Tissue-Specific Promoters Regulate Aromatase Cytochrome P450 Gene Expression in Human Ovary and Fetal Tissues
Gary D. Means*, Michael W. Kilgore*, Mala S. Mahendroo*, Carole R. Mendelson, and Evan R. Simpson Cecil H. and Ida Green Center for Reproductive Biological Sciences Departments of Obstetrics and Gynecology and Biochemistry University of Texas Southwestern Medical Center Dallas, Texas 75235-9051
JEG-3 cells, hydatid moles, and fetal liver appear to use promoter 1.1 and, to a limited extent, promoter I.2. These results suggest that the tissue-specific regulation of the human P450 ARO M gene is in part the consequence of the utilization of tissue-specific promoters. (Molecular Endocrinology 5: 2005-2013, 1991)
The formation of estrogens from C19 steroids is catalyzed by a specific form of cytochrome P450, aromatase cytochrome P450 (P450AROM; the product of the CYP19 gene). Previous studies have demonstrated that aromatase activity in human adipose and ovarian granulosa cells is subject to complex multifactorial regulation and that changes in activity are correlated with changes in the levels of mRNA encoding P450 ARO M. We have previously isolated the human CYP19 gene. Two unique untranslated first exons (exons 1.1 and I.2) have been identified in mRNA specific for P450AROM in human placenta. Although the proportion of transcripts encoding exon I.2 is very small, genomic clones encoding the sequences of both exons 1.1 and I.2 have recently been isolated. The corpus luteum of human ovary differs in that promoters 1.1 and 1.2 are completely inactive. Sequence analysis of the DNA immediately 5' of exon II (which contains the start site of translation) demonstrates the presence of a TATAA sequence beginning 149 basepairs 5' of the ATG initiation codon identified in placental exon II. Using a combination of primer extension and S1 nuclease protection analysis, it appears that the initiation site of ovarian P450 ARO M transcripts aligns 26 basepairs down-stream of the sequence TATAA. It appears, therefore, that the expression of P450AROM-specific mRNA in corpus luteum is regulated by an additional promoter (promoter II), which is located just 5' of exon II. Consistent with these observations, Northern analysis of poly(A)+ RNA isolated from placenta and corpus luteum demonstrates that the major promoter of placental P450AROM is promoter 1.1, while the major promoter in the corpus luteum is promoter II. Analysis of the tissue-specific utilization of these promoters was accomplished by means of the polymerase chain reaction to amplify specific 5'-termini from mRNA templates. In addition to placenta,
INTRODUCTION
The formation of estrogens from C19 steroids is catalyzed by an enzyme complex termed aromatase, which is present in the endoplasmic reticulum of cells in which it is expressed. This enzyme complex is comprised of a specific form of cytochrome P450, aromatase cytochrome P450 (P450 ARO M; P450XIX; the product of the CYP19 gene), and a flavoprotein, NADPH-cytochrome P450 reductase, which is a ubiquitous component of most cells (1-5). Aromatase activity is present in a number of human tissues and cell types, including syncytiotrophoblast of placenta (6), hydatid moles (7), JEG-3 cells (a choriocarcinoma-derived cell line) (8), fetal hepatocytes (9), adipocytes (10), and ovarian granulosa cells (11). Additionally, the activity is present in Sertoli (12) and Leydig cells (13, 14) in the male and several sites in the brain of both sexes in the rat, including the hypothalamus and amygdyla (15,16). In the human, the principal estrogen formed differs in different tissues. For example, in placenta, the major estrogen product is estriol; in adipose tissue, estrone; and in ovarian granulosa cells, 17/3-estradiol. Studies using a full-length cDNA encoding human P450 ARO M expressed in COS-1 cells indicate that a single polypeptide is capable of catalyzing the aromatization of all three classes of C19 substrates, namely 16«-hydroxylated androgens, androstenedione, and testosterone. These results suggest that the tissue-specific differences in estrogen formation reflect the presentation of these different Ci 9 substrates to the same enzyme,
0888-8809/91 /2005-2013$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
2005
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
Vol5No. 12
MOL ENDO-1991 2006
rather than the existence of different forms of P450 ARO M (17). Previous studies have demonstrated that aromatase activity in human adipose and ovarian granulosa cells is subject to complex multifactorial regulation and that changes in activity are correlated with changes in the levels of mRNA encoding P450AROM (18-21). To study the molecular mechanisms involved in the tissue-specific and multifactorial regulation of P450AROM, we have isolated the gene encoding P450AROM using a full-length insert isolated from a primer-extended human placental cDNA library (22). The aromatase P450 gene is unique among steroidogenic cytochrome P450 genes in that the first exon of the placental mRNA is untranslated, with exon II encoding the ATG from which translation is initiated (22-24). Analysis of genomic clones suggests that the nine coding exons of the P450 ARO M gene span 36 kilobases (kb). A genomic clone encoding the first exon (exon 1.1) has been identified which resides at least 35 kb 5' of exon II. The gene encoding P450 ARO M is, therefore, at least 72 kb in length. Northern and primer extension analyses of placental mRNA suggest that the promoter up-stream of exon 1.1 is the major promoter in placenta (21). A second unique cDNA insert encoding P450 ARO M mRNA was isolated from the primer-extended placental library that differed in its 5'-untranslated region from the cDNA that encodes the sequences for exon 1.1. The two species of cDNAs are identical down-stream from the sequence identified as the splice junction separating exon 1.1 from exon II, suggesting that the unique 5' leader sequence of the second form of placental P450AROM mRNA is encoded by an alternative untranslated first exon which is presumably controlled by a unique promoter (tentatively termed promoter 1.2). A genomic clone encoding the sequence of exon 1.2 has recently been isolated; mRNAs containing sequence encoded by exon 1.2, however, account for only a minor portion of the transcripts present in placenta (Kilgore, M. W., et al., unpublished observations). Transcripts containing exon 1.1 are the more abundant form of P450 ARO M encoded in placental mRNA (21). A schematic map of the orientation of the two differentially expressed first exons with respect to exon II is shown in Fig. 1. In addition to these alternatively spliced first
HBR
V
-H-fif
AATAAA ATTAAA
VIVIIVIII IX X
Fig. 1. Schematic Representation of the Human P450AROM Gene The closed bars represent translated sequences. The septum in the open bar in exon II represents the splice junction for exons 1.1 and I.2; sequences to the left of the septum would be present in mature RNA only when a putative TATA box 149 bp 5' of the ATG is used to promote transcription.
exons and their respective promoters, a TATAA sequence has been identified that begins 149 bp 5' of the ATG initiation codon identified in placental exon II (2224), suggesting the presence of an additional promoter (promoter II). The objective of the present investigation was to identify the promoter regulating the transcription of P450AROM mRNA in the human ovary. Additionally, tissue-specific utilization of the promoters of untranslated exons 1.1 and 1.2 as well as the putative promoter element identified 5' of the ATG initiation codon in exon II was evaluated by seeking to detect the unique sequences they encode in RNA extracted from tissues expressing aromatase activity. The results suggest that the tissue-specific regulation of the human P450 ARO M gene is in part the consequence of the utilization of tissue-specific promoters.
RESULTS
The 5'-flanking sequences up-stream of promoter II were analyzed to identify potential c/s-regulatory sequences of ovarian P450 ARO M expression. Sequences similar to the consensus sequences for binding of known transcriptional activators are indicated by boxes (Fig. 2), positions of divergence from the consensus sequences are marked by an asterisk. In addition to the TATA sequence previously described, a potential CAAT box was identified beginning at - 6 6 . Two half-sites for glucocorticoid receptor binding were identified at -855 and - 6 7 1 . Two potential cAMP response elementbinding protein-binding sites were identified beginning at -522 and -292. A consensus AP-1 site was found beginning at -498, and a sequence with similarity at all but one position was present beginning at -932. Two sites similar to the consensus sequence for NF-1 were found beginning at -700 and -553, and an SP-1 site at -583 was also indentified. A sequence identical to the consensus sequence reported by Kerr et al. (25) for conferring negative regulation by transforming growth factor-/? was identified beginning at -469. An additional motif has been identified at -132. This sequence, CCAAGGTCA (indicated by underlining), matches exactly the motif defined by Honda et al. (26) and Bogerd et al. (27), which has been found to be an important regulator of the genes of several steroidogenic enzymes. A schematic representation of the oligonucleotides and probes used in this study is shown in Fig. 3. Northern analysis of total placental RNA demonstrated that a probe specific for exon 1.1 hybridizes strongly to transcripts of 3.4 and 2.9 kb, which is consistent with the expected size for P450AROM-specific RNA (21). To determine whether exon 1.1-containing transcripts are present in corpus luteum, Northern analysis was conducted on poly(A)+ RNA isolated from placenta and corpora lutea. Whereas exon 1.1 is readily detectable in poly(A)+ RNA from placenta, mRNA transcripts containing this sequence were undetectable
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
2007
Tissue-Specific Promoters of Human P450AROM
-950
ACTGAAATGC ATTAATGWTG ACTCA|CTCTT CCTCACTCTA CAAGTTGTCA
-900
GRE ACCTACACCT CTTCAGCTAC AGACTACCTA CCATCCCTGA AACTqrGTTC
-850
ifcAGAGTAAA
TCTTAAAAAA CACAGCAGAA CCAGCACATC AGACTGTAAA TTGATTGTCT
-750
TGCACAGGAT GTTAGCTGCT CTTCGAATGA GGTTCCTGAG TGGCACCTGA GRE NF-1
-650 -600 -550 -500 -450 -400 -350 -300
GCC^ATJTGCT
Primer extension
GTTCTICTCTT
TCTTATAATT
-200
CTACCCACTC
-150
AGGAACCTGA
-100
GATCTTTCTT
-50
CCTAAATGTC
+1
GCACCCTCTG
+ 51
TAATTTTTCT AAGCAACAGG
AAGTCAKAAA TTTGGCTTTC TGATAAGGTT TGGCAAGAAA AGGTCAGAAA AAGGGCAAGA GTTTTGACTT GACTCTACCA ATAAflJACAGT GGGCTTCCTT AGCTATAGAT TGATCACATJF
y
GCAATTCTTT TTCTTCCTCT
S1 protection
TAATCTGECC GGATACTGTT
CRE AGATGGCCTG AAGTCAbTAG TGATGGAAGG CTCTGAGAAG ACTGTAGAAA CTTACTATTT TGF-B AGTTGGGA CJATTGCCTTT
Fig. 3. Schematic Representation of the Various Oligonucleotides Used
AGAAGATTGC CTAAACAAAA CTCTGGAATG AGCTTTATTT
Promoter
AATTGGGAAT GCACGTCACT
II
CTATCAGACC AAGCGTCTAA TGCTGCAATT CAAGCCAAAA GTA^CCATKA ATTAGTCTTG AAGTGAATCT GTACTGTACA GAACCTTTTA GGGGATTCTG ACTCTAAATT GCCCCCTCTG
TTTCCACAGG GTTTTGGAAA TGCTGAACCC GATACATTAT oIleHisTyr ValLeuGluM etLeuAsnPr
+ 151
AGGTCAAGGA GTCCCTTTGA ACACAAGATG Met AACATCACCA GCATCGTGCC AsnlleThrS erlleValPr
+201
CCTGCTCCTC ACTGGCCTTT TTCTCTTGGT GTGGAATTAT GAGGGCACAT lLeuLouLeu ThrGlyLeuP heLeuLeuVa lTrpAsnTyr GluGlyThrS
+251
CCTCAATACC AGgtaagtca gtcatttatt tctgtatcta aggagattat erSorllePr OG
+ 301
Northern probes
TCTTCCTCCT
GGTGGCATCC TATTCTGCCJF
TCCCCATTCC TTTCATTCTC TTCTCCCTTA r-Sl TTTTCCACAC CGGlTCCCTAG TACCGTTpGJ JATJGGTACAA ATCTTCAiffG~ GAGATTTTAG AGJTGAGTCAb ATAGACAAAC TTTGAATTCA GRE ACCTCAACGA ATdTGTTCitr TGCCCAAGAA TGATCAAAAA GTCAAAAAGG AGTCATTTTG TTGTTTTGAA CTTCAAGGGA ATTGATTTGG CCTGCTG, ATGACTCCAC £EE
-250
+ 101
PCR reactions
GGGATTACAA AACCTGGCTG AAAAGACAGA TTCAATGGCA
-800
-700
Procedures 52,50^
TGAAGCCATG CCTGCTGCCA CCATGCCAGT oGluAlaMet ProAlaAlaT hrMetProVa
ttacttggga ttttggtcca tcatggtaaa gaaaaatttt gcaaaaagga
C +400
tgaggacttg gagaaatttc agaccaatta tctg
Fig. 2. Sequence of the 5'-Flanking DNA Up-Stream of Exon II Corresponding to Promoter II Sequences with similarity to the consensus sequences for binding of known transcriptional activators are boxed, as are the putative TATA and CAAT boxes. Nucleotides that diverge from the consensus sequences are indicated with an asterisk. The underlined sequence has been found to be important in the regulation of several steroidogenic genes (26, 27), notably 11 i8-hydroxylase. The arrows indicate the positions of the placental intron/exon II boundary and the boundary between exon II and the first intron in the coding region. The bases are numbered such that +1 represents the start of transcription using promoter II, as defined by the primer extension and S1 nuclease protection experiments described in this manuscript.
using poly(A)+ RNA isolated from corpora lutea (Fig. 4). By contrast, when a radiolabeled probe extending upstream from the 5'-splice junction of placental exon II and through the TATA sequence identified at - 2 6 (Figs. 2 and 3) was used as a hybridization probe, a strong hybridization signal was evident in mRNA from the corpus luteum, but not from placenta (Fig. 4). S1 nuclease protection analysis of poly(A)+ RNA isolated from placenta, fetal liver, and corpus luteum was performed to determine whether the TATA sequence identified at - 2 6 is involved in promotion of the P450AROM gene in the corpus luteum. A 594-basepair (bp) probe was synthesized (Fig. 3), encompassing
P
C
P
Fig. 4. Northern Analysis of Poly(A)+ RNA Isolated from Corpus Luteum (C; 10 ^g) and Placenta (P; 10 ^g) PCR-generated probes specific for exon 1.1 or for the 5'untranslated region specific for putative promoter II (up-stream of the placental intron/exon II splice junction) were employed.
exon II and extending 5' of the putative promoter II transcription start site (estimated to be 26 bp downstream of the potential TATA sequence). When poly(A)+ RNA from fetal liver or placenta was used, the protected fragment corresponded in size (183 bp) to exon II (Fig. 5A), indicating a splice boundary between exon II and exons 1.1 or 1.2. By contrast, when RNA from corpus luteum was employed, the major protected fragment observed (262 bp) encompassed exon II, extending to a region 26 bp 3' of the TATA sequence, which is consistent with the use in this tissue of putative promoter II. Extensive overexposure of the film suggested that a promoter ll-specific mRNA transcript was not present in either fetal liver or placenta. These data are consistent with the utilization of putative promoter II in corpus luteum, whereas in fetal liver and placenta, promoters up-stream of exons 1.1 and 1.2 are being used. In addition to the major fragment discussed above, when mRNA isolated from corpus luteum was analyzed by S1 nuclease protection analysis, a minor protected fragment was observed which aligned with the placental exon II splice boundary (Fig. 5A). The
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
Vol5No. 12
MOL ENDO-1991 2008
TATA Probe
TATA
Gary D. Means*, Michael W. Kilgore*, Mala S. Mahendroo*, Carole R. Mendelson, and Evan R. Simpson Cecil H. and Ida Green Center for Reproductive Biological Sciences Departments of Obstetrics and Gynecology and Biochemistry University of Texas Southwestern Medical Center Dallas, Texas 75235-9051
JEG-3 cells, hydatid moles, and fetal liver appear to use promoter 1.1 and, to a limited extent, promoter I.2. These results suggest that the tissue-specific regulation of the human P450 ARO M gene is in part the consequence of the utilization of tissue-specific promoters. (Molecular Endocrinology 5: 2005-2013, 1991)
The formation of estrogens from C19 steroids is catalyzed by a specific form of cytochrome P450, aromatase cytochrome P450 (P450AROM; the product of the CYP19 gene). Previous studies have demonstrated that aromatase activity in human adipose and ovarian granulosa cells is subject to complex multifactorial regulation and that changes in activity are correlated with changes in the levels of mRNA encoding P450 ARO M. We have previously isolated the human CYP19 gene. Two unique untranslated first exons (exons 1.1 and I.2) have been identified in mRNA specific for P450AROM in human placenta. Although the proportion of transcripts encoding exon I.2 is very small, genomic clones encoding the sequences of both exons 1.1 and I.2 have recently been isolated. The corpus luteum of human ovary differs in that promoters 1.1 and 1.2 are completely inactive. Sequence analysis of the DNA immediately 5' of exon II (which contains the start site of translation) demonstrates the presence of a TATAA sequence beginning 149 basepairs 5' of the ATG initiation codon identified in placental exon II. Using a combination of primer extension and S1 nuclease protection analysis, it appears that the initiation site of ovarian P450 ARO M transcripts aligns 26 basepairs down-stream of the sequence TATAA. It appears, therefore, that the expression of P450AROM-specific mRNA in corpus luteum is regulated by an additional promoter (promoter II), which is located just 5' of exon II. Consistent with these observations, Northern analysis of poly(A)+ RNA isolated from placenta and corpus luteum demonstrates that the major promoter of placental P450AROM is promoter 1.1, while the major promoter in the corpus luteum is promoter II. Analysis of the tissue-specific utilization of these promoters was accomplished by means of the polymerase chain reaction to amplify specific 5'-termini from mRNA templates. In addition to placenta,
INTRODUCTION
The formation of estrogens from C19 steroids is catalyzed by an enzyme complex termed aromatase, which is present in the endoplasmic reticulum of cells in which it is expressed. This enzyme complex is comprised of a specific form of cytochrome P450, aromatase cytochrome P450 (P450 ARO M; P450XIX; the product of the CYP19 gene), and a flavoprotein, NADPH-cytochrome P450 reductase, which is a ubiquitous component of most cells (1-5). Aromatase activity is present in a number of human tissues and cell types, including syncytiotrophoblast of placenta (6), hydatid moles (7), JEG-3 cells (a choriocarcinoma-derived cell line) (8), fetal hepatocytes (9), adipocytes (10), and ovarian granulosa cells (11). Additionally, the activity is present in Sertoli (12) and Leydig cells (13, 14) in the male and several sites in the brain of both sexes in the rat, including the hypothalamus and amygdyla (15,16). In the human, the principal estrogen formed differs in different tissues. For example, in placenta, the major estrogen product is estriol; in adipose tissue, estrone; and in ovarian granulosa cells, 17/3-estradiol. Studies using a full-length cDNA encoding human P450 ARO M expressed in COS-1 cells indicate that a single polypeptide is capable of catalyzing the aromatization of all three classes of C19 substrates, namely 16«-hydroxylated androgens, androstenedione, and testosterone. These results suggest that the tissue-specific differences in estrogen formation reflect the presentation of these different Ci 9 substrates to the same enzyme,
0888-8809/91 /2005-2013$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
2005
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
Vol5No. 12
MOL ENDO-1991 2006
rather than the existence of different forms of P450 ARO M (17). Previous studies have demonstrated that aromatase activity in human adipose and ovarian granulosa cells is subject to complex multifactorial regulation and that changes in activity are correlated with changes in the levels of mRNA encoding P450AROM (18-21). To study the molecular mechanisms involved in the tissue-specific and multifactorial regulation of P450AROM, we have isolated the gene encoding P450AROM using a full-length insert isolated from a primer-extended human placental cDNA library (22). The aromatase P450 gene is unique among steroidogenic cytochrome P450 genes in that the first exon of the placental mRNA is untranslated, with exon II encoding the ATG from which translation is initiated (22-24). Analysis of genomic clones suggests that the nine coding exons of the P450 ARO M gene span 36 kilobases (kb). A genomic clone encoding the first exon (exon 1.1) has been identified which resides at least 35 kb 5' of exon II. The gene encoding P450 ARO M is, therefore, at least 72 kb in length. Northern and primer extension analyses of placental mRNA suggest that the promoter up-stream of exon 1.1 is the major promoter in placenta (21). A second unique cDNA insert encoding P450 ARO M mRNA was isolated from the primer-extended placental library that differed in its 5'-untranslated region from the cDNA that encodes the sequences for exon 1.1. The two species of cDNAs are identical down-stream from the sequence identified as the splice junction separating exon 1.1 from exon II, suggesting that the unique 5' leader sequence of the second form of placental P450AROM mRNA is encoded by an alternative untranslated first exon which is presumably controlled by a unique promoter (tentatively termed promoter 1.2). A genomic clone encoding the sequence of exon 1.2 has recently been isolated; mRNAs containing sequence encoded by exon 1.2, however, account for only a minor portion of the transcripts present in placenta (Kilgore, M. W., et al., unpublished observations). Transcripts containing exon 1.1 are the more abundant form of P450 ARO M encoded in placental mRNA (21). A schematic map of the orientation of the two differentially expressed first exons with respect to exon II is shown in Fig. 1. In addition to these alternatively spliced first
HBR
V
-H-fif
AATAAA ATTAAA
VIVIIVIII IX X
Fig. 1. Schematic Representation of the Human P450AROM Gene The closed bars represent translated sequences. The septum in the open bar in exon II represents the splice junction for exons 1.1 and I.2; sequences to the left of the septum would be present in mature RNA only when a putative TATA box 149 bp 5' of the ATG is used to promote transcription.
exons and their respective promoters, a TATAA sequence has been identified that begins 149 bp 5' of the ATG initiation codon identified in placental exon II (2224), suggesting the presence of an additional promoter (promoter II). The objective of the present investigation was to identify the promoter regulating the transcription of P450AROM mRNA in the human ovary. Additionally, tissue-specific utilization of the promoters of untranslated exons 1.1 and 1.2 as well as the putative promoter element identified 5' of the ATG initiation codon in exon II was evaluated by seeking to detect the unique sequences they encode in RNA extracted from tissues expressing aromatase activity. The results suggest that the tissue-specific regulation of the human P450 ARO M gene is in part the consequence of the utilization of tissue-specific promoters.
RESULTS
The 5'-flanking sequences up-stream of promoter II were analyzed to identify potential c/s-regulatory sequences of ovarian P450 ARO M expression. Sequences similar to the consensus sequences for binding of known transcriptional activators are indicated by boxes (Fig. 2), positions of divergence from the consensus sequences are marked by an asterisk. In addition to the TATA sequence previously described, a potential CAAT box was identified beginning at - 6 6 . Two half-sites for glucocorticoid receptor binding were identified at -855 and - 6 7 1 . Two potential cAMP response elementbinding protein-binding sites were identified beginning at -522 and -292. A consensus AP-1 site was found beginning at -498, and a sequence with similarity at all but one position was present beginning at -932. Two sites similar to the consensus sequence for NF-1 were found beginning at -700 and -553, and an SP-1 site at -583 was also indentified. A sequence identical to the consensus sequence reported by Kerr et al. (25) for conferring negative regulation by transforming growth factor-/? was identified beginning at -469. An additional motif has been identified at -132. This sequence, CCAAGGTCA (indicated by underlining), matches exactly the motif defined by Honda et al. (26) and Bogerd et al. (27), which has been found to be an important regulator of the genes of several steroidogenic enzymes. A schematic representation of the oligonucleotides and probes used in this study is shown in Fig. 3. Northern analysis of total placental RNA demonstrated that a probe specific for exon 1.1 hybridizes strongly to transcripts of 3.4 and 2.9 kb, which is consistent with the expected size for P450AROM-specific RNA (21). To determine whether exon 1.1-containing transcripts are present in corpus luteum, Northern analysis was conducted on poly(A)+ RNA isolated from placenta and corpora lutea. Whereas exon 1.1 is readily detectable in poly(A)+ RNA from placenta, mRNA transcripts containing this sequence were undetectable
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
2007
Tissue-Specific Promoters of Human P450AROM
-950
ACTGAAATGC ATTAATGWTG ACTCA|CTCTT CCTCACTCTA CAAGTTGTCA
-900
GRE ACCTACACCT CTTCAGCTAC AGACTACCTA CCATCCCTGA AACTqrGTTC
-850
ifcAGAGTAAA
TCTTAAAAAA CACAGCAGAA CCAGCACATC AGACTGTAAA TTGATTGTCT
-750
TGCACAGGAT GTTAGCTGCT CTTCGAATGA GGTTCCTGAG TGGCACCTGA GRE NF-1
-650 -600 -550 -500 -450 -400 -350 -300
GCC^ATJTGCT
Primer extension
GTTCTICTCTT
TCTTATAATT
-200
CTACCCACTC
-150
AGGAACCTGA
-100
GATCTTTCTT
-50
CCTAAATGTC
+1
GCACCCTCTG
+ 51
TAATTTTTCT AAGCAACAGG
AAGTCAKAAA TTTGGCTTTC TGATAAGGTT TGGCAAGAAA AGGTCAGAAA AAGGGCAAGA GTTTTGACTT GACTCTACCA ATAAflJACAGT GGGCTTCCTT AGCTATAGAT TGATCACATJF
y
GCAATTCTTT TTCTTCCTCT
S1 protection
TAATCTGECC GGATACTGTT
CRE AGATGGCCTG AAGTCAbTAG TGATGGAAGG CTCTGAGAAG ACTGTAGAAA CTTACTATTT TGF-B AGTTGGGA CJATTGCCTTT
Fig. 3. Schematic Representation of the Various Oligonucleotides Used
AGAAGATTGC CTAAACAAAA CTCTGGAATG AGCTTTATTT
Promoter
AATTGGGAAT GCACGTCACT
II
CTATCAGACC AAGCGTCTAA TGCTGCAATT CAAGCCAAAA GTA^CCATKA ATTAGTCTTG AAGTGAATCT GTACTGTACA GAACCTTTTA GGGGATTCTG ACTCTAAATT GCCCCCTCTG
TTTCCACAGG GTTTTGGAAA TGCTGAACCC GATACATTAT oIleHisTyr ValLeuGluM etLeuAsnPr
+ 151
AGGTCAAGGA GTCCCTTTGA ACACAAGATG Met AACATCACCA GCATCGTGCC AsnlleThrS erlleValPr
+201
CCTGCTCCTC ACTGGCCTTT TTCTCTTGGT GTGGAATTAT GAGGGCACAT lLeuLouLeu ThrGlyLeuP heLeuLeuVa lTrpAsnTyr GluGlyThrS
+251
CCTCAATACC AGgtaagtca gtcatttatt tctgtatcta aggagattat erSorllePr OG
+ 301
Northern probes
TCTTCCTCCT
GGTGGCATCC TATTCTGCCJF
TCCCCATTCC TTTCATTCTC TTCTCCCTTA r-Sl TTTTCCACAC CGGlTCCCTAG TACCGTTpGJ JATJGGTACAA ATCTTCAiffG~ GAGATTTTAG AGJTGAGTCAb ATAGACAAAC TTTGAATTCA GRE ACCTCAACGA ATdTGTTCitr TGCCCAAGAA TGATCAAAAA GTCAAAAAGG AGTCATTTTG TTGTTTTGAA CTTCAAGGGA ATTGATTTGG CCTGCTG, ATGACTCCAC £EE
-250
+ 101
PCR reactions
GGGATTACAA AACCTGGCTG AAAAGACAGA TTCAATGGCA
-800
-700
Procedures 52,50^
TGAAGCCATG CCTGCTGCCA CCATGCCAGT oGluAlaMet ProAlaAlaT hrMetProVa
ttacttggga ttttggtcca tcatggtaaa gaaaaatttt gcaaaaagga
C +400
tgaggacttg gagaaatttc agaccaatta tctg
Fig. 2. Sequence of the 5'-Flanking DNA Up-Stream of Exon II Corresponding to Promoter II Sequences with similarity to the consensus sequences for binding of known transcriptional activators are boxed, as are the putative TATA and CAAT boxes. Nucleotides that diverge from the consensus sequences are indicated with an asterisk. The underlined sequence has been found to be important in the regulation of several steroidogenic genes (26, 27), notably 11 i8-hydroxylase. The arrows indicate the positions of the placental intron/exon II boundary and the boundary between exon II and the first intron in the coding region. The bases are numbered such that +1 represents the start of transcription using promoter II, as defined by the primer extension and S1 nuclease protection experiments described in this manuscript.
using poly(A)+ RNA isolated from corpora lutea (Fig. 4). By contrast, when a radiolabeled probe extending upstream from the 5'-splice junction of placental exon II and through the TATA sequence identified at - 2 6 (Figs. 2 and 3) was used as a hybridization probe, a strong hybridization signal was evident in mRNA from the corpus luteum, but not from placenta (Fig. 4). S1 nuclease protection analysis of poly(A)+ RNA isolated from placenta, fetal liver, and corpus luteum was performed to determine whether the TATA sequence identified at - 2 6 is involved in promotion of the P450AROM gene in the corpus luteum. A 594-basepair (bp) probe was synthesized (Fig. 3), encompassing
P
C
P
Fig. 4. Northern Analysis of Poly(A)+ RNA Isolated from Corpus Luteum (C; 10 ^g) and Placenta (P; 10 ^g) PCR-generated probes specific for exon 1.1 or for the 5'untranslated region specific for putative promoter II (up-stream of the placental intron/exon II splice junction) were employed.
exon II and extending 5' of the putative promoter II transcription start site (estimated to be 26 bp downstream of the potential TATA sequence). When poly(A)+ RNA from fetal liver or placenta was used, the protected fragment corresponded in size (183 bp) to exon II (Fig. 5A), indicating a splice boundary between exon II and exons 1.1 or 1.2. By contrast, when RNA from corpus luteum was employed, the major protected fragment observed (262 bp) encompassed exon II, extending to a region 26 bp 3' of the TATA sequence, which is consistent with the use in this tissue of putative promoter II. Extensive overexposure of the film suggested that a promoter ll-specific mRNA transcript was not present in either fetal liver or placenta. These data are consistent with the utilization of putative promoter II in corpus luteum, whereas in fetal liver and placenta, promoters up-stream of exons 1.1 and 1.2 are being used. In addition to the major fragment discussed above, when mRNA isolated from corpus luteum was analyzed by S1 nuclease protection analysis, a minor protected fragment was observed which aligned with the placental exon II splice boundary (Fig. 5A). The
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 August 2015. at 19:43 For personal use only. No other uses without permission. . All rights reserved.
Vol5No. 12
MOL ENDO-1991 2008
TATA Probe
TATA
