- Email: [email protected]
Exon-specific northern analysis and rapid amplification of cDNA ends (RACE) reveal that the proximal promoter II (PII) is responsible for aromatase cytochrome P450 (CYP19) expression in human ovary
Rl
iWecufar and CeUular Endwi~~kvgy, 97 (1993) RI-R6 Q 1993 &evitz Scientific Publishers Ireland, Ltd. ~3~3-7~7/93/$06~~ MCX?.03128
Rapid Paper
Summary Estrogens are synthesized from Cl9 steroids by a unique form of cyto&rome PdsO,aromatase cytochrome P-,,, the product of the CXP19 gene). We have shown that tissue-specific expression of human P-450,,, (P-450,,,,; is dete~i~ed, in par&, by the use of alternative promoters. Previous methods of anafysis for dete~ining the specific Y-tern&i of the different transcripts included Sl nuclease protection, primer extension, and Northern analysis. fn the present study we have used the RACE procedure (rapid ampii~~atio~ of cDNA ends) to amplify and &one the 5’ tern&i af P-450m,, transcripts expressed in human corpus luteum (CL). Sequencing of the resulting clones supports the results of the previously performed studies. Specificahy, the proximaf promoter, PII, is the predominant promoter utihzed in CL, such that the start of ascription occurs 26 bp downstream of the putative TATA sequence. A minority of the clones possess an alternative 5’-end, namely 1.3. Exon-specific Northern analysis confirms that the majority of the P-450,,o, transcripts in CL tissue contain sequence specific for promoter II. Similarly, exon-specific Northern analysisindicates that transcripts in human follicles, as well as granulosa celfs in culture, contain prima~ly sequence specific for promoter II.
The biosynthesis of estrogens from androgens requires a specific enzyme e.ompIex of the ~ndopIas~~ reticulum known as aromatase ~MendeIs~~ et al., 1985). This complex is composed of a form of cytochrome P450 known as aromatase cytochrome P&e (P-450,,,, the product of the CWl9 gene) and a ubiquitous fIavoprotein, NADPH-cytochrome P450reductase (Mendelson et al., 1985; Nakajin et al., 1986; KeIIis and Vickery, 1987; Qsawa et al., 1987; Nebert et al., 1987). Aromatase cytochrome &-,a binds the C,, steroid substrate and receives reducing equivalents from ~i#tinamid~-
present~~~~: ’ Dept. of ~i~he~st~, Vanderbitt University Scftoof of Medicine, ~~shv~l~~~ TN_ USA, ’ Dept. of Molecular Genetics, University of Texas Southwestern Medical Center at Dallas, TX, USA.
adenine di~~~l~otide phosphate (NADPHS vb the reductase. P-45O~~o~, in the presence of molecular oxygen, catalyzes aromatization of the A ring of the substrate with ~ncomitant removaI of the C,, angular methyl group ~~ornp~~ and Siiteri, 1974; Akhtar et al., 1982, Caspi et al., 1984; CoIe and Robinson, 1988; Goto and Fishman, 1977) to form the corresponding estrogenic C,, product. P-450,,oM mRNA is expressed in a number of ceils and tissues, including ovarian granulosa cells (McNatty et al., 19761, testicular Leydig (V~ladares and Payne, 1979; Tsai-Morris et al., 19841, and Sertoli cells (Fritz et al., 1976), placenta ‘~Means et al., 1989; Kilgore et al., 19921, various fetaI tissues (Tapanainen et al., 1989; Price et al., 19921, adipoeytes (Grodin et at., 1973; Mahendr~~ et al., 19911, skin, and the amygdaia and h~oth~~~s of the brain {Naftolin et al., 1975; Roselli et al., 1985). During the human menstrual cycle, granuloss cells of the pre-o~iato~ folIicIe synthesize increasing amounts of estrogen, primariIy 17~“estradiol, in response to follicle-stimulating hormone (FSH). This
R2
occurs through a cyclic AMP-mediated pathway which results in increased expression of P-45O,,o, mRNA. The precursor C,, steroids are supplied by the surrounding luternizing hormone (LH)-stimulated thecal cells. Following the gonadotropin surge and ovulation, androgen and estrogen secretion decrease, but then a secondary phase of estrogen secretion occurs during the luteal phase, an event marked by a dramatic increase in levels of mRNA encoding P-450,,oM in the corpus luteum (CL) (Doody et al., 1990). Comparison of a P-450,,,, cDNA insert isolated from a human placental library with genomic clones indicated that the CYP19 gene is composed of nine single-copy coding exons, distributed over 35 kb (Fig. 1). Additionally, there are several untranslated first exons, distributed over at least 35 kb 5’ of the start of translation (Means et al., 1989). These are spliced in a tissue-specific fashion into P-450,,,, transcripts at a common 3’-splice junction upstream of the start of translation. Based on these findings, it has been proposed that regulation of CYP19 expression in the human ovary, placenta, and adipose involves the use of tissue-specific promoters upstream of these untranslated exons, and that the various tissue-specific 5’-termini present in transcripts arise as a consequence of alterp-450,,0, ative splicing (Thompson and Siiteri, 1974; Evans et al., 1987; Steinkampf et al., 1988; Simpson et al., 1981, 1989). Placental P-450,,,, transcripts have been shown to contain, by means of Northern analysis, as well as direct sequencing, sequence corresponding to either one of two untranslated first exons (I.1 or 1.2) with transcripts containing exon I.1 representing by far the predominant form in placenta (Kilgore et al., 1992; Means et al., 1991). In contrast, two other untranslated first exons (I.3 and 1.4) and their corresponding promoters may be utilized in adipose tissue in vivo or in
Promoter 1.2 Promoter Promoter
I.1
I. 1
Promoter
adipose stromal cells in culture (Mahendroo et al., 1993). In the case of the ovary, primer extension analysis and Sl nuclease protection assays employing P450 *nOM mRNA derived from CL have indicated that promoter II (PII), the promoter most proximal to exon II, dictates the transcriptional start site location (26 bp downstream of the PI1 TATA sequence) in the ovary (Means et al., 1991). These results are summarized in Fig. 1. However, the existence of other minor promoters and/or untranslated exons could not be ruled out by these studies (Means et al., 1991). Clearly, based on PCR and Northern analysis (Means et al., 19911, exons I.1 and I.2 were not being spliced onto exon II in CL tissue. In qrder to determine the nature of the 5’termini present in P-450,,,, transcripts in human corpus luteum as well as determine whether similar or different 5’-termini were present in ovarian P-450,,,, transcripts during the follicular phase of the ovarian cycle, we decided to use the rapid amplification of cDNA ends (RACE) procedure and subsequent exonspecific Northern analysis to yield the most definitive results in determining and quantitating the sequences of the various 5’-ends of P-450,,,, expressed in ovarian cells and tissue. Materials
and methods
Human Ovarian Cells and Tissue. Corpora lutea and follicles were obtained from women undergoing hysterectomy or bilateral oophorectomy for benign gynecological disease. Written consent was given preoperatively for the use of these tissues using a consent form approved by the Human Research Review Committee of the University of Texas Southwestern Medical Center at Dallas. The luteal phase was confirmed by histological examination of the endometrium. CL specimens and follicles were carefully removed from
1.3
I.4
1.4
1.2
1.3
II
I
I
I
Ill
Ill
IV
v
vIvIIvlll
I
HBR AATAAA ATTAAA
IX x
(10
KB 1
Fig.1. Schematic representation of the human P450 gene. The four untranslated exons and first coding exon (exon II) are indicated. Promoters I.1 and II and putative promoters 1.4, 1.2, and I.3 are also indicated. The size of the genomic region shown spans a distance at least 70 kb, but since the genomic clones containing exons I.1 and I.4 on the one hand, and exon I.2 on the other have not been overlapped, the true distance is still unknown. The closed bars represent translated sequences. The septum in the open bar in exon II represents the splice junction for untranslated exons 1.1-1.4. Sequences immediately to the left of the septum would be present in mature mRNA only when promoter II is utilized to promote transcription. The heme-binding region (HBR) is indicated in exon X, as are two alternative polyadenylation sites which give rise to the two transcript sizes of 3.4 and 2.9 kb.
R3
the remaining ovarian tissue, frozen rapidly in liquid nitrogen, and stored at-70°C until RNA isolation. Human granulosa-lutein cells were obtained and cultured as previously described (McAllister et al., 1990). RNA ~ul~fion. Corpora lutea and follicles were ground into a fine powder in liquid nitrogen using a mortar and pestle before being homogenized in 4.2 M guanidinium thiocyanate (pH 7.0). Following the removal of cell debris, the homogenate was layered over a 5.7 M cesium chloride solution and centrifuged at 1~ Xg for 18 h at 20°C. Granulosa cells in culture were frozen in liquid nitrogen and RNA was extracted as described (Mendelson et al., 1987). Total RNA pellets were resuspended in diethylpyrocarbonate (DEPC)-treated water and then precipitated with the addition of ethanol and potassium acetate (Chirgwin et al., 1979), Poly(A)+ RNA was isolated using olig~dT) affinity chromatography (Davis et al., 1986). RACE procedure. Complementary DNA libraries were constructed from corpora lutea by the RACE method. Construction of RACE cDNAs was performed with minor modifications as described by Frohman (Frohman, 1990; Frohman et al., 1988). The first step, reverse transcription, was performed using 1 RNA, 1 pmol primer 17 (5’Fg of poly@)+ ACITGCTGATAATGAGTGTT-3’, located in exon III), reverse transcriptase buffer (BRL), 1 mM DTT, 1 mM dNTPs, 200 units M-MLV reverse transcriptase (Bethesda Research Laboratories) in a 20 ~1 volume. The primer extension was carried out at 44°C for 1.5 h. Prior to the tailing step, the single-stranded cDNA was denatured at 65°C for 5 min, placed on ice, and then the tailing reaction was carried out using 0.1 mM dATP and terminal deo~nucleotidyl transferase (Tdt). The amplifications were performed suing the Perkin Elmer Cetus buffer system, P-450,,oM primer 24 (5’CTGGTATTGAGGATATGCCCTCATAAT-3’, located in exon II), and Tuq polyrnerase (Cetus). An aliquot of the amplified product was run on a 1.8% agarose gel to visually estimate DNA concentrations. PCR amplified cDNA Cloning of cDNA fragments. fragments were directly ligated into pCRlOO0 (Invitrogen TA cloning vector) using T4 ligase. Competent E. cc& cells (Invitrogen) were transformed with the recombinant plasmids and colonies selected on LB medium containing X-gal and kanamycin. White colonies were screened with 32P-end-labeled oligo-51 (5’-CAGGCACGATGCTGGTGATG-3’1, located in the coding region of exon II, using a modification of published methods (Wood et al., 1985). Plasmids hybridiz~equenc~g of cLXUA fragments. ing to 32P-end-labeied oligo-51 were isolated and sequenced by the dideoxy method (Sanger et al., 1977). Northern analysis. Northern analysis was performed using standard formaldehyde and sodium phosphate/glyoxai systems (Maniatis et al., 1982). Samples
PI1
I.3
1.4
IX
IX
1x
Fig. 2. Northern analysis of poly(A)+ RNA extracted from human corpus luteum tissue. CL poIy(A)+ RNA (20 pg) was probed with oligonucleotides specific for sequence expressed from promoter II (PII), exon 1.3, and exon 1.4. The blot was then stripped and reprobed with an oligonucleotide specific for exon IX. The sizes of the oligonucleotides and the incorporation of radiolabel were approximately the same in each case to permit qualitative comparison of the densities of each band. The two bands represent the two sizes of human P-450ARoM mRNA, of 3.4 and 2.9 kb.
were run on 1.1% or 1.25% agarose gels. CL poly(A)+ RNA (20 pg) (Fig. 2) and granulosa cell and CL poly(A1’ RNA (1.5 pg) (Fig. 3) were probed with oligonucleotides specific for exon IX, exon 1.3, exon 1.4, and the region between promoter II and the 3’acceptor splice junction upstream of the translation start site on exon II. A probe specific for exon I.3 was generated by PCR using oligo-72 (5’-GATAAGGTTCTATCAGACC-3’) and oligo-67 (5’-GCAGCATTTCTGACCITGG-3’). A probe specific for exon I.4 was generated by PCR using oligo-68 (5’-GTAGAACGTGACCAACTGG-3’) and oligo-69 (5’-GGTTTGACL
C
F
PII
28s. >: ,; -”
_,
-.
cDNA
18s -
1.3
Fig. 3. Northern analysis of poly(A)’ RNA extracted from human corpus luteum tissue (CL), as well as cultured granulosa cells maintained in the absence (Cl and presence (F) of 10 PM forskolin for 48 h. Poiy(A)’ RNA (1.5 pg) was probed with an oligonucleotide specific for sequence expressed from promoter II (PII). The blot was then stripped and probed with sequence specific for exon I.3 (1.3). Finally the blot was stripped and probed with the exon IX-specific probe. The positions of 2% and 18s RNA are indicated. The two bands represent the two sizes of human P-4.50,,, mRNA of 3.4 and 2.9 kb.
R4
TGGGCTGACCAG-3’). A probe specific for the region downstream from PI1 was generated by PCR using oligo-52 (5’-GCAACAGGAGCTATAGAT-3’) and oligo-56 (5’-TGTGGAAATCAAAGGGACAGA3’). A probe specific for exon IX was generated by PCR using oligo-18 (5’-CCAATATTCAGGATAATGTT-3’) and oligo-25 (5’-AGCCTTAGAAGATGATG3’). Results
A library of RACE-amplified clones was prepared using poly(A)+ RNA from pooled corpora lutea from several individuals. RACE products appeared as a smeared band centered at approximately 600 bp molecular weight. Table 1 summarizes the characteristics of the clones obtained. Of the twelve clones obtained, eleven contained sequences directly 5’ of the 3’-acceptor splice junction upstream of the translational start site in exon II, and thus specific for promoter II. Eight of these eleven clones isolated from this library of cDNA fragments extended to a position 26 bp downstream from promoter II, identical to the transcriptional start site previously indicated by primer extension and Sl nuclease protection assays (Means et al., 1991). Three longer clones were also obtained, extending to 24, 21 and 4 bp downstream from promoter II. Whereas those extending to 24 and 21 bp downstream of promoter II conceivably represent wobble in the transcriptional start site, the longest clone may contain unspliced sequence. The twelfth clone had a 5’terminus whose sequence corresponded to an incomplete extension of untranslated first exon I.3 (Mahendroo et al., 1993). Included within exon I.3 is an 11 bp element containing the promoter II TATA box. A full I.3 exon would include an additional upstream 200 bp of sequence. Premature termination by reverse transcriptase during primer extension likely resulted in the PCR amplification of the shortened 1.3-specific terminus. The translational start sites (ATG) for the P-450,R,, transcripts represented by the clones could be identified in all sequences.
In order to confirm this distribution of 5’-ends, as well as to determine if a similar distribution occurred in granulosa cells, northern analysis of poly(A)+ RNA using PII-, 1.3-, and exon IX-specific probes was employed. Corpora lutea from several different individuals were pooled for poly(A)+ RNA isolation and analysis. Poly(A)+ RNA was also prepared from granulosalutein cells maintained in the presence and absence of forskolin (10 FM). As expected, the PII-specific probe hybridized to P-450,,oM mRNA from CL (Fig. 2). The 3.4 and 2.9 kb bands are indicative of the two species of aromatase mRNA in CL which differ in the use of alternative polyadenylation signals (Toda et al., 1989). Low-level hybridization was detected between the 1.3specific probe and CL P-450,,,, poly(A)+ RNA, consistent with the isolation of a I.3-type clone by means of the RACE procedure. A Northern blot probed with 1.4-specific oligonucleotides revealed no hybridization, indicating that the level of 1.4-containing P_45O,RO, transcripts in CL was too low to be detectable by Northern analysis. Northern blotting of RNA from freshly harvested granulosa cells indicated that the transcripts contained promoter II-specific sequence (data not shown). When RNA from granulosa cells maintained in culture for 48 h in the absence or presence of forskolin was examined for the presence of P-450,,,, transcripts, the PIIspecific probe hybridized to the two mRNA species of 3.4 and 2.9 kb, as it did in the case of CL RNA (Fig. 3). Moreover, the intensity of hybridization was increased several-fold when RNA from forskolin-treated cells was employed, similar to the situation when the exon IX-specific probe was used. This is indicative that the levels of PII-specific transcripts were increased following forskolin treatment. On the other hand, when the blot was stripped and re-probed with the 1.3-specific probe, no detectable hybridization was observed (Fig. 3).
Discussion
Human TABLE
CYP19 is unique
among the cytochrome
Pdsosuperfamily of genes in that tissue-specific expres-
1
CLONES ISOLATED (A)+ RNA 5’ ENDS
FROM
Clone type
Number
I1 IIa IIb IIC I.3
8 1 1 1 1
a Distance downstream NA, not applicable.
cDNA
isolated
LIBRARY
CL POLY
5’-end of clone a 26 24 21 4 NA
of promoter
OF
II in base pairs.
sion is regulated, at least in part, by the use of different promoter regions, employing alternative splicing mechanisms. In the case of expression in placenta, the promoter used (PI.11 is distal to the start of translation, while that in CL (PII) is proximal, and in this case, no untranslated exon is spliced into the 3’-splice acceptor site upstream of the translational start site during transcript maturation. The current investigation, using the RACE procedure, has provided additional evidence for the use of promoter II in expression of ovarian P-450,,o, transcripts in both granulosa cells
R5
and CL. Furthermore, exon-specific Northern analysis of CL and granulosa cell poly(A)+ RNA supports the conclusions of the RACE experiments and provides a rough estimate of the relative abundance of promoter I.3-driven &45O,,o, transcripts (previously unreported) relative to PII-driven messages. Thus it appears that exon 1.3~specific sequence is present in very low abundance in CL transcripts, since the sequence was present in a single RACE-generated clone, and was barely detectable by Northern analysis of CL tissue (Fig. 2). It is also important to note that the same appears true of the P-450,,,, transcripts present in granulosa cells in situ and in culture. In the latter case, because only 1.5 pg of polyfAY RNA was available from the cultured cells for analysis, 1.3-containing transcripts were undetectable. Furthermore, the probes of the blots shown in Fig. 2 were of approximately the same size and specific radioactivity, therefore the densities may be directly compared. Visual inspection indicates that the abundance of the RNA hybridizing to the PI&specific probe was as great as that hybridizing to the exon IX (a coding exon) probe. Therefore we may conclude that essentially all of the CL transcripts contain PI&specific sequence. In most species examined (e.g. bovine), estrogens are formed during the follicular phase of the ovarian cycle, not by the CL, thus the process of luteinization involves a cessation of P-45O,,o, expression. The fact that the re-establishment of P-450,,,, expression during the luteal phase of the human ovarian cycle involves the same genomic region as that used in the follicular phase may imply similar mechanisms of regulation before and after luteinization in this case. Since a proximal promoter is also involved in P-450,,,, expression in ovaries of rat and chicken (Hickey et al., 1990; Matsumine et al., 1990, this genomic region is firmly established as the ovarian promoter of the 07’19 gene, and likely is the primordial promoter of the 07’19 gene, because of its proximity to the start of translation. Not surprisingly, the complex tissue-specific and temporal expression of human estrogen biosynthesis is mediated by equally complex mechanisms of CW19 gene expression. As much of this complexity is attributable to the utilization of alternative promoters, it is clearly essential to define which promoters are employed in each tissue site of expression prior to initiating studies designed to investigate the role of cis-acting elements and transcription factors in the regulation of CYP19 gene expression. The RACE procedure is a powerful technique which, if used in combination with Sl nuclease protection, primer extension, and northern analyses, can provide essential information regarding the character~ation and distribution of the 5’ termini of the different transcripts, and thus facilitate definition of the promoters responsible for their expression.
Acknowledgements This work was supported, in part, by USPHS Grants HD13234 and AG08174. Dodson Michael and Mala Mahendroo were supported, in part, by USPHS Training Grant .5-T32-HD07190. The skilled editorial assistance of Melissa Meister is gratefully acknowledged. References Akhtar, M., Calder, M.R., Corina, D.L. and Wright, J.N. (1982) B&hem. J. 201,569-580. Caspi, E., Wicha, J., ~unachal~, T., Nelson, P. and Spiteher, G. (1984) J. Am. Chem. Sot. 106, 7282-7283. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Biochemistry 18, 5294-5299. Cole, P.A. and Robinson, C.H. (1988) J. Am. Chem. Sec. 110, 1284-1285. Davis, L.G., Dibner, M. and Battey, J. (19863 in Methods in Molecular Biology, pp. 44-46, Efsevier, New York. Doody, K., Lorence, M.C., Mason, J.I. and Simpson, E.R. (1990) J. Clin. Endocrinol. Metab. 70, 1041-1045. Evans, C.T., Corbin, C.J., Saunders, C.T., Merrill, J.C., Simpson, E.R. and Mendelson, C.R. (1987) J. Biol. Chem. 262, 6914-6920. Fritz, LB., Griswald, M.D., Louis, B.F. and Dorrington, J.H. (1976) Mol. Cell. Endocrinol. 5, 289-294. Goto, J. and Fishman, J. (1977) Science 195, 80-81. Grodin, J.M., Siiteri, PK. and MacDonald, PC. (1973) J. Clin. Endocrinol. Metab. 36, 207-214. Hickey, G.T., Krasnow, J.S., Beattie, W.G. and Richards, J.S. (1990) Mol. Endocrinol. 4, 3-12. Kellis, J.T., Jr. and Vickery, L.E. (1987) J. Biol. Chem. 262, 44134420. Kilgore, M.W., Means, G.D., Mendelson, C.R. and Simpson, E.R. (1992) Mol. Cell. Endocrinol. 83, R9-R16. Mahendroo, M.S., Means, G.D., Mendelson, CR. and Simpson, E.R. (1991) J. Biol. Chem. 266, 11276-11281. Mahendroo, MS., Mendelson, CR. and Simpson, E.R. (1993) J. Biol. Chem., in press. Matsumine, H., Herbs& M.A., Ignatius Ott, S.-H., Wilson, J.D. and M~Phaul, M.J. (1991) J. Biol. Chem. 266, 19~-19~7. McAllister, J.M., Mason, J.I., Byrd, W., Trant, J.M., Waterman, M.R. and Simpson, E.R. (1990) J. Clin. Endocrinol. Metab. 71, 26-33. McNatty, K.P., Baird, D.T., Bolton, A., Chambers, P., Corker, C.S. and MacLean, H. (1976) J. Endocrinol. 71, 77-85. Means, G.D., Mahendroo, M., Corbin, C.J., Mathis, J.M., Powell, F.E., Mendelson, C.R. and Simpson, E.R. (1989) J. Biol. Chem. 264, 19385-19391. Means, G.D., Kilgore, M.W., Mahendroo, M.S., Mendelson, C.R. and Simpson, E.R. (1991) Mol. Endocrinol. 5, 2005-2013. Mendelson, CR., Wright, E.E., Porter, J.C., Evans, CT. and Simpson, E.R. (1985) Arch. Biochem. Biophys. 243,480-491. Mendelson, CR., Evans, CT. and Simpson, E.R. (1987) J. Steroid B&hem. 27, 753-757. Naftolin, F., Ryan, K.J., Davies, I.J., Reddy, V.V., Flares, F., Petro, Z., Kuhn, M., White, R.J., Takaoka, Y. and Wolin, L. (1975) Rec. Prog. Horm. Res. 31, 295-319. Nakajin, S., Shimoda, M. and Hall, P.F. (1986) Biochem. Biophys. Res. Commun. 134, 704-710. Nebert, D.W., Adesnik, M., C&m, M.J., Estabrook, R.W., Gonzaiez, F-J., Guengerich, F.P., Gunsalus, I.C., Johnson, E.F., Kemper, B., Levin, W., Philips, I.R., Sato, R. and Waterman, M.R. (1987) DNA 6, l-11.
Rh Osawa, Y., Yoshida. N.. Franckowiak, M. and Kitawaki, J. (1987) Steroids 50. 1I-28. Price, T., Aitken, J. and Simpson, E.R. (1992) J. Clin. Endocrinol. Metab. 74, 879-883. Roselli. C.E., Horton, L.E. and Resko, J.A. (1985) Endocrinology 117, 2471-2474. Sanger, F., Nicklen, S. and Co&on, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467. Simpson, E.R.. Ackerman, G.E., Smith, M.E. and Mendelson, C.R. (1981) Proc. Natl. Acad. Sci. USA 78, 5690-5694. Simpson, E.R., Merrill, J.C., Hollub, A.J., Graham-Lorence, S. and Mendelson, CR. (1989) Endocr. Rev. 10. 136-148. Steinkampf, M.P., Mendelson, C.R. and Simpson, E.R. (1988) Mol. Cell. Endocrinol. 59, 93-99.
Tapanainen, J., Voutilainen. R. and Jaffe. R.B. (1989) J. Steroid Biochem. 33, 7- Il. Thompson, E.A.. Jr. and Siiteri, P.K. (1974) J. Biol. Chem. 249. 5373-5378. Toda, K., Terashima, M.. Mitsuuchi. Y., Yamasaki, Y., Yokoyama. Y., Nojiona, S., Ushiro. H.. Maeda, T., Yamamoto, Y., Sagara. Y. and Shizuta, Y. (1989) FEBS Lett. 247, 371-376. Tsai-Morris, C.H., Aquilano. D.R. and Dufau, M.L. (1984) Ann. NY Acad. Sci. 438, 666-669. Valladares, L.E. and Payne, A.H. (1979) Endocrinology 105.431-436. Wood, W.I., Gitschier, J., Lasky, L.A. and Lawn, R.M. (1985) Proc. Natl. Acad. Sci. USA 82, 1585-1588.
iWecufar and CeUular Endwi~~kvgy, 97 (1993) RI-R6 Q 1993 &evitz Scientific Publishers Ireland, Ltd. ~3~3-7~7/93/$06~~ MCX?.03128
Rapid Paper
Summary Estrogens are synthesized from Cl9 steroids by a unique form of cyto&rome PdsO,aromatase cytochrome P-,,, the product of the CXP19 gene). We have shown that tissue-specific expression of human P-450,,, (P-450,,,,; is dete~i~ed, in par&, by the use of alternative promoters. Previous methods of anafysis for dete~ining the specific Y-tern&i of the different transcripts included Sl nuclease protection, primer extension, and Northern analysis. fn the present study we have used the RACE procedure (rapid ampii~~atio~ of cDNA ends) to amplify and &one the 5’ tern&i af P-450m,, transcripts expressed in human corpus luteum (CL). Sequencing of the resulting clones supports the results of the previously performed studies. Specificahy, the proximaf promoter, PII, is the predominant promoter utihzed in CL, such that the start of ascription occurs 26 bp downstream of the putative TATA sequence. A minority of the clones possess an alternative 5’-end, namely 1.3. Exon-specific Northern analysis confirms that the majority of the P-450,,o, transcripts in CL tissue contain sequence specific for promoter II. Similarly, exon-specific Northern analysisindicates that transcripts in human follicles, as well as granulosa celfs in culture, contain prima~ly sequence specific for promoter II.
The biosynthesis of estrogens from androgens requires a specific enzyme e.ompIex of the ~ndopIas~~ reticulum known as aromatase ~MendeIs~~ et al., 1985). This complex is composed of a form of cytochrome P450 known as aromatase cytochrome P&e (P-450,,,, the product of the CWl9 gene) and a ubiquitous fIavoprotein, NADPH-cytochrome P450reductase (Mendelson et al., 1985; Nakajin et al., 1986; KeIIis and Vickery, 1987; Qsawa et al., 1987; Nebert et al., 1987). Aromatase cytochrome &-,a binds the C,, steroid substrate and receives reducing equivalents from ~i#tinamid~-
present~~~~: ’ Dept. of ~i~he~st~, Vanderbitt University Scftoof of Medicine, ~~shv~l~~~ TN_ USA, ’ Dept. of Molecular Genetics, University of Texas Southwestern Medical Center at Dallas, TX, USA.
adenine di~~~l~otide phosphate (NADPHS vb the reductase. P-45O~~o~, in the presence of molecular oxygen, catalyzes aromatization of the A ring of the substrate with ~ncomitant removaI of the C,, angular methyl group ~~ornp~~ and Siiteri, 1974; Akhtar et al., 1982, Caspi et al., 1984; CoIe and Robinson, 1988; Goto and Fishman, 1977) to form the corresponding estrogenic C,, product. P-450,,oM mRNA is expressed in a number of ceils and tissues, including ovarian granulosa cells (McNatty et al., 19761, testicular Leydig (V~ladares and Payne, 1979; Tsai-Morris et al., 19841, and Sertoli cells (Fritz et al., 1976), placenta ‘~Means et al., 1989; Kilgore et al., 19921, various fetaI tissues (Tapanainen et al., 1989; Price et al., 19921, adipoeytes (Grodin et at., 1973; Mahendr~~ et al., 19911, skin, and the amygdaia and h~oth~~~s of the brain {Naftolin et al., 1975; Roselli et al., 1985). During the human menstrual cycle, granuloss cells of the pre-o~iato~ folIicIe synthesize increasing amounts of estrogen, primariIy 17~“estradiol, in response to follicle-stimulating hormone (FSH). This
R2
occurs through a cyclic AMP-mediated pathway which results in increased expression of P-45O,,o, mRNA. The precursor C,, steroids are supplied by the surrounding luternizing hormone (LH)-stimulated thecal cells. Following the gonadotropin surge and ovulation, androgen and estrogen secretion decrease, but then a secondary phase of estrogen secretion occurs during the luteal phase, an event marked by a dramatic increase in levels of mRNA encoding P-450,,oM in the corpus luteum (CL) (Doody et al., 1990). Comparison of a P-450,,,, cDNA insert isolated from a human placental library with genomic clones indicated that the CYP19 gene is composed of nine single-copy coding exons, distributed over 35 kb (Fig. 1). Additionally, there are several untranslated first exons, distributed over at least 35 kb 5’ of the start of translation (Means et al., 1989). These are spliced in a tissue-specific fashion into P-450,,,, transcripts at a common 3’-splice junction upstream of the start of translation. Based on these findings, it has been proposed that regulation of CYP19 expression in the human ovary, placenta, and adipose involves the use of tissue-specific promoters upstream of these untranslated exons, and that the various tissue-specific 5’-termini present in transcripts arise as a consequence of alterp-450,,0, ative splicing (Thompson and Siiteri, 1974; Evans et al., 1987; Steinkampf et al., 1988; Simpson et al., 1981, 1989). Placental P-450,,,, transcripts have been shown to contain, by means of Northern analysis, as well as direct sequencing, sequence corresponding to either one of two untranslated first exons (I.1 or 1.2) with transcripts containing exon I.1 representing by far the predominant form in placenta (Kilgore et al., 1992; Means et al., 1991). In contrast, two other untranslated first exons (I.3 and 1.4) and their corresponding promoters may be utilized in adipose tissue in vivo or in
Promoter 1.2 Promoter Promoter
I.1
I. 1
Promoter
adipose stromal cells in culture (Mahendroo et al., 1993). In the case of the ovary, primer extension analysis and Sl nuclease protection assays employing P450 *nOM mRNA derived from CL have indicated that promoter II (PII), the promoter most proximal to exon II, dictates the transcriptional start site location (26 bp downstream of the PI1 TATA sequence) in the ovary (Means et al., 1991). These results are summarized in Fig. 1. However, the existence of other minor promoters and/or untranslated exons could not be ruled out by these studies (Means et al., 1991). Clearly, based on PCR and Northern analysis (Means et al., 19911, exons I.1 and I.2 were not being spliced onto exon II in CL tissue. In qrder to determine the nature of the 5’termini present in P-450,,,, transcripts in human corpus luteum as well as determine whether similar or different 5’-termini were present in ovarian P-450,,,, transcripts during the follicular phase of the ovarian cycle, we decided to use the rapid amplification of cDNA ends (RACE) procedure and subsequent exonspecific Northern analysis to yield the most definitive results in determining and quantitating the sequences of the various 5’-ends of P-450,,,, expressed in ovarian cells and tissue. Materials
and methods
Human Ovarian Cells and Tissue. Corpora lutea and follicles were obtained from women undergoing hysterectomy or bilateral oophorectomy for benign gynecological disease. Written consent was given preoperatively for the use of these tissues using a consent form approved by the Human Research Review Committee of the University of Texas Southwestern Medical Center at Dallas. The luteal phase was confirmed by histological examination of the endometrium. CL specimens and follicles were carefully removed from
1.3
I.4
1.4
1.2
1.3
II
I
I
I
Ill
Ill
IV
v
vIvIIvlll
I
HBR AATAAA ATTAAA
IX x
(10
KB 1
Fig.1. Schematic representation of the human P450 gene. The four untranslated exons and first coding exon (exon II) are indicated. Promoters I.1 and II and putative promoters 1.4, 1.2, and I.3 are also indicated. The size of the genomic region shown spans a distance at least 70 kb, but since the genomic clones containing exons I.1 and I.4 on the one hand, and exon I.2 on the other have not been overlapped, the true distance is still unknown. The closed bars represent translated sequences. The septum in the open bar in exon II represents the splice junction for untranslated exons 1.1-1.4. Sequences immediately to the left of the septum would be present in mature mRNA only when promoter II is utilized to promote transcription. The heme-binding region (HBR) is indicated in exon X, as are two alternative polyadenylation sites which give rise to the two transcript sizes of 3.4 and 2.9 kb.
R3
the remaining ovarian tissue, frozen rapidly in liquid nitrogen, and stored at-70°C until RNA isolation. Human granulosa-lutein cells were obtained and cultured as previously described (McAllister et al., 1990). RNA ~ul~fion. Corpora lutea and follicles were ground into a fine powder in liquid nitrogen using a mortar and pestle before being homogenized in 4.2 M guanidinium thiocyanate (pH 7.0). Following the removal of cell debris, the homogenate was layered over a 5.7 M cesium chloride solution and centrifuged at 1~ Xg for 18 h at 20°C. Granulosa cells in culture were frozen in liquid nitrogen and RNA was extracted as described (Mendelson et al., 1987). Total RNA pellets were resuspended in diethylpyrocarbonate (DEPC)-treated water and then precipitated with the addition of ethanol and potassium acetate (Chirgwin et al., 1979), Poly(A)+ RNA was isolated using olig~dT) affinity chromatography (Davis et al., 1986). RACE procedure. Complementary DNA libraries were constructed from corpora lutea by the RACE method. Construction of RACE cDNAs was performed with minor modifications as described by Frohman (Frohman, 1990; Frohman et al., 1988). The first step, reverse transcription, was performed using 1 RNA, 1 pmol primer 17 (5’Fg of poly@)+ ACITGCTGATAATGAGTGTT-3’, located in exon III), reverse transcriptase buffer (BRL), 1 mM DTT, 1 mM dNTPs, 200 units M-MLV reverse transcriptase (Bethesda Research Laboratories) in a 20 ~1 volume. The primer extension was carried out at 44°C for 1.5 h. Prior to the tailing step, the single-stranded cDNA was denatured at 65°C for 5 min, placed on ice, and then the tailing reaction was carried out using 0.1 mM dATP and terminal deo~nucleotidyl transferase (Tdt). The amplifications were performed suing the Perkin Elmer Cetus buffer system, P-450,,oM primer 24 (5’CTGGTATTGAGGATATGCCCTCATAAT-3’, located in exon II), and Tuq polyrnerase (Cetus). An aliquot of the amplified product was run on a 1.8% agarose gel to visually estimate DNA concentrations. PCR amplified cDNA Cloning of cDNA fragments. fragments were directly ligated into pCRlOO0 (Invitrogen TA cloning vector) using T4 ligase. Competent E. cc& cells (Invitrogen) were transformed with the recombinant plasmids and colonies selected on LB medium containing X-gal and kanamycin. White colonies were screened with 32P-end-labeled oligo-51 (5’-CAGGCACGATGCTGGTGATG-3’1, located in the coding region of exon II, using a modification of published methods (Wood et al., 1985). Plasmids hybridiz~equenc~g of cLXUA fragments. ing to 32P-end-labeied oligo-51 were isolated and sequenced by the dideoxy method (Sanger et al., 1977). Northern analysis. Northern analysis was performed using standard formaldehyde and sodium phosphate/glyoxai systems (Maniatis et al., 1982). Samples
PI1
I.3
1.4
IX
IX
1x
Fig. 2. Northern analysis of poly(A)+ RNA extracted from human corpus luteum tissue. CL poIy(A)+ RNA (20 pg) was probed with oligonucleotides specific for sequence expressed from promoter II (PII), exon 1.3, and exon 1.4. The blot was then stripped and reprobed with an oligonucleotide specific for exon IX. The sizes of the oligonucleotides and the incorporation of radiolabel were approximately the same in each case to permit qualitative comparison of the densities of each band. The two bands represent the two sizes of human P-450ARoM mRNA, of 3.4 and 2.9 kb.
were run on 1.1% or 1.25% agarose gels. CL poly(A)+ RNA (20 pg) (Fig. 2) and granulosa cell and CL poly(A1’ RNA (1.5 pg) (Fig. 3) were probed with oligonucleotides specific for exon IX, exon 1.3, exon 1.4, and the region between promoter II and the 3’acceptor splice junction upstream of the translation start site on exon II. A probe specific for exon I.3 was generated by PCR using oligo-72 (5’-GATAAGGTTCTATCAGACC-3’) and oligo-67 (5’-GCAGCATTTCTGACCITGG-3’). A probe specific for exon I.4 was generated by PCR using oligo-68 (5’-GTAGAACGTGACCAACTGG-3’) and oligo-69 (5’-GGTTTGACL
C
F
PII
28s. >: ,; -”
_,
-.
cDNA
18s -
1.3
Fig. 3. Northern analysis of poly(A)’ RNA extracted from human corpus luteum tissue (CL), as well as cultured granulosa cells maintained in the absence (Cl and presence (F) of 10 PM forskolin for 48 h. Poiy(A)’ RNA (1.5 pg) was probed with an oligonucleotide specific for sequence expressed from promoter II (PII). The blot was then stripped and probed with sequence specific for exon I.3 (1.3). Finally the blot was stripped and probed with the exon IX-specific probe. The positions of 2% and 18s RNA are indicated. The two bands represent the two sizes of human P-4.50,,, mRNA of 3.4 and 2.9 kb.
R4
TGGGCTGACCAG-3’). A probe specific for the region downstream from PI1 was generated by PCR using oligo-52 (5’-GCAACAGGAGCTATAGAT-3’) and oligo-56 (5’-TGTGGAAATCAAAGGGACAGA3’). A probe specific for exon IX was generated by PCR using oligo-18 (5’-CCAATATTCAGGATAATGTT-3’) and oligo-25 (5’-AGCCTTAGAAGATGATG3’). Results
A library of RACE-amplified clones was prepared using poly(A)+ RNA from pooled corpora lutea from several individuals. RACE products appeared as a smeared band centered at approximately 600 bp molecular weight. Table 1 summarizes the characteristics of the clones obtained. Of the twelve clones obtained, eleven contained sequences directly 5’ of the 3’-acceptor splice junction upstream of the translational start site in exon II, and thus specific for promoter II. Eight of these eleven clones isolated from this library of cDNA fragments extended to a position 26 bp downstream from promoter II, identical to the transcriptional start site previously indicated by primer extension and Sl nuclease protection assays (Means et al., 1991). Three longer clones were also obtained, extending to 24, 21 and 4 bp downstream from promoter II. Whereas those extending to 24 and 21 bp downstream of promoter II conceivably represent wobble in the transcriptional start site, the longest clone may contain unspliced sequence. The twelfth clone had a 5’terminus whose sequence corresponded to an incomplete extension of untranslated first exon I.3 (Mahendroo et al., 1993). Included within exon I.3 is an 11 bp element containing the promoter II TATA box. A full I.3 exon would include an additional upstream 200 bp of sequence. Premature termination by reverse transcriptase during primer extension likely resulted in the PCR amplification of the shortened 1.3-specific terminus. The translational start sites (ATG) for the P-450,R,, transcripts represented by the clones could be identified in all sequences.
In order to confirm this distribution of 5’-ends, as well as to determine if a similar distribution occurred in granulosa cells, northern analysis of poly(A)+ RNA using PII-, 1.3-, and exon IX-specific probes was employed. Corpora lutea from several different individuals were pooled for poly(A)+ RNA isolation and analysis. Poly(A)+ RNA was also prepared from granulosalutein cells maintained in the presence and absence of forskolin (10 FM). As expected, the PII-specific probe hybridized to P-450,,oM mRNA from CL (Fig. 2). The 3.4 and 2.9 kb bands are indicative of the two species of aromatase mRNA in CL which differ in the use of alternative polyadenylation signals (Toda et al., 1989). Low-level hybridization was detected between the 1.3specific probe and CL P-450,,,, poly(A)+ RNA, consistent with the isolation of a I.3-type clone by means of the RACE procedure. A Northern blot probed with 1.4-specific oligonucleotides revealed no hybridization, indicating that the level of 1.4-containing P_45O,RO, transcripts in CL was too low to be detectable by Northern analysis. Northern blotting of RNA from freshly harvested granulosa cells indicated that the transcripts contained promoter II-specific sequence (data not shown). When RNA from granulosa cells maintained in culture for 48 h in the absence or presence of forskolin was examined for the presence of P-450,,,, transcripts, the PIIspecific probe hybridized to the two mRNA species of 3.4 and 2.9 kb, as it did in the case of CL RNA (Fig. 3). Moreover, the intensity of hybridization was increased several-fold when RNA from forskolin-treated cells was employed, similar to the situation when the exon IX-specific probe was used. This is indicative that the levels of PII-specific transcripts were increased following forskolin treatment. On the other hand, when the blot was stripped and re-probed with the 1.3-specific probe, no detectable hybridization was observed (Fig. 3).
Discussion
Human TABLE
CYP19 is unique
among the cytochrome
Pdsosuperfamily of genes in that tissue-specific expres-
1
CLONES ISOLATED (A)+ RNA 5’ ENDS
FROM
Clone type
Number
I1 IIa IIb IIC I.3
8 1 1 1 1
a Distance downstream NA, not applicable.
cDNA
isolated
LIBRARY
CL POLY
5’-end of clone a 26 24 21 4 NA
of promoter
OF
II in base pairs.
sion is regulated, at least in part, by the use of different promoter regions, employing alternative splicing mechanisms. In the case of expression in placenta, the promoter used (PI.11 is distal to the start of translation, while that in CL (PII) is proximal, and in this case, no untranslated exon is spliced into the 3’-splice acceptor site upstream of the translational start site during transcript maturation. The current investigation, using the RACE procedure, has provided additional evidence for the use of promoter II in expression of ovarian P-450,,o, transcripts in both granulosa cells
R5
and CL. Furthermore, exon-specific Northern analysis of CL and granulosa cell poly(A)+ RNA supports the conclusions of the RACE experiments and provides a rough estimate of the relative abundance of promoter I.3-driven &45O,,o, transcripts (previously unreported) relative to PII-driven messages. Thus it appears that exon 1.3~specific sequence is present in very low abundance in CL transcripts, since the sequence was present in a single RACE-generated clone, and was barely detectable by Northern analysis of CL tissue (Fig. 2). It is also important to note that the same appears true of the P-450,,,, transcripts present in granulosa cells in situ and in culture. In the latter case, because only 1.5 pg of polyfAY RNA was available from the cultured cells for analysis, 1.3-containing transcripts were undetectable. Furthermore, the probes of the blots shown in Fig. 2 were of approximately the same size and specific radioactivity, therefore the densities may be directly compared. Visual inspection indicates that the abundance of the RNA hybridizing to the PI&specific probe was as great as that hybridizing to the exon IX (a coding exon) probe. Therefore we may conclude that essentially all of the CL transcripts contain PI&specific sequence. In most species examined (e.g. bovine), estrogens are formed during the follicular phase of the ovarian cycle, not by the CL, thus the process of luteinization involves a cessation of P-45O,,o, expression. The fact that the re-establishment of P-450,,,, expression during the luteal phase of the human ovarian cycle involves the same genomic region as that used in the follicular phase may imply similar mechanisms of regulation before and after luteinization in this case. Since a proximal promoter is also involved in P-450,,,, expression in ovaries of rat and chicken (Hickey et al., 1990; Matsumine et al., 1990, this genomic region is firmly established as the ovarian promoter of the 07’19 gene, and likely is the primordial promoter of the 07’19 gene, because of its proximity to the start of translation. Not surprisingly, the complex tissue-specific and temporal expression of human estrogen biosynthesis is mediated by equally complex mechanisms of CW19 gene expression. As much of this complexity is attributable to the utilization of alternative promoters, it is clearly essential to define which promoters are employed in each tissue site of expression prior to initiating studies designed to investigate the role of cis-acting elements and transcription factors in the regulation of CYP19 gene expression. The RACE procedure is a powerful technique which, if used in combination with Sl nuclease protection, primer extension, and northern analyses, can provide essential information regarding the character~ation and distribution of the 5’ termini of the different transcripts, and thus facilitate definition of the promoters responsible for their expression.
Acknowledgements This work was supported, in part, by USPHS Grants HD13234 and AG08174. Dodson Michael and Mala Mahendroo were supported, in part, by USPHS Training Grant .5-T32-HD07190. The skilled editorial assistance of Melissa Meister is gratefully acknowledged. References Akhtar, M., Calder, M.R., Corina, D.L. and Wright, J.N. (1982) B&hem. J. 201,569-580. Caspi, E., Wicha, J., ~unachal~, T., Nelson, P. and Spiteher, G. (1984) J. Am. Chem. Sot. 106, 7282-7283. Chirgwin, J.M., Przybyla, A.E., MacDonald, R.J. and Rutter, W.J. (1979) Biochemistry 18, 5294-5299. Cole, P.A. and Robinson, C.H. (1988) J. Am. Chem. Sec. 110, 1284-1285. Davis, L.G., Dibner, M. and Battey, J. (19863 in Methods in Molecular Biology, pp. 44-46, Efsevier, New York. Doody, K., Lorence, M.C., Mason, J.I. and Simpson, E.R. (1990) J. Clin. Endocrinol. Metab. 70, 1041-1045. Evans, C.T., Corbin, C.J., Saunders, C.T., Merrill, J.C., Simpson, E.R. and Mendelson, C.R. (1987) J. Biol. Chem. 262, 6914-6920. Fritz, LB., Griswald, M.D., Louis, B.F. and Dorrington, J.H. (1976) Mol. Cell. Endocrinol. 5, 289-294. Goto, J. and Fishman, J. (1977) Science 195, 80-81. Grodin, J.M., Siiteri, PK. and MacDonald, PC. (1973) J. Clin. Endocrinol. Metab. 36, 207-214. Hickey, G.T., Krasnow, J.S., Beattie, W.G. and Richards, J.S. (1990) Mol. Endocrinol. 4, 3-12. Kellis, J.T., Jr. and Vickery, L.E. (1987) J. Biol. Chem. 262, 44134420. Kilgore, M.W., Means, G.D., Mendelson, C.R. and Simpson, E.R. (1992) Mol. Cell. Endocrinol. 83, R9-R16. Mahendroo, M.S., Means, G.D., Mendelson, CR. and Simpson, E.R. (1991) J. Biol. Chem. 266, 11276-11281. Mahendroo, MS., Mendelson, CR. and Simpson, E.R. (1993) J. Biol. Chem., in press. Matsumine, H., Herbs& M.A., Ignatius Ott, S.-H., Wilson, J.D. and M~Phaul, M.J. (1991) J. Biol. Chem. 266, 19~-19~7. McAllister, J.M., Mason, J.I., Byrd, W., Trant, J.M., Waterman, M.R. and Simpson, E.R. (1990) J. Clin. Endocrinol. Metab. 71, 26-33. McNatty, K.P., Baird, D.T., Bolton, A., Chambers, P., Corker, C.S. and MacLean, H. (1976) J. Endocrinol. 71, 77-85. Means, G.D., Mahendroo, M., Corbin, C.J., Mathis, J.M., Powell, F.E., Mendelson, C.R. and Simpson, E.R. (1989) J. Biol. Chem. 264, 19385-19391. Means, G.D., Kilgore, M.W., Mahendroo, M.S., Mendelson, C.R. and Simpson, E.R. (1991) Mol. Endocrinol. 5, 2005-2013. Mendelson, CR., Wright, E.E., Porter, J.C., Evans, CT. and Simpson, E.R. (1985) Arch. Biochem. Biophys. 243,480-491. Mendelson, CR., Evans, CT. and Simpson, E.R. (1987) J. Steroid B&hem. 27, 753-757. Naftolin, F., Ryan, K.J., Davies, I.J., Reddy, V.V., Flares, F., Petro, Z., Kuhn, M., White, R.J., Takaoka, Y. and Wolin, L. (1975) Rec. Prog. Horm. Res. 31, 295-319. Nakajin, S., Shimoda, M. and Hall, P.F. (1986) Biochem. Biophys. Res. Commun. 134, 704-710. Nebert, D.W., Adesnik, M., C&m, M.J., Estabrook, R.W., Gonzaiez, F-J., Guengerich, F.P., Gunsalus, I.C., Johnson, E.F., Kemper, B., Levin, W., Philips, I.R., Sato, R. and Waterman, M.R. (1987) DNA 6, l-11.
Rh Osawa, Y., Yoshida. N.. Franckowiak, M. and Kitawaki, J. (1987) Steroids 50. 1I-28. Price, T., Aitken, J. and Simpson, E.R. (1992) J. Clin. Endocrinol. Metab. 74, 879-883. Roselli. C.E., Horton, L.E. and Resko, J.A. (1985) Endocrinology 117, 2471-2474. Sanger, F., Nicklen, S. and Co&on, A.R. (1977) Proc. Natl. Acad. Sci. USA 74, 5463-5467. Simpson, E.R.. Ackerman, G.E., Smith, M.E. and Mendelson, C.R. (1981) Proc. Natl. Acad. Sci. USA 78, 5690-5694. Simpson, E.R., Merrill, J.C., Hollub, A.J., Graham-Lorence, S. and Mendelson, CR. (1989) Endocr. Rev. 10. 136-148. Steinkampf, M.P., Mendelson, C.R. and Simpson, E.R. (1988) Mol. Cell. Endocrinol. 59, 93-99.
Tapanainen, J., Voutilainen. R. and Jaffe. R.B. (1989) J. Steroid Biochem. 33, 7- Il. Thompson, E.A.. Jr. and Siiteri, P.K. (1974) J. Biol. Chem. 249. 5373-5378. Toda, K., Terashima, M.. Mitsuuchi. Y., Yamasaki, Y., Yokoyama. Y., Nojiona, S., Ushiro. H.. Maeda, T., Yamamoto, Y., Sagara. Y. and Shizuta, Y. (1989) FEBS Lett. 247, 371-376. Tsai-Morris, C.H., Aquilano. D.R. and Dufau, M.L. (1984) Ann. NY Acad. Sci. 438, 666-669. Valladares, L.E. and Payne, A.H. (1979) Endocrinology 105.431-436. Wood, W.I., Gitschier, J., Lasky, L.A. and Lawn, R.M. (1985) Proc. Natl. Acad. Sci. USA 82, 1585-1588.