Epidermal Growth Factor Increases Cortisol Production and Type II 30-Hydroxysteroid Dehydrogenase/45-44- Isomerase Expression in Human Adrenocortical Carcinoma Cells: Evidence for a Stat5- Dependent Mechanism
F. ALEX FELTUS, WILLIAM J. KOVACS, WENDELL NICHOLSON, CORRINE M. SILVA, SUBIR K. NAGDAS, NICOLE A. DUCHARME, AND MICHAEL H. MELNER
Departments of Obstetrics and Gynecology (F.A.F., N.A.D., M.H.M.) and Cell Biology (F.A.F., S.K.N., N.A.D., M.H.M.), and Division of Endocrinology (W.J.K., W.N.), Vanderbilt University School of Medicine, and Veterans Affairs (W.J.K., W.N.), Nashville, Tennessee 37232; and Department of Internal Medicine, University of Virginia (C.M.S.), Charlottesville, Virginia 22908
We tested the ability of epidermal growth factor (EGF) to regulate a key enzyme in the adrenal synthesis of glucocor- ticoids: human type II 36-hydroxysteroid dehydrogenase/45- 44-isomerase (36HSD). EGF treatment (25 ng/ml) of human adrenocortical carcinoma cells (H295R) resulted in a 5-fold increase in cortisol production and a corresponding 2-fold increase in 36HSD mRNA. Experiments were performed to determine whether EGF is acting through a previously iden- tified signal transducer and activator of transcription 5 (Stat5)-responsive element located from -110 to -118 in the human type II 36HSD promoter. A Stat5 expression construct was cotransfected with a 3ßHSD-chloramphenol acetyltrans-
ferase (CAT) reporter construct comprised of nucleotides -301→+45 of the human type II 36HSD promoter linked to the CAT reporter gene sequence. The addition of EGF at doses as low as 10 ng/ml resulted in an 11- to 15-fold increase in CAT activity. The introduction of 3-bp point mutations into critical nucleotides in the Stat5 response element obviated the EGF response. Either Stat5a or Stat5b isoforms induced CAT re- porter expression upon treatment with EGF. These results demonstrate the ability of EGF to regulate the expression of a critical enzyme (36HSD) in the production of cortisol and suggest a molecular mechanism by which this regulation occurs. (Endocrinology 144: 1847-1853, 2003)
A CRITICAL STEP in the adrenal production of glucocor- ticoids and mineralocorticoids is the conversion of pregnenolone into progesterone by the enzyme, 3ß-hydroxy- steroid dehydrogenase/45-44-isomerase (3ßHSD). 3ßHSD exists as two isoforms encoded by separate genes in humans. Type I 3ßHSD (1) is primarily expressed in the placenta and at a lower level in other tissues, such as prostate, breast, and skin, whereas type II 3ßHSD (2, 3) is expressed in adrenal, ovary, and testis. Type II 3ßHSD is known to be regulated by ACTH (4) in adult human adrenal cells, and the mechanism by which this occurs is thought to require the orphan nuclear receptor, steroidogenic factor-1 (SF-1; Ref. 5). However, ACTH-independent regulation of 3ßHSD expression has been demonstrated in the adrenal (6, 7), and alternative reg- ulatory pathways may exist.
The primate fetal adrenal is composed of an inner fetal zone and an outer neocortical (definitive) zone. The defini- tive zone differentiates into the zona glomerulosa, zona fas- ciculata, and zona reticularis, whereas the fetal zone re-
gresses after birth. A third zone, termed the transitional zone, arises from the outer edge of the fetal zone late in gestation (8). As the fetal zone does not express 3BHSD, steroids are metabolized via the 45-steroid pathway, leading to the pro- duction of dehydroepiandrosterone and dehydroepiandro- sterone sulfate, which are the primary steroid products of this zone; these precursor steroids serve as a reservoir for placental estrogen biosynthesis (9). The definitive and tran- sitional zones express 3BHSD and are thought to be the sites of fetal aldosterone and cortisol biosynthesis, respectively (10, 11). Neocortical expression of 3ßHSD appears at 22 wk, and by 28 wk gestation, 3ßHSD is widely distributed throughout the neocortex (12).
The mechanisms responsible for the differential expres- sion of 3ßHSD in the fetal adrenal are unknown, but evidence exists for the control of 3BHSD expression in the primate fetal adrenal to occur in the absence of ACTH. Induction of neo- cortical 3ßHSD expression in the fetal baboon adrenal occurs in the absence of ACTH at midgestation (13), and the anen- cephalic human fetus, which lacks ACTH production, ex- hibits normal definitive zone development (14). Thus, fetal adrenal growth and function are not completely dependent on ACTH until late in gestation, and definitive zone (a site of 3ßHSD expression) development occurs independently of ACTH.
Evidence suggests that epidermal growth factor (EGF) can
Abbreviations: A-II, Angiotensin II; BODIPY CAM, borondipyr- romethene difluoride chloramphenicol; CAT, chloramphenol acetyl- transferase; CoA, coenzyme A; EGF, epidermal growth factor; FBS, fetal bovine serum; h, human; 3ßHSD, 3ß-hydroxysteroid dehydrogenase/ 45-44-isomerase; PKC, protein kinase C; PMA, 12-O-tetradecanoyl phor- bol 13-acetate; SF-1, steroidogenic factor-1; Stat5, signal transducer and activator of transcription 5; Stat5RE, Stat5 response element.
affect adrenal development and steroid output. EGF is mi- togenic in human fetal adrenals (6,7) and has been shown to increase 3ßHSD activity in rat granulosa (15) and porcine Leydig (16) cells. Infusion of EGF into ewes for 24 h resulted in a 700% increase in cortisol levels (17). Macaque fetuses treated with EGF increased 3ßHSD protein in the adrenal with an induction in the transitional zone and an increase in adrenal weight due to hypertrophy of the definitive zone. Forskolin or 12-O-tetradecanoyl phorbol 13-acetate (TPA) induce 3ßHSD activity in human fetal adrenocortical cells; either angiotensin II (A-II) or EGF can enhance the forskolin effect (18). These studies all point to a role for EGF in adrenal development and function.
Previous work has identified a functional signal trans- ducer and activator of transcription 5 (Stat5) response ele- ment in the 5’-flanking region of the human type II 3ßHSD gene (19). As EGF has been shown to be a potent activator of Stat5 in the mouse liver (20), it is possible that EGF could induce 3ßHSD mRNA through a Stat5-mediated mechanism. EGF activation of Stat5 is thought to occur by ligand-depen- dent activation of the tyrosine kinase, Src, resulting in re- cruitment of latent Stat5 molecules via Src homology 2 do- mains from the cytoplasm to the receptor complex (21). Stat5 becomes phosphorylated on tyrosine 694 (22), and the Stat5 molecules then dimerize via association with Src homology 2 domains, translocate to the nucleus, and bind to Stat5 response elements in the regulatory regions of target genes, thereby activating transcription. We demonstrate here that EGF increases cortisol production and 3ßHSD expression in H295R cells, and that this effect can be transduced via Stat5.
Materials and Methods
Cell culture
HeLa (human cervical carcinoma) cells were maintained in DMEM/ Ham’s F-12 (Life Technologies, Inc., Grand Island, NY) with 10% fetal calf serum (Hyclone Laboratories, Inc., Logan, UT) and 50 µg/ml gen- tamicin (Sigma-Aldrich Corp., St. Louis, MO). H295R (human adreno- cortical carcinoma) cells were cultured as described above, but in 2% fetal bovine serum (FBS) and 1% ITS-X (Life Technologies, Inc.). EGF treatment was performed with 25 ng/ml purified murine EGF (gift from Stanley Cohen, Vanderbilt University, Nashville, TN). Treatment was for 24 h for chloramphenol acetyltransferase (CAT) assays and for 48 h for Northern and cortisol measurements.
Cortisol RIA
Cortisol was measured by a specific liquid phase RIA using antisera from IgG Corp. (Nashville, TN). Reference standard cortisol was from Sigma-Aldrich Corp., and radiolabeled [125I]cortisol was from Diagnos- tic Products (Los Angeles, CA). Cross-reaction in the assay was 0.02% for progesterone, 0.4% for 17a-hydroxyprogesterone, and 5.9% for 11-deoxycortisol.
Total RNA isolation and Northern analysis
Isolation of H295R total RNA was performed using an RNeasy Mini Kit (QIAGEN) following the protocol supplied with the kit. Before RNA extraction, samples in 100-mm tissue culture dishes were washed with 1× PBS and trypsinized, and pellets were quick-frozen in dry ice. After extraction and column purification, samples were eluted in ribonucle- ase-free water, quantified by absorbance at A260, and stored under excess ethanol at -20 ℃ before analysis. Northern analysis was conducted using Me2SO-glyoxal RNA sample preparation and agarose gel electrophore- sis as previously described (23). Samples (20 µg) or RNA size standards (Life Technologies, Inc.) were denatured in 1.08 M deionized glyoxal,
54% (vol/vol) Me2SO, and 10 mM sodium phosphate (pH 7.0) in a reaction volume of 30 ul for 1 h at 50 C and resolved using agarose [1.5% (wt/vol) in 10 mM sodium phosphate, pH 7.0] gel electrophoresis with 10 mM sodium phosphate (pH 7.0). RNA was transferred to a nylon membrane (Duralon-UV, Stratagene, La Jolla, CA) by capillary action with 20× SSC (1 × SSC = 0.15 M AND, 15 mM Na3 citrate) and fixed by UV cross-linking and baking for 1 h at 80 C. The blot was deglyoxalated in 0.1 M sodium phosphate (pH 9.0) for 1 h at 65 C, followed by pre- hybridization in 100 ml hybridization buffer [50% (vol/vol) deionized formamide, 5× SSC, 0.025 M sodium phosphate buffer (pH 7.0), 5% (wt/vol) sodium dodecyl sulfate, 0.1% (wt/vol) BSA, 0.1% (wt/vol) Ficoll 400, 0.1% (wt/vol) polyvinylpyrrolidone, and 0.05% (wt/vol) salmon sperm DNA] for 1 h at 42 C before the addition of 50 uCi [y-32P]deoxy-ATP-labeled human type II 3ßHSD cDNA probe. cDNA probe was generated using random primers and exo ” Klenow fragment (Prime-It II kit, Stratagene). Probes were purified from unincorporated nucleotides using Nuc-Trap columns (Stratagene). After hybridization (19 h), the blot was washed once with 2× SSC/1% (wt/vol) sodium dodecyl sulfate at 45 C for 15 min, twice with 2× SSC/0.1% (wt/vol) sodium dodecyl sulfate at 45 ℃ for 15 min, and once with 0.1x SSC/0.1% (wt/vol) sodium dodecyl sulfate for 15 min at 55 C. Autoradiography was performed at -70 ℃ with intensifying screens using Kodak BioMax MR film (Eastman Kodak Co., Rochester, NY). After autoradiography, blots were stripped of probe by washing for 5 min at 90 C in 10 mM sodium phosphate (pH 7.0), and rehybridized (19 h) with a [y-32P]deoxy- ATP-labeled constitutive probe (625-bp fragment of CHO-B) encoding ribosomal protein S2 (24). After hybridization, the blot was washed as described above. Autoradiographs were scanned using an Agfa Arcus II (Agfa Corp., Ridgefield Park, NJ) flatbed scanner and saved as TIFF (Tag Image File Format) files. RNA expression quantification was per- formed using Image software (Scion Corp., Frederick, MD).
Transient Transfection
Hela cells were transiently transfected using a modification of the calcium phosphate coprecipitation method (25). Plasmid constructs em- ployed were oStat5 (26), HA-hStat5A, and HA-Stat5B. Human type II 3ßHSD promoter fragments were inserted into pCAT-Basic (Promega Corp., Madison, WI) reporter plasmids as described previously (5). The construct containing the Stat5 response element (Stat5RE) point muta- tions (5’-TTCTGAGAA-3’ to 5’-TTTTGATTA-3’) was generated as pre- viously described (underlined nucleotides are mutated; Ref. 19). Adher- ent Hela cells were cultured to 55-65% confluence in 100-mm tissue culture dishes (Corning, Inc., Corning, NY) in 10 ml of the appropriate medium. Calcium phosphate-DNA coprecipitates were formed by drop- wise addition of equal volumes (0.5 ml) of solution A (0.24 M CaCl2 containing 15 µg of the appropriate plasmid constructs) to solution B (2X HEPES-buffered saline: 50 mM HEPES; 1.4 mM Na2HPO4; and 0.28 M NaCl, pH 7.1). Calcium phosphate:DNA precipitates were incubated at 23 C for at least 20 min and added to single 100-mm dishes of cells containing 9 ml fresh medium. Hela cells were then incubated with precipitate for 4 h at 37 C (5% CO2 and 95% air), shocked for 1 min with 15% (vol/vol) glycerol in Dulbecco’s PBS (0.137 M NaCl, 0.137 M NaCl, 0.5 mm MgCl2, 6.45 mm Na2HPO4, and 1.5 mm K2HPO4), washed three times with Dulbecco’s PBS, and incubated at 37 C for 24 h. During the final 24 h of incubation, cells were cultured in the presence or absence of appropriate treatments. Cells were then harvested using trypsin/ EDTA (Life Technologies, Inc.), pelleted, resuspended in 0.25 M Tris-HCI (pH 7.4), and stored at -70 C until assayed for CAT activity. Transfec- tions were performed in triplicate with mock negative controls. Internal transfection efficiency was monitored by cotransfection of 0.5 µg cyto- megalovirus promoter-ß-galactosidase and measurement of ß-galacto- sidase activity. Individual samples were corrected for transfection effi- ciency based upon their relative ß-galactosidase activity.
CAT assays
Frozen cell pellets were thawed on ice and lysed by sonication. Ex- tracts were heated to 60 C for 5 min to denature any endogenous acetylase/deacetylase enzymes. Soluble extracts were then separated from cell debris by centrifugation, divided into aliquots for CAT assays, and stored at -70 C before use. Fluorescent CAT assays were performed as previously described (27) with some modifications using the FLASH
CAT assay kit (Stratagene). Acetyl coenzyme A (CoA) was synthesized by reaction of CoA (Pharmacia Biotech, Piscataway, NJ) with acetic anhydride (Sigma-Alrich Corp.) as described previously (28) and stored at -70 C until use. Cell extracts (10-20 pl) were incubated in 0.25 M Tris-HC1 (pH 7.4) in a total reaction volume of 50 pl with acetyl-CoA (8.2 [M) and fluorescent borondipyrromethene difluoride chloramphenicol (BODIPY CAM) substrate (1:12.5 dilution) at 37 C for 4-8 h. Reactions were terminated by the addition of cold ethyl acetate (850 pl), followed by vigorous vortexing. An aliquot (800 ul) of extracted substrate and acetylated products was removed (organic phase), dried under vacuum, and resuspended in ethyl acetate (20 pl) before separation on thin layer chromatography plates (LK6, Whatman, Clifton, NJ) with chloroform/ methanol (9:1) for 30 min. Substrate and products were visualized under long-wave UV light (366 nm) and photographed (type 55 positive/ negative film, Polaroid, Cambridge, MA). Substrate and combined prod- uct bands were scraped from the plates, extracted, and diluted 1:10 in methanol before quantification by fluorescence spectrophotometry at excitation and emission wavelengths of 490 and 512 nm, respectively, using a fluorometer. The percent conversion of BODIPY CAM substrate to 1-, 3-, and 1,3-acetylated BODIPY CAM products was computed.
Western blot analysis
HeLa nuclear and whole cell extracts were prepared as previously described (19). The indicated volumes of nuclear and whole cell extracts were separated by 12% SDS-PAGE (29), immunoblotted with antibodies to human (h) Stat5A or hStat5B, and detected by enhanced chemilumi- nescence as previously described (30, 31). The Stat5b antibody (poly- clonal) was obtained from Santa Cruz Biotechnology, Inc. (catalog no. SC-835), and the Stat5a (monoclonal) was obtained from Transduction Laboratories, Inc. (catalog no. S21520, Lexington, KY).
Statistical analysis
Statistical significance was determined by single-factor ANOVA, fol- lowed by Bonferroni correction for multiple comparisons. Sample dif- ferences were not considered statistically significant unless P < 0.05 (divided by the number of treatment groups) according to the Bonferroni correction for multiple comparisons.
Results
EGF enhances cortisol production and 3BHSD mRNA expression in a human adrenocortical carcinoma cell line (H295R)
Treatment of H295R cells with 200 nM 12-O-tetradecanoyl phorbol 13-acetate (PMA) resulted in a 9.6-fold increase in cortisol production relative to basal, whereas 25 ng/ml EGF treatment resulted in a 5.5-fold increase (Fig. 1). Increases in cortisol production correlate with increases in 3BHSD
Cortisol (ng/ml)
30
25
T
20
15
T
10
5
0
BASAL
PMA
EGF
mRNA expression. Northern analysis showed an increase in 3ßHSD expression after PMA and EGF treatment (Fig. 2A). Densitometric analysis of band intensities showed that EGF treatment resulted in a 2.6-fold increase in mRNA, whereas PMA treatment resulted in a 7-fold increase (Fig. 2B). These data demonstrate the ability of EGF and phorbol esters to increase 3ßHSD mRNA and cortisol output by H295R cells.
Low levels of EGF induce human type II 3BHSD transcriptional activity in the presence of Stat5
We have previously characterized a functional Stat5RE located from -110 to -118 in the 5’-flanking region of the human type II 3ßHSD promoter (19). Experiments were per- formed to test the ability of this element to mediate EGF activation of the 3ßHSD gene. A -301 to +45 fragment of the human type II 3BHSD 5’-flanking region linked to the chloramphenicol aceytltransferase (CAT) reporter gene (-301CAT) was transiently transfected into Hela cells with or without an expression vector for Stat5 (Fig. 3). Treatment of the cells with 25 ng/ml EGF showed no increase in basal activity without transfection of the Stat5 expression vector. Expression of Stat5 and treatment with EGF resulted in a 15-fold increase in reporter activity after correction for trans- fection efficiency. An EGF dose-response curve in Hela cells transfected with Stat5 and-301CAT reveals that activation of 3BHSD reporter activity becomes statistically significant at 5 ng/ml EGF, reaches a maximal response at 15 ng/ml, and plateaus up to 100 ng/ml (Fig. 4). These data show that low levels of EGF can functionally activate type II 3ßHSD and that this response requires Stat5.
Both Stat5A and Stat5B can mediate EGF transduction of human type II 3BHSD promoter activity
Humans express two isoforms of Stat5 encoded by two separate genes: hStat5A and hStat5B (32, 32). We have tested the ability of hStat5A and hStat5B to mediate EGF signaling on the human type II 3BHSD CAT reporter construct (Fig. 5). Treatment with 25 ng/ml EGF and expression of hStat5A resulted in a 31-fold increase in reporter activity, whereas hStat5B expression resulted in a 22-fold increase in reporter activity. The endogenous protein expression levels of Stat5A and Stat5B in Hela cells are shown in Fig. 6. Both Stat5A and Stat5B were detected in nuclear extracts and whole cell ex- tracts of Hela cells. These data suggest that both Stat5A and Stat5B are capable of mediating EGF transduction of human type II 3BHSD promoter activity with their levels being crit- ical for this activation.
Functional activation of human type II 3BHSD transcription by EGF is mediated by the Stat5RE
To verify that the functional activation of the type II 3ßHSD gene by EGF is mediated through the Stat5RE, a CAT construct containing 3-bp point mutations in the -301→+45 CAT reporter construct converting the element from 5’- TTCTGAGAA-3’ to 5’-TTTTGATTA-3’ was employed (un- derlined nucleotides are mutated). These point mutations are in base pairs known to be critical for functional Stat5 regu- lation. Comparison of the -301(mutant) CAT promoter con-
A
B
KB
4.4+
(Ratio of 3ß-HSD/Ribosomal Protein S2)
0.4
*
T
2.4+
0.3
1.4+
+ 36-HSD
3ß-HSD mRNA
0.2
*
T
0.24 -
0.1
T
0
CHO-B
CONTROL
EGF
PMA
TTT
CONTROL EGF
PMA
struct to the -301(wild-type) CAT promoter construct is shown in Fig. 7. Both reporter constructs were transiently cotransfected into HeLa cells overexpressing Stat5. Basal and EGF-activated reporter activities were diminished by muta- tion of the Stat5 element.
Discussion
Our results demonstrate that EGF induces 36HSD mRNA expression and cortisol biosynthesis in H295R cells. We also present evidence for a mechanism by which EGF might elicit these effects on steroidogenesis by transduction through Stat5. Low doses of EGF (10-25 ng/ml) activate the 36HSD CAT reporter construct in Hela cells and require the ex- pression of Stat5. The Stat5 response element located from -110 to-118 in the type II 3ßHSD promoter region is critical to EGF activation, because point mutations abrogate this effect. The induction of transcriptional activity occurred through either Stat5A or Stat5B. These results suggest that the majority of EGF transcriptional activation of 3ßHSD oc- curs through the Stat5RE.
A number of growth factors have been implicated in the control of adrenal development. These include EGF, fibro- blast growth factor, IGF-I, IGF-II, inhibin, activin, and TGFß (8). EGF has been shown to enhance 3ßHSD activity in hu- man fetal adrenal cells (18), and EGF treatment of human fetal zone and definitive zone cells results in proliferation with a higher mitogenic response seen in the definitive zone cells (6). As 3ßHSD is expressed in definitive, but not fetal,
35%
CAT Enzymatic Activity % Conversion
30%
☐ BASAL
15
☒ EGF
I
25%
20%
15%
10%
5%
0%
Endogenous Stat5
Stat5 Overexpression
30%
CAT Enzymatic Activity % Conversion
25%
20%
15%
10%
5%
0%
0
20
40
60
80
100 120
EGF (ng/ml)
CAT Enzymatic Activity % Conversion
60%
31
BASAL
50%
T
EGF
40%
30%
20%
22
10%
0%
hStat5A
hStat5B
zone cells, it is possible that EGF is mediating this effect. Other studies have shown that injection of EGF into the fetus of rhesus monkeys results in hypertrophy of the definitive
kD
97
hSTAT5A
68
48
97
68
hSTAT5B
48
1 2
3
4
5 6
NE
WCE
CAT Enzymatic Activity % Conversion
60%
☐ BASAL
50%
T
☒ EGF
40%
30%
20%
10%
0%
wild-type
mutant
zone with an increase in 3ßHSD protein in the definitive and transitional zones (34). The mitogenic effect of EGF in the adrenal appears to be species dependent, as EGF treatment of bovine adrenocortical cells does not result in a mitogenic response (35). It should be noted that TGFa, but not EGF, is expressed in adult and fetal human adrenals (36), and acti- vation of the EGF receptor by TGFa (or other EGF receptor ligands) might mediate the induction of 3ßHSD in the ad- renal gland in vivo.
As EGF has been shown to be a potent activator of Stat5 (20), Stat5 activation could be the mechanism by which EGF increases 3ßHSD transcription in the adrenal. Previous stud-
ies on type II 3ßHSD regulation have examined a functional Stat5RE located from -110 to -118 (5’-TTCTGAGAA-3’) in the type II 3BHSD promoter region (19). In these studies it was shown that PRL-activated Stat5 binds to the Stat5RE and functionally activates 3BHSD transcription. Evidence that EGF is activating 3ßHSD through the Stat5RE presented here suggests a molecular mechanism for EGF regulation of ad- renal 3ßHSD, yet it cannot be ruled out that parallel signal transduction cascades initiated by EGF in vivo could also be affecting 3ßHSD transcription.
A-II also has the potential for regulation of 3ßHSD through a Stat5-mediated pathway. A-II acts through the AT1 and AT2 receptors and activates cAMP/protein kinase A, phospho- inositol turnover/Ca2+ flux/protein kinase C (PKC), and Janus kinase/Stat signaling pathways (37). In the zona glo- merulosa, A-II modulates aldosterone production by acti- vating AT1 receptors which involves increases in PKC ac- tivity (38). Phorbol ester increases PKC activity, and induction of the human type II 3ßHSD gene maps to a re- sponse element (5’-TCAAGGTAA-3’) that binds the orphan nuclear receptor, SF-1, and is located from -64 to -56 in the 5’-flanking sequence of the transcription initiation (5). In- creases in 3BHSD mRNA by A-II and TPA in H295R cells have been demonstrated (39). Interestingly, AT1 receptors that transduce A-II signals in zona glomerulosa cells have been shown to activate Stat5 in cardiac myocytes via the Janus kinase/Stat pathway (40). This opens the possibility that A-II could regulate 3BHSD directly through a Stat5 mechanism independently or in parallel with PKC mechanisms.
Besides EGF and PRL induction of 3ßHSD transcription via Stat5, it has been shown that the cytokine IL-4 requires the Stat5RE for functional activity in conjunction with a Stat6 response element located 5’ of the Stat5RE from -160 to-151 (Cote, S., S. Gingras, F. A. Feltus, M. Freeman, M. H. Melner, and J. Simard, manuscript in preparation). Cytokine activa- tion of 3ßHSD has important implications for the regulation of steroidogenesis in the ovary due to changes in the immune cell complement during the follicular cycle. Multiple extra- cellular factors therefore are capable of activating Stat pro- teins and converging at the Stat5 element in the human type II 3ßHSD gene. This element is independent of the region of trophic hormone action, which occurs through the SF-1 re- sponse element.
Regulation of the steroidogenic pathway by Stat proteins is not limited to 3ßHSD. Stat3 has been shown to induce aromatase activity in human adipose tissue (41). This sug- gests that control of the steroidogenic pathway can occur through Stat protein binding in the regulatory regions of steroid enzyme genes at multiple levels in the pathway. In the case of adrenal steroidogenesis, an important study would be to examine the localization patterns of the Stat5 isoforms in a primate fetal adrenal to track expression changes during 3ßHSD induction in the definitive and tran- sitional zones as well as expression differences in other ste- roidogenic enzymes. In addition, as knockout mice exist for both Stat5A (42) and Stat5B (43), defects in adrenal devel- opment and steroid production could be examined with the caveat that compensatory changes may occur.
The increase in cortisol synthesis observed could reflect
changes in the expression of multiple steroidogenic enzyme genes, including 3ßHSD. However, it is also possible EGF could alter the expression of other genes that could either directly or indirectly influence steroid synthesis. For exam- ple, the expression of steroidogenesis acute regulatory pro- tein could be altered directly by EGF. Alternatively, EGF could alter the availability of cholesterol substrate via ele- vation of low density lipoprotein receptor expression or in- creases in the de novo synthesis of cholesterol. We show that EGF treatment increases 3ßHSD expression, but the exact steps responsible for increased cortisol production will re- quire a systematic search.
In conclusion, these data demonstrate the ability of EGF to induce 3ßHSD mRNA and cortisol biosynthesis in human adrenocortical carcinoma (H295R) cells. This model system was used to test the capacity of EGF to directly induce the expression of a steroidogenic enzyme in the human adrenal gland. We have also presented a molecular mechanism by which EGF can transduce its signal to the type II 3ßHSD gene through the activation of Stat5. Induction of 3ßHSD reporter activity required the expression of Stat5, and this increase in transcriptional activity is not isoform specific. Stat5 media- tion of EGF signaling requires an intact Stat5-responsive element located from -110 to-118 bp in the regulatory region of the type II 3ßHSD gene, as point mutations in this element abrogated EGF induction. These data confirm the ability of EGF to regulate a gene that is critical to the production of adrenal steroids.
Acknowledgments
Received December 13, 2000. Accepted January 9, 2003.
Address all correspondence and requests for reprints to: Dr. Michael H. Melner, Department of Obstetrics and Gynecology, B-1100 Medical Center North, Vanderbilt University, Nashville, Tennessee 37232. E- mail: mike.melner@vanderbilt.edu.
This work was supported by NIH Training Grant 5T3-HD-07043 (to F.A.F.) and in part by NIH Grants U54-HD-37321 and EPA G7041337 (to M.H.M.). W.J.K. is the recipient of a Career Development Award (Clin- ical Investigator) from the U.S. Department of Veterans Affairs.
References
1. Simard J, Melner MH, Breton N, Low KG, Zhao HF, Periman LM, Labrie F 1991 Characterization of macaque 3HSD-hydroxy-5-ene steroid dehydroge- nase/45-44 isomerase: structure and expression in steroidogenic and periph- eral tissues in primate. Mol Cell Endocrinol 75:101-110
2. Lachance Y, Luu-The V, Verreault H, Dumont M, Rheame E, Leblanc G, Labrie F 1991 Structure of the human type II 3HSD-hydroxysteroid dehydro- genase/45-44 isomerase (3(HSD) gene: adrenal and gonadal specificity. DNA Cell Biol 10:701-711
3. Rheaume E, Lachance Y, Zhao HF, Breton N, Dumont M, deLaunoit Y, Trudel C, Luu-The V, Simard J, Labrie F 1991 Structure and expression of a new complementary DNA encoding the almost exclusive 3HSD-hydroxysteroid dehydrogenase/45-44-isomerase in human adrenals and gonads. Mol Endo- crinol 5:1147-1157
4. Doody KM, Carr BR, Rainey WE, Byrd W, Murry BA, Strickler RC, Thomas JL, Mason I 1990 3HSD-hydroxysteroid dehydrogenase/isomerase in the fetal zone and neocortex of the human fetal adrenal gland. Endocrinology 126: 2487-2492
5. Leers-Sucheta S, Morohashi K, Mason JI, Melner MH 1997 Synergistic ac- tivation of the human type II 3HSD-hydroxysteroid dehydrogenase/45-44 isomerase promoter by the transcription factor steroidogenic factor-1/adrenal 4-binding protein and phorbol ester. J Biol Chem 272:7960-7967
6. Crickard K, Ill CR, Jaffe RB 1981 Control of proliferation of human fetal adrenal cells in vitro. J Clin Endocrinol Metab 53:790-796
7. Hornsby PJ, Sturek M, Harris SE, Simonian MH 1983 Serum and growth factor requirements for proliferation of human adrenocortical cells in culture: comparison with bovine adrenocortical cells. In Vitro 19:863-869
8. Mesiano S, Jaffe RB 1997 Role of growth factors in the developmental reg- ulation of the human fetal adrenal cortex. Steroids 62:62-72
9. Siiteri PL, MacDonald PC 1963 The utilization of circulating dehydroepi- androsterone sulfate for estrogen synthesis during human pregnancy. Steroids 2:713-716
10. Seron-Ferre M, Lawrence CC, Siiteri PK, Jaffe RB 1978 Steroid production by definitive and fetal zones of the human fetal adrenal gland. J Clin Endocrinol Metab 47:603-609
11. Mesiano S, Coulter CL, Jaffe RB 1993 Localization of cytochrome P450 cho- lesterol side-chain cleavage, cytochrome P450 17@-hydroxylase/17,20-lyase, and 3HSD-hydroxysteroid dehydrogenase isomerase steroidogenic enzymes in human and rhesus monkey fetal adrenal glands: reappraisal of functional zonation. J Clin Endocrinol Metab 77:1184-1189
12. Dupont E, Luu-The V, Labrie F, Pelletier G 1990 Ontogeny of 3ß-hydroxy- steroid dehydrogenase/45-44 isomerase (3ß-HSD) in human adrenal gland performed by immunocytochemistry. Mol Cell Endocrinol 74:R7-R10
13. Leavitt MG, Albrecht ED, Pepe GJ 1999 Development of the baboon fetal adrenal gland: regulation of the ontogenesis of the definitive and transitional zones by adrenocorticotropin. J Clin Endocrinol Metab 84:3831-3835
14. Benirschke K 1956 Adrenals in anencephaly and hydrocephaly. Obstet Gy- necol 8:412-425
15. Bendell JJ, Dorrington JH 1990 Epidermal growth factor influences growth and differentiation of rat granulosa cells. Endocrinology 127:533-540
16. Sordoillet C, Chauvin MA, de Peretti E, Morera AM, Benahmed M 1991 Epidermal growth factor directly stimulates steroidogenesis in primary cul- tures of porcine Leydig cells: actions and sites of action. Endocrinology 128: 2160-2168
17. Singh-Asa P, Waters MJ 1983 Stimulation of adrenal cortisol biosynthesis by epidermal growth factor. Mol Cell Endocrinol 30:189-199
18. McAllister JM, Hornsby PJ 1988 Dual regulation of 3HSD-hydroxysteroid dehydrogenase, 17«-hydroxylase, and dehydroepiandrosterone sulfotrans- ferase by adenosine 3’,5’-monophosphate and activators of protein kinase C in cultured human adrenocortical cells. Endocrinology 122:2012-2018
19. Feltus FA, Groner B, Melner MH 1999 Stat5-mediated regulation of the human type II 3HSD-hydroxysteroid dehydrogenase/45-44 isomerase gene: activa- tion by prolactin. Mol Endocrinol 13:1084-1093
20. Ruff-Jamison S, Chen K, Cohen S 1995 Epidermal growth factor induces the tyrosine phosphorylation and nuclear translocation of Stat 5 in mouse liver. Proc Natl Acad Sci USA 92:4215-4218
21. Olayioye MA, Beuvink I, Horsch K, Daly JM, Hynes NE 1999 ErbB receptor- induced activation of stat transcription factors is mediated by Src tyrosine kinases. J Biol Chem 274:17209-17218
22. Gouilleux F, Wakao H, Mundt M, Groner B 1994 Prolactin induces phos- phorylation of Tyr694 of Stat5 (MGF), a prerequisite for DNA binding and induction of transcription. EMBO J 13:4361-4369
23. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press
24. Harpold MM, Evans RM, Salditt-Georgieff M, Darnell JE 1979 Production of mRNA in Chinese hamster cells: relationship of the rate of synthesis to the cytoplasmic concentration of nine specific mRNA sequences. Cell 17:1025-1035
25. Low KG, Chu HM, Schwartz PM, Daniels GM, Melner MH, Comb MJ 1994 Novel interactions between human T-cell leukemia virus type I Tax and ac- tivating transcription factor 3 at a cyclic AMP-responsive element. Mol Cell Biol 14:4958-4974
26. Wakao H, Gouilleux F, Groner B 1994 Mammary gland factor (MGF) is a novel member of the cytokine regulated transcription factor gene family and confers the prolactin response. EMBO J 13:2182-2191
27. Young SL, Barbera L, Kaynard AH, Haugland RP, Kang CH, Brinkley M, Melner MH 1991 A nonradioactive assay for transfected chloramphenicol acetyltransferase activity using fluorescent substrates. Anal Biochem 197:401- 407
28. Stadtman ER 1957 Preparation and assay of acyl coenzyme A and other thiol esters; use of hydroxyamine. Methods Enzymol 3:931-941
29. Towbin H, Staehelin T, Gordon J 1992 Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some ap- plications. Biotechnology 24:145-149
30. Olson GE, Nagdas SK, Winfrey VP 1997 Temporal expression and localization of protein farnesyltransferase during spermiogenesis and posttesticular sperm maturation in the hamster. Mol Reprod Dev 48:71-76
31. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685
32. Lin JX, Mietz J, Modi WS, John S, Leonard WJ 1996 Cloning of human Stat5B. Reconstitution of interleukin-2-induced Stat5A and Stat5B DNA binding ac- tivity in COS-7 cells. J Biol Chem 271:10738-10744
33. Silva CM, Lu H, Day RN 1996 Characterization and cloning of STAT5 from IM-9 cells and its activation by growth hormone. Mol Endocrinol 10:508-518
34. Coulter CL, Read LC, Carr BR, Tarantal AF, Barry S, Styne DM 1996 A role for epidermal growth factor in the morphological and functional maturation of the adrenal gland in the fetal rhesus monkey in vivo. J Clin Endocrinol Metab 81:1254-1260
35. Gospodarowicz D, Ill CR, Hornsby PJ, Gill GN 1977 Control of bovine adrenal cortical cell proliferation by fibroblast growth factor. Lack of effect of epidermal growth factor. Endocrinology 100:1080-1089
36. Sasano H, Suzuki T, Shizawa S, Kato K, Nagura H 1994 Transforming growth factor a, epidermal growth factor, and epidermal growth factor receptor ex- pression in normal and diseased human adrenal cortex by immunohistochem- istry and in situ hybridization. Mod Pathol 7:741-746
37. Sayeski PP, Ali MS, Semeniuk DJ, Doan TN, Bernstein KE 1998 Angiotensin II signal transduction pathways. Regul Pept 78:19-29
38. Balla T, Varnai P, Tian Y, Smith RD 1998 Signaling events activated by angiotensin II receptors: what goes before and after the calcium signals. Endocr Res 24:335-344
39. Bird IM, Imaishi K, Pasquarette MM, Rainey WE, Mason JI 1996 Regulation of 3HSD-hydroxysteroid dehydrogenase expression in human adrenocortical H295R cells. J Endocrinol 150:S165-S173
40. McWhinney CD, Dostal D, Baker K 1998 Angiotensin II activates Stat5 through Jak2 kinase in cardiac myocytes. J Mol Cell Cardiol 30:751-761
41. Zhao Y, Nichols JE, Valdez R, Mendelson CR, Simpson ER 1996 Tumor necrosis factor-« stimulates aromatase gene expression in human adipose stromal cells through use of an activating protein-1 binding site upstream of promoter 1.4. Mol Endocrinol 10:1350-1357
42. Liu X, Robinson GW, Wagner KU, Garrett L, Wynshaw-Boris A, Hennighausen L 1997 Stat5a is mandatory for adult mammary gland development and lacto- genesis. Genes Dev 11:179-186
43. Udy GB, Towers RP, Snell RG, Wilkins RJ, Park S-H, Ram PA, Waxman DJ, Davey HW 1997 Requirement of STAT5b for sexual dimorphism of body growth rates and liver gene expression. Proc Natl Acad Sci USA 94:7239-7244