collected from different experiments were assayed for P levels [48]. The accumulation of cAMP in media was determined between 0.5 and 2 h in the absence or presence of 0.5 mmol/L of 3-isobutyl-1-methyl xanthine (Sigma) using the cAMP [125I]RIA kit (NEN; Life Sciences Products, Inc., Boston, MA).
For promoter analyses, transfection studies were carried out in mLTC-1 cells using FuGENE 6 transfection reagent under optimized conditions [36, 42]. Briefly, different wild-type and mutant StAR constructs or the -151/-1-bp StAR promoter in relation to the specified AP-1 family members (c-Fos and Fra-1 cDNAs inserted into XhoI-BamHI and c-Jun and JunD cDNAs inserted into EcoRI-HindIII sites of pcDNA3.1 Myc-His vector; Invitrogen, San Diego, CA) were used. Transfections were carried out in the presence of the pRL-SV40 vector (a plasmid that constitutively expresses renilla luciferase) to normalize the transfection efficiency [39]. Luciferase activity in the cell lysates was determined by the Dual-luciferase reporter assay system (Promega).
Statistical Analysis
The results presented were analyzed by one-way ANOVA followed by Fisher least-significant difference tests using the Statview 5.1 program (Abacus Concepts, Inc., Berkeley, CA) fitted for the Macintosh computer. The results expressed represent the mean ± SEM, and P < 0.05 was considered to be statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The optimal conditions for mEGF effects on P production and StAR mRNA expression in mLTC-1 cells were first determined. The results summarized in Figure 1 demonstrate the time responses (0–24 h) of P production and StAR mRNA expression following mEGF treatment. Cells stimulated with mEGF (10 µg/L) showed a maximal 4.4- ± 0.5-fold increase in the level of StAR mRNA over nonstimulated cells (Fig. 1A). The magnitude of response was significant (P < 0.05) at 0.5 h, increased gradually up to 4 h, started decreasing between 4 and 8 h, and reached levels below those of controls at 18–24 h (Fig. 1B). The accumulation of P in the media was significant (P < 0.05) within 0.5 h, increased gradually (being maximal at 4 h; 12.4- ± 2.7-fold over control), and then decreased following the pattern of StAR mRNA, which indicated a close correlation between mEGF-stimulated P production and StAR mRNA expression (Fig. 1B). Because treatment with mEGF produced maximal P production and StAR gene expression between 4 and 6 h, a 5-h incubation period was chosen for subsequent experiments.
|
The dose-response pattern in response to mEGF was next examined (Fig. 2). The MLTC-1 cells stimulated for 5 h with increasing doses of mEGF (0–100 µg/L) showed a dose-dependent increase in the levels of StAR protein (Mr = 30 000) (Fig. 2A) and StAR mRNA (Fig. 2B). The maximal responses were 3.3- ± 0.2-fold (StAR protein) and approximately 4.8-fold (StAR mRNA) over the respective controls. The increases in StAR protein and StAR mRNA levels were significant (P < 0.05) at 0.1 µg/L, with the half-maximal stimulation occurring at 2–3 µg/L and the maximum increase obtained at 10 µg/L. Production of P with increasing doses of mEGF (maximally 13-fold) was qualitatively similar to the levels of StAR protein and StAR mRNA, demonstrating a direct correlation between StAR expression and P production.
|
Induction of mEGF-Mediated P Synthesis and StAR Gene Expression Require Protein Synthesis
The involvement of mEGF in P production and StAR mRNA expression was investigated for its dependence on protein synthesis. Northern blot analysis revealed that mLTC-1 cells stimulated with mEGF (10 µg/L) for 5 h showed a significant elevation of approximately fivefold in StAR mRNA expression in different transcripts (3.4, 2.7, and 1.6 kilobases [kb]) (Fig. 3A). Addition of the protein synthesis-inhibitor cycloheximide (CHX; 10 mg/L) to mEGF-stimulated cells significantly inhibited all StAR transcripts but had no effect on basal StAR mRNA expression, indicating that the induction of steroidogenic activity by mEGF in these cells is dependent on translation. The effect of mEGF (10 µg/L) on P synthesis in the media was approximately 12.8-fold over basal and was significantly inhibited (P < 0.01) by CHX (Fig. 3B).
|
Binding of [125I]mEGF to mLTC-1 Cells
Ligand-binding activity of mEGF was next analyzed in mLTC-1 cells. Displacement of [125I]mEGF binding increased gradually with increasing amounts of unlabeled mEGF (0–100 µg/L) (Fig. 4). The equilibrium dissociation constant (Kd, 0.53 nmol/L) demonstrated the presence of a high-affinity component, as calculated from Scatchard analysis (Fig. 4, inset), indicating the presence of a homogeneous receptor population.
|
Interaction of mEGF on hCG/cAMP-Stimulated StAR mRNA Expression
Because Leydig cell steroidogenesis is an LH/hCG-mediated process, we next studied the effect of mEGF on StAR mRNA expression in hCG- and (Bt2)cAMP-stimulated cells. The data presented in Figure 5 show that mLTC-1 cells stimulated with maximum effective doses of mEGF (10 µg/L), hCG (50 µg/L), and (Bt2)cAMP (1 mM; not illustrated in Fig. 5) demonstrated significant increases in StAR mRNA expression. Addition of maximally stimulating doses of mEGF and hCG together had no further effects. However, combining these agents at suboptimal concentrations (i.e., mEGF at 3 µg/L, hCG at 20 µg/L, and [Bt2]cAMP at 0.5 mM) resulted in an increase in StAR mRNA levels in an additive fashion (Fig. 5).
|
Role of mEGF on LHR mRNA Expression and on [125I]hCG Binding in mLTC-1 Cells
Given that the stimulatory effect of mEGF on P synthesis and StAR expression is acting through specific EGF receptors (Figs. 1 and 4), its effects on LHR mRNA expression and [125I]hCG binding were studied. Northern blot analysis with an LHR-specific probe revealed multiple transcripts of 6.9, 4.8, 3.2, 2.6, and 1.8 kb in these cells (Fig. 6). However, mEGF treatment (10 µg/L) resulted in a time-dependent decrease in the steady-state levels of LHR mRNA after a lag phase of 4–6 h. Using the same experimental paradigms, a modest but consistent increase (12–20%) in [125I]hCG binding was observed by 4–6 h of mEGF stimulation and then followed the time course of LHR mRNA expression (data not shown).
|
Effects of mEGF on P Production and StAR Gene Expression Involve AA and MAPK Signaling
To understand the potential mechanism of mEGF action in the regulation of steroidogenesis and StAR mRNA expression, the roles of mEGF in the AA- and MAPK-signaling pathways were investigated. As shown in Figure 7, mLTC-1 cells stimulated with mEGF (10 µg/L) demonstrated approximately 4.6- and 12.3-fold increases in StAR mRNA and P levels, respectively. Inclusion of AA (150 µM) in the incubations modestly but consistently increased StAR mRNA expression (Fig. 7A) and P (P < 0.05) accumulation (Fig. 7B). Blocking AA release with the phospholipase A2 (PLA2)-inhibitor dexamethasone (1 µM, [26]) significantly reduced (P < 0.01) StAR mRNA expression and P production in mEGF-stimulated cells. Interestingly, the addition of exogenous AA to the dexamethasone-inhibited cells partially reversed StAR mRNA expression and P production, indicating the involvement of AA in mEGF-mediated steroidogenic responses (Fig. 7).
|
To obtain more insight regarding these mechanisms, the effects of mEGF on P synthesis were studied using inhibitors of the PLA2, lipoxygenase, phospholipase C (PLC), and cyclooxygenase pathways (Fig. 7C). Synthesis of P significantly increased in response to mEGF. The inhibitors of PLA2 ([N-(p-amylcinnamoyl)anthranilic acid], 5 µM) and lipoxygenase (nordihydroguaiaretic acid [NDGA], 70 µM; AA861, 70 µM) strongly decreased (P < 0.01) mEGF-stimulated P levels. Conversely, P production was significantly increased (P < 0.05) by the cyclooxygenase pathway-inhibitor indomethacin (Indo; 10 µM). The PLC inhibitor (U-73122, 25 µM) had no effect on mEGF-induced P synthesis. None of the inhibitors used had significant effects on basal P production (data not shown). The AA, dexamethasone, NDGA, AA861, and Indo were obtained from Sigma and the ACA and U-73122 from Calbiochem-Novabiochem Corp. (San Diego, CA); their concentrations were optimized and/or chosen based on previous findings [24, 26].
The potential involvement of the MAPK pathway in mEGF-induced P synthesis and StAR protein expression was then assessed. It can be seen that mEGF treatment (10 µg/L, 5 h) clearly increased levels of StAR protein (3.4-fold) and P (13.2-fold) (Fig. 8). The addition of MAPK/ERK (MEK)-inhibitors PD098059 (50 µM) and UO126 (10 µM) (Calbiochem) significantly decreased (P < 0.01) mEGF-induced levels of StAR protein and P synthesis at concentrations shown to inhibit MEK activity in MA-10 Leydig and Y-1 adrenal tumor cells [49]. The MEK inhibitors also decreased cAMP-stimulated StAR protein and P levels. Whereas these inhibitors decreased mEGF-induced steroidogenic responses, they had no effect on basal levels. These results suggest involvement of the MAPK pathway on mEGF-induced steroidogenic activity.
|
To gain further insight regarding these mechanisms, nonsteroidogenic COS-1 cells were transiently transfected with the trans-reporting plasmids followed by mEGF stimulation. The fusion trans-activator protein that includes the activation domain of Elk1 is phosphorylated and activated by MAPK and reflects its activity in terms of luciferase response to mEGF. As shown in Figure 9, mEGF treatment (0.1–100 µg/L) increased MAPK-mediated luciferase activity in a dose-dependent manner that was maximally 3.6- ± 0.7-fold over control, indicating specific involvement of MAPK in mEGF responsiveness.
|
Analysis of the 5'-Flanking Region of the Mouse StAR Gene for mEGF Responsiveness
To identify mEGF-responsive segments, various truncations of the 5'-flanking region of the StAR gene were studied in mLTC-1 cells (Fig. 10). Cells transfected with the -966-bp fragment demonstrated an approximately twofold increase in luciferase activity over nonstimulated cells following mEGF treatment. Deletion of -966 to -426 bp modestly decreased basal (22–30%) luciferase response but did not affect the fold-stimulation by mEGF. The -254- and -151-bp fragments resulted in mEGF-induced responses of approximately twofold over basal, indicating that one or more elements responsive to mEGF remained within this region. Cells transfected with the -110-bp fragment showed a 52% decrease in basal activity without affecting the mEGF-induced luciferase response. Basal and mEGF-stimulated responses significantly decreased when cells were transfected with the -68-bp fragment of the mouse StAR gene.
|
Several elements within the -151-bp region were assessed for mEGF responsiveness by generating mutations in each of the putative binding sites, which have previously been demonstrated to influence StAR gene transcription. The elements studied within the -151/–1-bp region were: three SF-1 binding sites, an AP-1 site, a C/EBP site, a GATA site, an Sp1 site, and an SREBP site (Fig. 11A). The results show that mLTC-1 cells transfected with -151/–1-bp StAR/luc (wild-type) demonstrated an approximately twofold increase in luciferase activity in response to mEGF (Fig. 11B). Mutations in the SF-1/1 and SF-1/3 sites decreased basal expression by approximately 75% and 90%, respectively, without affecting the mEGF-induced luciferase responses, whereas the SF-1/2 site had essentially no effect when compared to the -151-bp wild-type. Importantly, mutations in the AP-1 site resulted in a 69% decrease in basal and an 83% decrease in mEGF-induced luciferase responses. Furthermore, basal activity was decreased by 62% by mutating the C/EBP-binding site, but this mutation had no effect on mEGF-stimulated activity. Alteration of bases in the GATA, SREBP, and Sp1 recognition sites decreased basal reporter responses by 54%, 42%, and 45%, respectively, but did not affect mEGF-induced luciferase responses. These data demonstrate involvement of multiple elements on mEGF-mediated StAR promoter function, where AP-1 appears to play the most important role.
|
To further ascertain the involvement of AP-1 in mEGF-dependent StAR promoter function, the role of AP-1 family (Fos/Jun) members c-Fos, Fra-1, c-Jun, and JunD was studied (Fig. 12). The results show that the mEGF-mediated increase of approximately twofold) in luciferase activity observed with wild-type -151-bp StAR/luc was attenuated by 40–53% when mLTC-1 cells were transfected in the presence of Fos/Jun members. Basal luciferase responses were decreased modestly but consistently by Fos members (c-Fos and Fra-1) and significantly increased (P < 0.05) by Jun members (c-Jun and JunD), documenting further the specific involvement of the AP-1 family with mEGF responsiveness in mouse Leydig cells.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The classic mechanism of tropic hormone-responsive steroid synthesis occurs through increased intracellular cAMP and depends on protein synthesis. Transcriptional or translational inhibition of StAR protein results in a 90% inhibition in steroid biosynthesis, whereas approximately 10% of steroid synthesis appears to be StAR protein independent [44, 52]. Our findings demonstrate that mEGF coordinately increases the levels of P, StAR protein, and StAR mRNA in a time- and concentration-dependent manner in mLTC-1 cells, and these responses require on-going protein synthesis [43, 44, 53]. Importantly, the observed steroidogenic responses of mEGF were mediated by a single class of high-affinity (Kd, 0.53 nmol/L) binding sites. Also, high-affinity EGF receptors have been previously demonstrated in adult mouse (Kd, 0.77 nmol/L) and in immature porcine (Kd, 0.16 and 0.23 nmol/L) Leydig cells [11, 15, 54]. However, mitogenic activity of mEGF can probably be ruled out as a cause of the increased steroidogenic response in mLTC-1 cells, because this effect has not been previously demonstrated in other Leydig cells [6, 8].
The expression of StAR protein in the adrenals and gonads is regulated by cAMP, but the StAR promoter appears to lack a consensus cAMP-response element (CRE) [30, 44, 55]. This seems to be a common phenomenon for many cAMP-regulated genes in which nonconsensus CREs mediate hormonal action. We recently identified and characterized the involvement of three functional CRE half-sites in steroidogenesis and StAR gene transcription within the -97/–67-bp region of the mouse StAR gene [39]. In the present findings, whereas mEGF significantly elevated steroid synthesis and StAR expression, this increase was not associated with an alteration in intracellular cAMP levels (not illustrated). These data clearly support previous findings that demonstrated a cAMP-independent action of mEGF in the modulation of steroid biosynthesis in MA-10 and primary cultures of rat and mouse Leydig cells [6, 20]. This is not particularly surprising, because the control of steroidogenesis has been shown to be mediated both by cAMP-dependent and cAMP-independent mechanisms. For example, in addition to the well-known signal transduction pathways involved in cAMP-dependent processes, several factors that do not require cAMP and/or protein synthesis have been identified as potent stimulators of steroidogenesis. These include various interleukins, steroidogenic inducing protein, the imidazole compound calmidazolium, and a lipophilic factor from macrophages [56].
The present findings also demonstrate that mEGF is capable of potentiating the effects of hCG and cAMP on StAR gene expression, but only when these factors are present in suboptimal concentrations [6, 20]. Also, in prepubertal rat and mouse Leydig cells, EGF has been shown to acutely increase the levels of P, 17-hydroxyprogesterone, testosterone, and androstenedione, whereas its chronic effects have been shown to inhibit the stimulatory effects of LH and cAMP by decreasing 17-hydroxylase activity [20]. Our data also demonstrate that under acute conditions, mEGF has little or no effect on [125I]hCG binding and LHR mRNA expression in mLTC-1 cells, but these responses are strongly repressed following longer exposure to mEGF [22]. The chronic exposure to mEGF that markedly inhibited LHR expression appeared to be associated, at least in part, with the decrease in steroidogenic responses observed in the present findings. It has previously been demonstrated that the LHR mRNA level in Leydig cells is regulated in a complex, species-specific fashion and can be down-regulated by its own ligand (LH/hCG), cAMP, and by phorbol esters [4, 57].
Because mEGF has no effect on intracellular cAMP [19], [Ca2+]i [24], inositol phosphates, or diacylglycerol [58] levels, a pertinent question in the present study concerns the mechanism of mEGF-induced steroid synthesis and StAR expression. The binding of EGF to its receptor in the plasma membrane activates receptor tyrosine kinase activity, which is thought to mediate biological functions through several mechanisms, including receptor autophosphorylation, phosphorylation of intracellular protein, ribosomal protein S-6 phosphorylation, receptor clustering, and differentiation of target cells. However, the receptor-independent tyrosine kinase and/or a kinase-inactive (Lys721Met substitution) function of the EGF receptor have been demonstrated to induce MAPK, the prosurvival kinase Akt, c-Fos gene expression, and DNA synthesis [59–61]. The present data document that the AA pathway plays an important role in mEGF-mediated steroidogenic responses. Using inhibitors of the PLA2, lipoxygenase, PLC, and cyclooxygenase pathways, these findings clearly demonstrate that the lipoxygenase class of AA metabolites is predominantly involved in mEGF-induced P levels and StAR mRNA expression [24, 26].
In recent studies, the MEK cascade was shown to be involved in the regulation of steroid synthesis and StAR protein expression [49, 62]. Using mLTC-1 cells, our current data show that the induction of P synthesis and StAR protein expression by mEGF were decreased by MEK inhibitors. Moreover, nonsteroidogenic COS-1 cells expressing a pathway-specific Elk1 trans-activator protein that is phosphorylated and activated by MAPK resulted in a significant increase of luciferase activity in response to mEGF, demonstrating involvement of the MAPK pathway. It was also observed that MEK inhibitors significantly decreased cAMP-stimulated steroidogenic responses. In adrenal and gonadal cells, involvement of the MEK pathway has been shown to enhance StAR gene transcription and steroid production in response to forskolin [49]. Whereas MEK inhibitors attenuated these responses, they had no effect on 22R-hydroxycholesterol-stimulated pregnenolone production or on forskolin-induced P450scc mRNA levels, suggesting that their site of action is proximal to cholesterol delivery to the mitochondrial inner membrane, probably involving StAR protein expression. Conversely, inhibition of ERK activity has been demonstrated to enhance gonadotropin-stimulated P and StAR protein levels in an LHR- and FSH-receptor (FSHR)-expressing rat granulosa-derived cell line [62] and to reverse prostaglandin F2-mediated, hCG-induced inhibition of P synthesis in human granulosa-luteal cells [63]. In a positive manner, it has been found that gonadotropins, cAMP, and EGF activates the responses of MAPKs (ERK1 and ERK2) in granulosa cells [3, 64]. More specifically, EGF has been clearly shown to increase both the ERK isoforms in porcine granulosa cells, with its efficiency being higher regarding ERK2 in comparison to ERK1 [3]. These data suggest that multiple pathways are associated with cell surface receptor-effector coupling that may vary in a species- and tissue-specific manner.
The promoter sequences in the mouse, human, and rat StAR gene demonstrate extensive homology and possess recognition motifs for several transcription factors within the -254-bp region that are highly instrumental in the regulation of cAMP-mediated StAR gene expression [38, 55, 65–67]. The present data demonstrate that the mEGF-responsive region in the 5'-flanking region of the mouse StAR gene is located within the -151 bp upstream of the transcriptional start site, a region containing recognition sites for multiple cis-elements. Mutation in several of these putative binding sites resulted in a decrease in basal promoter activity without affecting mEGF-induced fold response. An exception to this was noted when a mutation in the AP-1-binding site significantly decreased basal as well as mEGF-stimulated StAR promoter function. The specific involvement of AP-1 family members (Fos/Jun) was found to be associated with a decrease in StAR promoter function in response to mEGF, an observation that is consistent with a recent finding [68]. Moreover, EGF has also been shown to increase the AP-1 family members c-Fos, c-Jun, and Jun-B mRNA levels in pig Leydig cells [69]. In addition, our data demonstrate that the elements involved in mEGF-induced StAR promoter function are the same regulatory elements that play important roles in cAMP-mediated StAR gene transcription. Endogenous cAMP most likely plays a role in mEGF-mediated steroid synthesis, because mEGF stimulation does not affect intracellular cAMP levels.
Taken together, the present findings demonstrate that mEGF significantly increases the levels of steroid synthesis and StAR expression, and that the AA and MAPK pathways play important roles. Furthermore, analysis of the 5'-flanking region of the StAR gene identified multiple mEGF-responsive elements within the -151-bp region. Each element may be sufficient to mediate part of the mEGF response. Both the present study and those of others [6, 9, 10] suggest that EGF is an important physiological regulator of steroidogenesis and, hence, testicular function. Whether EGF acting on the testis is of circulatory origin or locally produced remains to be explored.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence. FAX: 806 743 2990; doug.stocco{at}ttmc.ttuhsc.edu
Received: 3 May 2002.
First decision: 30 May 2002.
Accepted: 30 May 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
action in human granulosa-luteal cells. J Clin Endocrinol Metab 2001 86:375-380
