J. Steroid Biochem. Molec. Biol. Vol.44, No. 4-6, pp. 429-433, 1993
0960-0760/93$6.00+ 0.00 Copyright© 1993PergamonPressLtd
Printed in GreatBritain.All rightsreserved
REGULATION OVARIAN
OF
THE
GRANULOSA
RAT CELLS
AROMATASE AND
R2C
GENE
LEYDIG
IN CELLS
SUSANL. FITZPATRICK*and JOANNES. RICHARDS Department of Cell Biology,Baylor Collegeof Medicine,One Baylor Plaza, Houston, TX 77030, U.S.A. Summary--Aromatase cytochrome P450 is regulated in granulosa cells of ovarian follicles by the synergistic action of FSH and steroids. The effect of FSH can be mimicked by forskolin suggesting that transcription of the aromatase gene is regulated by cAMP. In contrast, aromatase is constitutively expressed in the rat R2C Leydig cells. To characterize the functional regions of the promoter in these two cell types, a fragment containing 534 bp of the aromatase promoter sequence and various deletion mutants were ligated to a reporter gene, chloramphenicol acetyl transferase and used in transient transfection assays. The results suggest that the region between -176 and -31 bp is essential both for cAMP regulation in granulosa cells and constitutive expression in R2C cells. Nuclear proteins from granulosa and R2C cells specifically bind the -176 fragment in an electrophoretic mobility shift assay. Binding was completed by an oligonucleotide ( - 9 0 / - 66 bp) containing a hexameric sequence, AGGTCA, which has been found in the promoters of other steroidogenic genes. These results suggest that cAMP regulation and constitutive expression of the rat aromatase promoter requires sequences between -176 and -31 bp, particularly the sequence AGGTCA at - 8 2 / - 7 7 and nuclear proteins binding to these sequences.
INTRODUCTION Ovarian production of estradiol is the physiological regulator of numerous reproductive functions; including release of the ovulatory surge of luteinizing hormone (LH), determining behavioral estrus, and priming the uterus for implantation [1]. Aromatase cytochrome P450, the enzyme which converts testosterone to estradiol, has been localized to the granulosa cells of antral follicles in the ovary [2, 3]. During follicular development, there is an increase in aromatase m R N A and activity [3], this induction is hormonally regulated by follicle stimulating hormone (FSH)[4] which acts by stimulating adenylyl cyclase and cAMP-dependent protein kinase A [5, 6]; these effects can be mimicked by forskolin [7]. In addition, various steroids, including estrogen, testosterone and dihydrotestosterone have been shown to augment FSH-stimulated aromatase activity [8-10]. Receptors for estrogens[l l] and androgens[12] have been identified in rat granulosa cells. The specific mechanisms by which peptide and steroid hormones interact remain to be determined. High levels of cAMP associated with the L H Proceedings o f the Third International Arornatase Conference. Basic and Clinical Aspects of Aromatase, Bologna,
Italy, 14-17 June 1992. *To whom correspondenceshould be addressed.
surge, however, inhibit aromatase m R N A and activity [3]. Thus, the amount of aromatase expression during follicular development is differentially regulated by low and high concentrations of cAMP. The promoters for the rat and human aromatase gene have recently been isolated. Several tissue-specific promoters have been identified for the human gene [13-17] but only an ovarian promoter has been identified for the rat gene[18]. Of the human promoters, only the placental one has been initially characterized and found to contain a tissue-specific enhancer element [19]. There is little sequence homology between the human placental promoter and that of either the human or rat ovarian promoters. However there are regions of striking homology between the human and rat ovarian promoters. Considering that expression of aromatase m R N A is regulated by cAMP and estradiol in granulosa cells, it is surprising that the rat promoter does not contain a cAMP response element (CRE, TGACGTCA)[20] nor an estrogen response element (ERE, G G T C A N N N T G A C C ) [ 2 1 ] . Nevertheless, an ERE half-site, A G G T C A , was identified in both the rat [18] and human [16] ovarian aromatase promoters. This sequence has been implicated in basal and cAMP regulated expression of several cytochrome P450 enzymes involved in steroid
429
430
SUSAN L. FITZPATRICK and JOANNE S. RICHARDS
metabolism including side chain cleavage (scc), 11/~ hydroxylase, and 21 hydroxylase [22, 23]. In contrast to the hormonal regulation of aromatase in granulosa cells, aromatase is constitutively expressed in the rat R2C Leydig cell line. Forskolin has little effect on aromatase activity except at pharmacological levels (100-500 #M) which are inhibitory [24]. Therefore, this cell line has provided a useful model for comparing the molecular mechanisms controlling constitutive expression versus hormonal induction of the aromatase gene in gonadal cells. The purpose of this study was to (I) determine the DNA sequences required for cAMP regulation in granulosa cells and constitutive expression in R2C Leydig cells and (2) identify regions of the promoter which specifically bind nuclear proteins and are believed to play a role in transcription.
EXPERIMENTAL
Transfection assays
The rat aromatase promoter ( - 5 3 4 / + 105 bp) was ligated 5' of the chloramphenicol acetyl transferase (CAT) gene in the vector pCAT Basic (Promega). Various 5' deletion mutants and two 3' deletion mutants ( - 31/+ 9; A- 176 = - 176/+ 13) of the promoter were created using restriction enzymes (Boehringer Mannheim Biochemicals, New England Biolabs). The sequence and orientation of each mutant were confirmed and the plasmids purified on CsC1 gradients. Granulosa cells were isolated from day 26 immature female rats (Holtzman) as described previously[25] and plated at a density of 1.5 x 106 viable cells/100mm dish. The cells were cultured in DMEM-F-12 media ( l : l , GIBCO) containing penicillin-streptomycin (GIBCO) and I% fetal bovine serum (FBS, Hyclone). The next day the cells were transfected with 20/~g of plasmid DNA by the calcium phosphate precipitation method[26]. Half of the cultures were treated with 7.5/~M forskolin. The CAT assay was performed with 25/~g protein for 2h according to standard protocols [26]. The radioactivity was quantified using a Betascope 603 Blot Analyzer (Betagen Corp.). R2C Leydig cells (American Type Culture Collection) were grown in Ham's F-10 (GIBCO) media supplemented with 12.5% equine serum
(GIBCO) and 2.5% FBS and plated at a density of 1.5 × 10 6 cells/60 mm dish. The transfection assay was performed as for the granulosa cells with the following exceptions. The cells were shocked with 20% glycerol in Hank's complete balanced salt solution for 2 min at 37~C and then fed with fresh media that did not contain forskolin. In addition the CAT assay was incubated tbr only 1 h. Gel shift assays
Nuclear extracts were prepared [27, 28] from granulosa cells of hypophysectomized female rats (Johnson Laboratories) treated in vivo with estradiol and FSH. A fragment of the aromatase promoter ( - 1 7 6 / + 13) was labeled with [32p]_ dCTP (ICN) and DNA polymerase I (Klenow enzyme, large fragment, New England Biolabs). The labeled fragment (13,000 cpm) was incubated with 1.4/~g crude nuclear proteins in the presence of 10 pg poly (d[I-C]) (Pharmacia) for 30 rain at 25'C according to standard protocol [29]. The bound and free DNA were separated on a 5% polyacrylarnide gel in 0.5 x TBE (0.5 × 0.089M Tris borate pH 8.3, 2 m M EDTA). The gels were dried and subjected to autoradiography. To determine the specificity of binding, 20-fold (oligonucleotides) or 50-fold (DNA fragments) molar excess unlabeled competitor DNA was added to the binding reaction. The competitors included an aromatase oligonucleotide ( - 9 0 / - 66, GAGTCTCCC A A G G T C A T C C T T G T T T c t g c a ) , a mutant aromatase oligonucleotide (GAGTCTCCCAA taTCA TCCTTGTTTctgca), and Mae III digested aromatase fragments (--176/-150, -150/---55, - 5 5 / - 3 1 , - 3 1 ' + 9 bp).
RESULTS
Promoter s'equences conferring c A M P induction o/" reporter constructs in granulosa cells
The aromatase CAT vector containing - 534 bp of 5' flanking sequence was expressed 8-fold above background levels of pCAT Basic (4 vs 0.5% conversion, data not shown) when transfected into granulosa cells. Addition of" 7.5#M forskolin to the cultures transfected with - 5 3 4 aromCAT resulted in a 7- to 8-fold increase in CAT activity (Fig. 1). In contrast, CAT activity was induced ~2-fold when forskolin was added to cultures transfected with pCAT Basic (Fig. 1). These results indicate that a regulatory region is present within - 534 bp of
Rat aromatase promoter the start of transcription. Forskolin induction of CAT activity was maintained at 6- to 7-fold even when the promoter was reduced to 176 bp (Fig. 1). Further deletion to - 8 2 bp reduced forskolin induction to 4-fold and vectors containing - 3 1 bp of 5' flanking sequence were only induced 2.5-fold by forskolin. Unlike the other deletion mutants, which all had the same 3' end ( + 1 0 5 b p ) , clone - 3 1 contained less 3' sequence ( + 9 bp). Therefore, to determine whether the loss of forskolin induction of - 3 1 / + 9 aromCAT was due to removal of 5' and not 3' sequences, 3' sequences were removed from the - 176 aromCAT (-176/+105) construct to create A-176 ( - 1 7 6 / + 13). There was little difference in the level of forskolin induction between - 1 7 6 and A-176 suggesting that transcribed sequences are not required for cAMP regulation. These results suggest that two regions, - 1 7 6 / - 8 2 and - 8 2 / - 3 1 bp, are involved in mediating cAMP regulation of aromatase gene expression in granulosa cells. -
Promoter sequences required for constitutive expression in R2C cells To determine which aromatase promoter sequences are required for constitutive expression in R2C cells, the same aromatase CAT vectors used in the granulosa cell experiments were used to transfect R2C cells. Deletion of the promoter from - 534 to - 176 bp reduced CAT activity 2to 3-fold (Fig. 1). CAT activity decreased an additional 4-fold (20 vs 5% conversion) with - 8 2 bp of 5' sequence and further decreased another 4-fold (5.4 to 1.3%) when the promoter contained only - 3 1 bp. Deletion of transcribed sequences from - 1 7 6 aromCAT did not affect constitutive expression as seen in the level of CAT activity from A-176 aromCAT (17 vs 24% conversion). Therefore, at least three regions, - 5 3 4 / - 176, - 1 7 6 / - 8 2 , and -82/31 bp, are involved in constitutive expression of aromatase in R2C cells and two of these regions correspond to those sequences required for cAMP regulation in granulosa cells.
Characterization of the protein binding region of the rat aromatase promoter
loi c 8.go=
.~,~
431
e.
~.t" 42-
-534
-302 -257 -210 .176 6-176 ,,82
-31
pCAT
Rat Aromatmm Promoter - CAT Vectors
Fig. 1. Expression of rat aromatase p r o m o t e r - - C A T vectors in granulosa cells and R2C Leydig cells. Upper panel: CAT activity for granulosa cells transfected with the deletion-reporter constructs is expressed as fold-induction in the presence (vs absence) of 7.5 # M forskolin. The results of 3-5 experiments with duplicate or triplicate samples were averaged as mean _+ SD. Lower panel: CAT activity derived from transfected R2C cells (which do not respond to forskolin) is expressed as percent conversion of [14C]chloramphenicol to the acetylated forms. AromCAT vectors contained - 5 3 4 to - 3 1 bp of 5' flanking sequence. pCAT Basic (pCAT) lacks a promoter and was a negative control. N D = not determined.
To determine if nuclear proteins from granulosa cells and R2C cells specifically bind sequences between - 1 7 6 and + 1 3 b p of the aromatase promoter, gel shift assays were performed. As shown in Fig. 2, nuclear proteins from granulosa cells bind the 32p labeled A-176 fragment in the absence of any competitor DNA (lane 1). However, addition of a 50-fold molar excess of unlabeled A-176 D N A greatly reduced the formation of all protein complexes. In order to determine the region of the A-176 fragment to which the protein(s) were binding, the A-176 fragment was digested with into four subfragments ( - 1 7 6 / - 150, - 1 5 0 / - 55, - 55/ - 31, - 31 / + 13 bp) and each was used as competitor DNA in the gel shift assay. As shown in Fig. 2 lanes 2-6, only the fragment - 1 5 0 / - 55 (lane 3) effectively competed for binding to the nuclear proteins. Because a hexameric sequence A G G T C A found to be important for basal and cAMP regulation of promoters for other steroidogenic enzymes [22, 23] is located within this region ( - 8 2 / - 7 7 bp), an oligonucleotide containing this hexamer was used as a competitor in the gel shift assay. The oligonucleotide ( - 9 0 / - 6 6 bp) competed for all protein complexes formed with the longer A-176 fragment (lane 7). In addition, mutation of the central GG's to ta (AtaTCA) within the oligonucleotide
432
St:SAY L. FITZPAIRICK and JOANNE S. RICHARDS Granulosa
o
if5
Competitor DNA
'~
Cell E x t r a c t
E
tO
I
I
tD
CD
I
I
I
I
5
6
7
8
&17E 1
2
3
4 L
Lane ~:
6
Compet!to" A
I n r ao
176,'- i~
E~
51
Fig. 2. Characterization of a protein binding region within the rat aromatase promoter. A fragment of the aromatase promoter ( A - 176/+ 13 bp) was radiolabeled and incubated with nuclear proteins from granulosa cells in an electrophoretic mobility shift assay. The products were separated on a 5% polyacrylamide gel and the gel exposed to X-ray film. Competitor D N A s , depicted below the autoradiograph, were added at a 20-fold (lanes 7 and 8) or 50-fold (lanes 2-6) molar excess to the binding reaction.
( - 9 0 / - 6 6 bp mutant) prevented it from competing for binding to nuclear proteins (lane 8). Similar gel shift results were obtained using nuclear extracts prepared from R2C cells and these same D N A fragments. Therefore these analyses indicate that nuclear proteins from granulosa cells and R2C cells specifically bind to a region of the aromatase promoter between - 9 0 and - 6 6 b p and require an A G G T C A sequence located within this region.
scriptional activation of the aromatase gene in these two cell types. However, using deletion constructs and transient transfection assays, one functional region of the rat aromatase promoter was localized to the region - 176 to --31 bp in both granulosa and R2C cells. In addition nuclear proteins from both cell types specifically bound this region of the promoter in a gel shift assay. Results from competition gel shift experiments suggested that the A G G T C A hexameric sequence at - 8 2 / - 7 7 b p was important for protein binding. A protein binding this hexameric sequence was recently isolated from a bovine adrenal cDNA library and termed SF-I [30]. The deduced amino acid sequence of this protein is similar to the Drosophila and mouse homologs offushi tarazu-factor I [31, 32] which is a member of the orphan steroid receptor superfamily. Other members of this family also bind the A G G T C A hexamer including nerve growth factor inducible gene-B [33]. The presence of both of these orphan receptors in ovarian cells [30, 33] suggest that they may also play a role in the transcriptional regulation of the aromatase gene. However, the amount of binding activity of nuclear proteins to the A G G T C A oligonucleotide was not associated with specific stages of granulosa cell differentiation or the induction of aromatase m R N A [34]. It seems likely, therefore, that the cell and developmental regulation of aromatase expression in the ovary. is dependent on transcriptional specificity which is regulated by other hexamer binding proteins. protein protein interactions, or modifications of hexameric binding proteins e.g. phosphorylation. Therefore the protein which binds the aromatase hexamer may be a member of a subfamily of orphan receptors involved in regulating cAMP induction and constitutive expression. Further studies will require isolation and characterization of the DNA-binding protein(s) from granulosa and R2C cells which is involved in regulating transcription of the aromatase gene. Acknowledgements--The authors gratefully thank Lisa O.
DISCUSSION
Expression of the aromatase gene is regulated in granulosa cells by the synergistic actions of FSH, via the cAMP signalling pathway, and estradiol. In contrast, aromatase expression is constitutively maintained in R2C cells suggesting that different cell-specific signalling pathways may be involved in regulating tran-
Levy and Rebecca Robker for technical assistance, Dr Jeffrey W. Clemens for helpful discussions, and David Scarff for art work.
REFERENCES 1. Martin C. R.: Endocrine Physiology. Oxford University Press, New York (1985). 2. Richards J. S.: Maturation of ovarian follicles: actions and interactions of pituitary and ovarian hormones on
Rat aromatase promoter follicular cell differentiation. Physiol. Rev. 60 (1980) 51-89. 3. Hickey G. J., Chen S., Besman M. J., Shively J. E., Hall P. F., Gaddy-Kurten D. and Richards J. S.: Hormonal regulation, tissue distribution, and content of aromatase cytochrome P450 messenger ribonucleic acid and enzyme in rat ovarian follicles and corpora lutea: relationship to estradiol biosynthesis. Endocrinology 122 (1988) 1426-1436. 4. Dorrington J. R., Moon Y. S. and Armstrong D. T.: Estradiol-17fl biosynthesis in cultured granulosa cells from hypophysectomized immature rats: stimulation by follicle-stimulating hormone. Endocrinology 97 (1975) 1328-1331. 5. Jonassen J. A., Bose K. and Richards J. S.: Enhancement and desensitization of hormone-responsive adenylate cyclase in granulosa cells of preantral and antral ovarian follicles: effects of estradiol and FSH. Endocrinology 111 (1982) 74-79. 6. Conti M., Kasson B. G. and Hsueh A. J. W.: Hormonal regulation of 3',5'-adenosine monophosphate phosphodiesterase in cultured rat granulosa cells. Endocrinology 114 (1984) 2361-2368. 7. Hsueh A. J. W., Adashi E. Y., Jones P. B. C. and Welsh T. H. Jr: Hormonal regulation of the differentiation of cultured ovarian granulosa cells. Endocrine Rev. 5 (1984) 76-127. 8. Daniel S. A. J. and Armstrong D. T.: Enhancement of follicle-stimulating hormone-induced aromatase activity by androgens in cultured rat granulosa cells. Endocrinology 107 (1980) 1027-1033. 9. Hillier S. G. and De Zwart F. A.: Evidence that granulosa cell aromatase induction/activation by follicle-stimulating hormone is an androgen receptor-regulated process in vitro. Endocrinology 109 (1981) 1303-1305. 10. Fitzpatrick S. L. and Richards J. S.: Regulation of cytochrome P450 aromatase messenger ribonucleic acid and activity by steroids and gonadotropins in rat granulosa ceils. Endocrinology 129 (1991) 1452-1462. 1 I. Richards J. S.: Estradiol receptor content in rat granulosa cells during follicular development: modification by estradiol and gonadotropins. Endocrinology 97 (1975) 1174-I 184. 12. Schreiber J. R. and Ross G. T.: Further characterization of a rat ovarian testosterone receptor with evidence for nuclear translocation. Endocrinology 99 (1976) 590-596. 13. Toda K., Terashima M., Kawamoto T., Sumimoto H., Yokoyama Y., Kuribayashi I., Mitsuuchi Y., Maeda T., Yamamoto Y., Sagara Y., Ikeda H. and Shizuta Y.: Structural and functional characterization of human aromatase P-450 gene. Fur. J. Biochem. 193 (1990) 559-565. 14. Harada N., Yamada K., Saito K., Kibe N., Dohmae S. and Takagi Y.: Structural characterization of the human estrogen synthetase (aromatase) gene. Biochem. Biophys. Res. Commun. 166 (1990) 365-372. 15. Mahendroo M. S., Means G. D., Mendelson C. R. and Simpson E. R.: Tissue-specific expression of human P-450arom, The promoter responsible for expression in adipose tissue is different from that utilized in placenta. J. Biol. Chem. 266 (1991) 11276-11281. 16. Means G. D., Kilgore M. W., Mahendroo M. S., Mendelson C. R. and Simpson E. R.: Tissue-specific promoters regulate aromatase cytochrome P450 gene expression in human ovary and fetal tissue. Molec. Endocr. 5 (1991) 2005-2013. 17. Kilgore M. W., Means G. D., Mendelson C. R. and Simpson E. R.: Alternative promotion of aromatase P-450 expression in the human placenta. Molec. Cell. Endocr. 83 (1992) 9-16. 18. Hickey G. J., Krasnow J. F., Beattie W. G. and Richards J. S.: Aromatase cytochrome P450 in rat
433
ovarian granulosa cells before and after luteinization: adenosine 3',5'-monophosphate-dependent and independent regulation. Cloning and sequencing of rat aromatase cDNA and 5' genomic DNA. Molec. Endocr. 4 (1990) 3-12. 19. Toda K., Miyahara K., Kawamoto T., Ikeda H., Sagara Y. and Shizuta Y.: Characterization of a cis-acting regulatory element involved in human-aromatase P-450 gene expression. Fur. J. Biochem. 205 (1992) 303-309. 20. Roesler W. J., Vandenbark G. R. and Hanson R. W.: Cyclic AMP and the induction of eukaryotic gene transcription. J. Biol. Chem. 263 (1988) 9063-9066. 21. Beato M.: Gene regulation by steroid hormones. Cell56 (1989) 335-344. 22. Honda S., Morahashi K. and Omura T.: Novel cAMP regulatory elements in the promoter region of bovine P450(I lb) gene. J. Biochem. 108 (1990) 1042-1049. 23. Price D. A., Mouw A. R., Bogerd A. M. and Parker K. L.: A shared promoter element regulates the expression of three steroidogenic enzymes. Molec. Endocr. 5 (1991) 1552-1561. 24. Lephart E. D., Peterson K. G., Noble J. F., George F. W. and McPhaul M. J.: The structure of cDNA clones encoding the aromatase P-450 isolated from a rat Leydig cell tumor line demonstrates differential processing of aromatase mRNA in rat ovary and a neoplastic cell line. Molec. Cell Endocr. 70 (1990) 31-40. 25. Kurten R. C. and Richards J. S.: An adenosine 3',5'monophosphate-responsive deoxyribonucleic acid element confers forskolin sensitivity on gene expression by primary rat granulosa cells. Endocrinology 125 (1989) 1345-1357. 26. Gorman C.: High efficiency gene transfer into mammalian cells. In DNA Cloning (Edited by D. M. Glover). IRL Press, Oxford, England, Vol. II (1985) pp. 143-190. 27. Hedin L., McKnight G. S., Liffka J., Durica J. M. and Richards J. S.: Tissue distribution and hormonal regulation of messenger ribonucleic acid for regulatory and catalytic subunits of adenosine 3',5'-monophosphatedependent protein kinases during ovarian follicular development and luteinization in the rat. Endocrinology 120 (1987) 1928-1935. 28. Hattori M., Tugores A., Veloz L., Karin M. and Brenner D. A.: A simplified method for the preparation of transcriptionally active liver nuclear extracts. DNA Cell Biol. 9 (1990) 777-781. 29. Chodosh L. A.: Mobility shift DNA-binding assay using gel electrophoresis. In Current Protocols in Molecular Biology (Edited by F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith and K. Struhl). Greene and Wiley-lnterscience, New York, Vol. II (1992) pp. 12.2.i-12.2.10. 30. Lala D. S., Rice D. A. and Parker K. L.: Steroidogenic factor l, a key regulator of steroidogenic enzyme expression, is the mouse homolog offushi tarazu-Factor I. Molec. Endocr. 6 (1992) 1249. 31. Ueda H., Sonoda S., Brown J. L., Scott M. P. and Wu C.: A sequence-specific DNA-binding protein that activates fushi tarazu segmentation gene expression. Genes Dev. 4 (1990) 624-635. 32. Tsukiyama T., Ueda H., Hirose S. and Niwa O.: Embryonal long terminal repeat-binding protein is a murine homolog of FTZ-FI, a member of the steroid receptor superfamily. Molec. Cell Biol. 12 (1992) 1286-1291. 33. Wilson T. E., Paulsen R. E., Padgett K. A. and Milbrandt J.: Participation of non-zinc finger residues in DNA binding by two nuclear orphan receptors. Science 256 (1992) 107-110. 34. Fitzpatrick S. L. and Richards J. S.: Cis-acting elements of the rat aromatase promoter required for cAMP induction in ovarian granulosa cells and constitutive expression in R2C Leydig cells. Molec. Endocr. In press.