help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2003-0342
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
18/5/1144    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow NURSA Molecule Pages Link
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flück, C. E.
Right arrow Articles by Miller, W. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Flück, C. E.
Right arrow Articles by Miller, W. L.
Molecular Endocrinology 18 (5): 1144-1157
Copyright © 2004 by The Endocrine Society

GATA-4 and GATA-6 Modulate Tissue-Specific Transcription of the Human Gene for P450c17 by Direct Interaction with Sp1

Christa E. Flück and Walter L. Miller

Department of Pediatrics and The Metabolic Research Unit, University of California San Francisco, San Francisco, California 94143-0978

Address all correspondence and requests for reprints to: Prof. Walter L. Miller, Department of Pediatrics, Building MR IV, Room 207, University of California San Francisco, San Francisco, California 94143-0978. E-mail: wlmlab{at}itsa.ucsf.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Cytochrome P450c17 catalyzes steroidogenic 17{alpha}-hydroxylase and 17,20 lyase activities. Expression of the gene for P450c17 is cAMP dependent, tissue specific, developmentally programmed, and varies among species. Binding of Sp1, Sp3, and NF1-C (nuclear factor 1-C) to the first 227 bp of 5'flanking DNA (–227/LUC) is crucial for basal transcription in human NCI-H295A adrenal cells. Human placental JEG-3 cells contain Sp1, Sp3, and NF1, but do not express –227/LUC, even when transfected with a vector expressing steroidogenic factor 1 (SF-1). Therefore, other factors are essential for basal expression of P450c17. Deoxyribonuclease I footprinting and EMSAs identified a GATA consensus site at –64/–58 and an SF-1 site at –58/–50. RT-PCR identified GATA-4, GATA-6, and SF-1 in NCI-H295A cells and GATA-2 and GATA-3, but not GATA-4, GATA-6, or SF-1 in JEG-3 cells. Cotransfection of either GATA-4 or GATA-6 without SF-1 activated –227/LUC in JEG-3 cells, but cotransfection of GATA-2 or GATA-3 with or without SF-1 did not. Surprisingly, mutation of the GATA binding site in –227/LUC increased GATA-4 or GATA-6 induced activity, whereas mutation of the Sp1/Sp3 site decreased it. Furthermore, promoter constructs including the GATA site, but excluding the Sp1/Sp3 site at –196/–188, were not activated by GATA-4 or GATA-6, suggesting an interaction between Sp1/Sp3 and GATA-4 or GATA-6. Glutathione-S-transferase pull-down experiments and coimmunoprecipitation demonstrated interaction between GATA-4 or GATA-6 and Sp1, but not Sp3. Chromatin immunoprecipitation assays confirmed that this GATA-4/6 interaction with Sp1 occurred at the Sp site in the P450c17 promoter in NCI-H295A cells. Demethylation with 5-aza-2-deoxycytidine permitted JEG-3 cells to express endogenous P450c17, SF-1, GATA-4, GATA-6, and transfected –227/LUC. Thus, GATA-4 or GATA-6 and Sp1 together regulate expression of P450c17 in adrenal NCI-H295A cells and methylation of P450c17, GATA-4 and GATA-6 silence the expression of P450c17 in placental JEG-3 cells.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
IN HUMAN STEROIDOGENIC tissues, P450c17 is the qualitative regulator, determining the class of steroid produced by both the transcriptional and posttranscriptional mechanisms. The gene encoding P450c17 (formally termed CYP17) is not expressed in the placenta or ovarian granulosa cells, directing steroidogenesis to progesterone (1), and is not expressed in the adrenal zona glomerulosa, directing steroidogenesis to mineralocorticoids (2). By contrast, expression of P450c17 in the adrenal zona fasciculata permits the production of C-21 17-hydroxy precursors of cortisol, and its expression in the adrenal zona reticularis, ovarian theca, and testicular Leydig cells permits the production of C-19 precursors of sex steroids. The 17{alpha}-hydroxylase and 17, 20 lyase activities of P450c17 are differentially regulated at the posttranslational level by the relative abundance of the obligate electron-donor P450 oxidoreductase (3, 4), by the allosteric action of cytochrome b5 (5), and by the phosphorylation of P450c17 itself (6, 7).

Expression of the single human gene for P450c17 on chromosome 10q24.3 (8, 9, 10, 11) is regulated developmentally, hormonally, and in a tissue-specific fashion (1, 12, 13, 14, 15). The human and rat genes for P450c17 share 76% nucleotide sequence identity in their exons (16), but only 43% in the first 1560 bp of 5'-untranslated region. Human P450c17 is expressed in the adrenals, but not in the placenta, whereas rodent P450c17 is not expressed in the adrenals but is expressed in the placenta (17, 18, 19). Consistent with these differences, the identity of the trans-acting factors that regulate P450c17 gene expression, and the locations of their binding sites, differ substantially in the bovine, rat, and human genes (15).

Previous studies of human P450c17 gene expression in human adrenal NCI-H295A cells have shown that most of the transcriptional regulatory elements lie in the first 227 bp of 5'-flanking DNA (13, 14, 15). This 227-bp contains the TATA box, several potential steroidogenic factor 1 (SF-1) sites, a GATA site, two nuclear factor 1 (NF1) sites, and an Sp1/Sp3 site; mutation of the two NF1 sites plus the Sp1/Sp3 site eliminates almost all detectable transcription, and other studies indicated that these NF1 sites bind the NF1-C splice variants CCAAT-binding transcription factor (CTF) 2 and CTF5 (15). Studies in the closely related (and probably equivalent) NCI-H295R cell line also showed that ACTH/cAMP-dependent transcriptional regulation, which works through the SF-1 site (20), is modified by phosphorylation of SF-1 by the MKP-1/ERK1/2 system (21, 22, 23, 24).

However, despite these recent advances in identifying the transcription factors that regulate human P450c17 gene transcription, the basis of its tissue-specific expression remains wholly unknown. To address this question, we compared the regulation of human P450c17 promoter/reporter constructs in steroidogenic human adrenal NCI-H295A cells, which endogenously express the P450c17 gene, and in steroidogenic human placental JEG-3 cells, which do not express P450c17.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
SF-1 Is Insufficient for Expression of P450c17 in Human Placental JEG-3 Cells
SF-1 is required for the transcription of apparently all the genes for steroidogenic enzymes in the adrenal and gonad but does not participate in placental expression of the cholesterol side chain cleavage enzyme, P450scc (25, 26, 27). The basal –227 bp P450c17 promoter has several potential SF-1 binding sites, and the one at –58/–50 clearly participates in basal and cAMP-induced expression in NCI-H295 cells (21, 22, 23, 24). RT-PCR amplification of SF-1 cDNA readily detected SF-1 mRNA in human adrenal NCI-H295A cells, but not in human placental JEG-3 cells (Fig. 1AGo). Southern blotting of the PCR products, which increases the sensitivity of detection substantially, followed by phosphorimager analysis showed that JEG-3 cells contain about 0.4% of the amount of SF-1 found in NCI-H295A cells (Fig. 1BGo). Transfecting JEG-3 cells with a vector expressing SF-1 was not able to transactivate the inactive –227/LUC promoter/reporter construct (Fig. 1CGo); by contrast, overexpression of SF-1 in NCI-H295A cells increased the substantial basal level of –227/LUC activity 17-fold (Fig. 1DGo). Thus, although SF-1 dramatically increases P450c17 expression in NCI-H295A cells, the absence of SF-1 in JEG-3 cells is insufficient to account for their lack of P450c17 gene transcription.



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 1. JEG-3 Cells Do Not Express SF-1, and SF-1 Cannot Activate the –227/LUC Construct in JEG-3 Cells

A, RT-PCR (30 cycles) of RNA from JEG-3 and NCI-H295A cells showed the presence of GAPDH mRNA in both cell types but SF-1 was only detected in NCI-H295A cells. B, Southern blot of PCR products probed with [32P]SF-1 cDNA confirms the presence of abundant SF-1 in NCI-H295A cells, but only a trace amount in JEG-3 cells. C, Expression of the –227/LUC construct in JEG-3 cells minus or plus transfection of an expression vector for SF-1. The –227/LUC construct is inactive in JEG-3 cells and cannot be transactivated by SF-1. D, Expression of the –227/LUC construct in NCI-H295A cells. There is abundant basal expression which is increased 17-fold by transfection of the expression vector for SF-1. In panels C and D, data are the means ± SEM of three experiments, each performed in triplicate. *, P <= 0.05 compared with the vector control; **, P <= 0.05 compared with –227/LUC without SF-1.

 
Effects of NF1 Proteins on the P450c17 Promoter
The NF1 family of transcription factors consists of four members, NF1-A, NF1-B, NF1-C, and NF1-X, encoded by four distinct genes, each of which can be differentially spliced, yielding a large number of distinct proteins (28, 29). The seven identified splice variants of NF1-C are called CTF1 to CTF7 (30). NF1 factors may repress or activate the transcription of many genes (30). The basal promoter of the P450c17 gene contains two NF1 binding sites which contribute about 50% of the activity seen when a –227-bp promoter/reporter construct (–227/LUC) is transiently expressed in NCI-H295A cells (15). The principal forms of NF1 found in adrenal NCI-H295A cells are CTF2 and CTF5 (15). RT-PCR analysis of the NF1 isoforms found in placental JEG-3 cells showed the presence of CTF1, 2, 3, and 5 (Fig. 2AGo). Because JEG-3 cells contain the same NF1 factors found in NCI-H295A cells, it is unlikely that the absence of P450c17 expression in JEG-3 cells is due to the lack of a needed NF1 transcription factor. Consistent with this, cotransfection of vectors that overexpress CTF1, 2, or 5 did not activate transcription of the –227/LUC construct in JEG-3 cells (Fig. 2BGo), and cotransfection of the vector for SF-1 with the CTF’s elicited no significant activity from –227/LUC (Fig. 2BGo). Similarly, –227/LUC was inactive in Drosophila S2 cells, which lack NF1 factors (30) and Sp factors (31), and cotransfection of S2 cells with CTF1 failed to activate –227/LUC (Fig. 2CGo). However, cotransfection of S2 cells with a vector expressing Sp1 activated expression of –227/LUC dramatically, and cotransfection of Sp1 and Sp3 increased activity further, but adding vectors expressing CTF1 and/or SF-1 did not increase expression from –227/LUC further (Fig. 2CGo). Thus, Sp1 is essential for basal P450c17 gene transcription, but NF1 factors are not.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Sp1, not NF1-C, Activates the –227/LUC Construct

A, RT-PCR for NF1-C/CTFs from JEG-3 and NCI-H295A cells. The identities of the CTF factors found in each cell type were inferred by comparing the sizes of the PCR products to PCR products amplified from plasmids expressing known variants and by direct sequencing of the PCR products. B, Expression of the –227/LUC construct in JEG-3 cells plus or minus vectors for CTFs 1, 2, and 5. Neither factor was able to activate the reporter construct (P > 0.05). C, Expression of the –227/LUC construct in Drosophila S2 cells plus or minus cotransfection with vectors for SF-1, Sp1, Sp3, and/or CTF1. Each bar represents the mean ± SEM of at least three independent transfections, each performed in duplicate or triplicate. *, P <= 0.05 compared with empty vector.

 
GATA Factors Are Required for P450c17 Gene Transcription
Because the 227-bp basal P450c17 promoter contains an apparent GATA binding site (AGATAA) at –64 /–59, GATA factors might participate in the tissue-specific transcription of P450c17. The GATA family of transcription factors consists of six members. GATA-1, –2 and –3 are mainly expressed in hematopoietic stem cells where they regulate differentiation-specific gene expression (32); GATA-4, –5, and –6 are expressed in tissues derived from mesoderm and endoderm, including heart, liver, lung, gonad, and gut where they participate in regulating tissue-specific gene expression, either alone or in concert with other factors (33). RT-PCR analysis showed that NCI-H295A cells express GATA-4 and GATA-6, but not GATA-2 and GATA-3, and that JEG-3 cells express GATA-2 and GATA-3, but not GATA-4 and GATA-6 (Fig. 3AGo). Neither GATA-1 nor GATA-5 was found in either NCI-H295A cells or JEG-3 cells (data not shown). Cotransfecting JEG-3 cells with GATA-2 or GATA-3 had no effect on –227/LUC, but cotransfecting with GATA-4 or GATA-6 activated the –227/LUC construct substantially (Fig. 3BGo). This activation was dose dependent up to 300 ng plasmid DNA/well, but further increase in GATA decreased reporter activity (Fig. 3CGo). Thus, GATA-4 and/or GATA-6 appear to be essential for tissue-specific transcription of P450c17.



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 3. GATA-4 or GATA-6 Are Required to Activate the –227/LUC Promoter Reporter Construct in JEG-3 Cells

A, RT-PCR for GATA factors from total RNA extracted from NCI-H295A cells and JEG-3 cells. GAPDH amplification served as an internal control. B, Transfection of JEG-3 cells with the –227/LUC construct and vectors expressing GATA factors. GATA-4 and GATA-6 transactivate the –227/LUC construct significantly. C, Dose-response curve of the transactivation of the –227/LUC construct by GATA-4 (0–1000 ng GATA-4 plasmid DNA per transfection well). RT-PCR for GATA-4 compared with GAPDH served as control that the transfected GATA-4 was expressed in a dose-dependent manner. Data are the means ± SEM of three experiments, each performed in triplicate. Asterisks designate P <= 0.05 compared with the vector control.

 
GATA Factors Do Not Act through the GATA Binding Site in the P450c17 Promoter
The ability of GATA-4 and GATA-6 to transactivate the –227/LUC construct (Fig. 3Go), and the presence of an apparent GATA binding site at –64/–59 in the proximal promoter, suggested that GATA factors acted through this site. To test the ability of GATA factors to bind to this site, we performed EMSAs using oligonucleotides encompassing bases –75/–49. Nuclear extracts from untransfected NCI-H295A and JEG-3 cells, and from JEG-3 cells transfected with an expression vector for GATA-4 formed three complexes, of which the two faster-migrating bands were markedly less prominent in nontransfected JEG-3 cells; all three complexes could be competed to varying degrees by cold probe (Fig. 4AGo). The wild-type –75/–49 probe but not the mut1 or mut2 mutants nor a mutant of the GATA consensus sequence formed three bands of equivalent migration and appearance with nuclear proteins from JEG-3 cells transfected with GATA-4 (Fig. 4BGo). The complexes formed by nuclear extract from transfected JEG-3 cells could be competed by the wild-type P450c17 GATA site and consensus oligonucleotides but not by the mutant P450c17 GATA site (mut1) or the mutant GATA consensus oligo (Fig. 4CGo). Antiserum to GATA-4, but not antisera to GATA-2, –3, or –6 supershifted or competed all three complexes from JEG-3 cells transfected with the GATA-4 expression vector (Fig. 4DGo). The EMSA pattern formed with the –75/–49 oligonucleotide containing the putative GATA-binding site was distinct from the pattern formed by the –219/–182 oligonucleotide including the Sp1/Sp3 site (Fig. 4EGo). The complex formed by this site is specific for Sp1/Sp3 (15) and could not be competed with wild-type GATA or GATA-consensus oligonucleotides (Fig. 4EGo).



View larger version (72K):
[in this window]
[in a new window]
  		Fig. 4. Electrophoretic Mobility Band-Shift Assays

A, Nuclear extracts from untransfected NCI-H295A and JEG-3 cells and from JEG-3 cells transfected with a GATA-4 expression vector (JEG/G4) were incubated with double-stranded radiolabeled –75/–49 probe in the absence or presence of 100-, 300-, or 1000-fold molar excess of unlabeled probe as a competitor. B, The three complexes are formed by JEG/G4 proteins with wild-type –75/–49, but not with mutant oligonucleotides. –, Absence; +, presence of 200-fold molar excess of competitor. C, The three complexes formed by JEG/G4 proteins and the –75/–49 oligo could be competed by 200-fold molar excess of wild-type and consensus GATA, but not mutant wild-type or mutant consensus GATA. D, The three complexes formed by JEG/G4 proteins and the –75/–49 oligo can be supershifted and competed by adding antibody to GATA-4. E, The EMSA pattern formed by the –219/–182 oligo forming the Sp1/Sp3 complex is distinct from the pattern formed by –75/–49. GATA does not participate in the Sp1/Sp3-DNA binding complex at –196/–188. *, GATA complex; **, Sp complex; n.s., nonspecific.

 
To determine whether the GATA binding site at –64/–59 is required for P450c17 gene expression, we generated the same mutations found in the –75/–49 mut1 oligo in the –227/LUC construct. The empty LUC vector, the –227/LUC vector and the mutant –227/LUC vector were inactive in control JEG-3 cells, but cotransfection of the cells with a GATA-4 or GATA-6 expression vectors increased the expression of –227/LUC about 13-fold or 7-fold, respectively (Fig. 5AGo). Surprisingly, mutation of the GATA site in –227/LUC did not abolish this GATA-induced transactivation but instead increased it approximately 3-fold more. Cotransfection of vectors expressing GATA-4 or GATA-6 and SF-1 did not affect –227/LUC significantly but ablated the increase seen with mutant –227/LUC (Fig. 5AGo). The site of action of GATA-4 or GATA-6 on –227/LUC was investigated with deletional mutants; –206/LUC responded to GATA-4 or GATA-6 indistinguishably from –227/LUC, but –184/LUC and all shorter constructs were unresponsive, even though all of these constructs retain the GATA binding site at –64/–59 (Fig. 5BGo). Thus, it appeared that the action of GATA-4 or GATA-6 to induce transcription of the –227/LUC construct was not mediated through the GATA site at –64/–59 but instead required the segment between –184 and –206, which includes the Sp1/Sp3 site at –196/–188 (15). To confirm that the action of GATA-4 or GATA-6 was not mediated via the GATA binding site at –64/–59, we fused the –75/–49 oligonucleotide to the TK32 minimal promoter/luciferase construct. Neither the wild-type nor the mutant –75/–49 oligo fused to TK32/LUC was able to respond to cotransfection of GATA-4 or GATA-6 in JEG-3 cells (Fig. 5CGo). Thus, we considered whether the action of GATA-4 or GATA-6 to induce P450c17 gene transcription was mediated by a protein/protein interaction with Sp1/Sp3 rather than by action mediated by the cis-acting GATA response element.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 5. GATA-4 and GATA-6 Do Not Function through the GATA-Binding Site at –64/–59

A, JEG-3 cells were transfected with the –227/LUC construct containing a normal or mutant GATA-binding site at –64/–59, and cotransfected with and without expression vectors for GATA-4 or GATA-6. Surprisingly, the construct containing the mutated GATA-site was more active than the wild type. This effect was abolished by cotransfection with a vector expressing SF-1. B, Cotransfection of JEG-3 cells with serially deleted P450c17 promoter constructs and expression vectors for GATA-4 or GATA-6. Only the –206/LUC and –227/LUC constructs showed activity in response to GATA-4 or GATA-6. C, Transfection of JEG-3 cells with expression vectors for GATA-4 or GATA-6 and the wild type, or mutated GATA-response element at –64/–59 fused to TK32/LUC. Neither GATA-4 nor GATA-6 could activate TK32/LUC reporter expression via the –64/–59 element. Data are the means ± SEM of three experiments, each performed in triplicate; *, P <= 0.05 compared with empty vector; **, P <= 0.05 compared with wild-type.

 
Interaction of GATA-4 and GATA-6 with Sp1 at the Sp Binding Site
To determine whether the Sp1/Sp3 DNA site was important for the action of GATA-4 or GATA-6, we mutated the Sp1/Sp3 site at –196/–188 in the –227/LUC reporter. When transfected into JEG-3 cells, there was no significant expression of any construct without cotransfection with GATA-4 or GATA-6. Expression of GATA-4 or –6 induced wild-type –227/LUC 13- or 7-fold, but transactivation of the mutant was reduced to less than 23% of that seen with wild-type –227/LUC (Fig. 6AGo). Similarly, GATA-4 or GATA-6 readily transactivated expression of vectors containing either one or two copies of the wild-type –219/–182 sequence fused to TK32/LUC, but GATA-4 or GATA-6 had no effect when one or two copies of the mutant oligonucleotide were fused to TK32/LUC (Fig. 6BGo). To test for direct protein-protein interaction between GATA-4 or GATA-6 and Sp1 or Sp3, we performed glutathione-S-transferease (GST) pull-down experiments. Previously described vectors for Sp1, Sp3, and a carboxy-terminal deletion of Sp1 that contains the N-terminal located zinc finger domain (34) were expressed in Escherichia coli, and the GST-fusion proteins were purified by affinity chromatography. The fusion proteins were mixed with [35S]labeled GATA-4 or GATA-6 prepared in a linked transcription/translation system. Analysis by sodium dodecyl sulfate (SDS) gel electrophoresis showed that GATA-4 and GATA-6 were both able to interact with Sp1 and with the C-terminally deleted Sp1, but neither interacted with Sp3 (Fig. 6Go, C and D). To confirm that the complex of GATA-4 or GATA-6 with Sp1 seen in vitro is present in NCI-H295A cells in vivo, we cross-linked proteins in cell lysates, immunoprecipitated complexes containing GATA proteins with GATA antibodies and harvested the immunocomplexes with protein A Sepharose. Western blotting of the products selected by the GATA antibodies using antiserum to Sp1 showed that each GATA factor coimmunoisolated with Sp1. Thus GATA-4 and GATA-6 appear to transactivate transcription of P450c17 by interacting with Sp1.



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 6. GATA-4 and GATA-6 Interact with Sp1 to Activate the P450c17 Promoter

A, Mutating the Sp site at –196/–188 in the –227/LUC construct greatly diminished the ability of GATA-4 or GATA-6 to transactivate expression. B, Both GATA-4 and GATA-6 can activate the TK32/LUC constructs containing one or two upstream copies of the wild-type –219/–182 Sp1/Sp3 site but did not activate these constructs when bases –196/–194 (CTC) were mutated to AAA. In panels A and B, data are the means ± SEM of three experiments, each performed in triplicate; *, P <= 0.05 compared with the vector control. C and D, GST pull-down experiments with bacterially expressed Sp proteins and radiolabeled GATA protein produced in a cell-free transcription/translation system. GATA-4 and GATA-6 interact with both full-length Sp1 and with a fragment of Sp1 containing only the zinc finger region (Sp1/Zn); GATA-4 and GATA-6 did not interact with Sp3. E, Coimmunoprecipitation showing GATA-4 and GATA-6 interacting with Sp1 in NCI-H295A cells. Cross-linked NCI-H295A cell lysate was immunoadsorbed to anti-GATA-4 or anti-GATA-6 bound to protein-A-Sepharose beads and analyzed on SDS-10% PAGE by Western blotting with antiserum to Sp1. F, ChIP. Both GATA-4 and GATA-6 interact with the –350/–101 DNA segment including the Sp site and with the –169/+45 DNA segment including the GATA and SF-1 sites, but not with the –934/–671 region of the P450c17 promoter in NCI-H295A cells. Both mock immunoprecipitation (no Ab) and immunoprecipitation with IgG served as negative controls.

 
To localize the promoter elements that interact with GATA-4 or GATA-6 in NCI-H295A cells, we performed chromatin immunoprecipitation (ChIP), which can identify DNA elements that interact with transcription factors even if they do not bind directly to the DNA (35). DNA-protein complexes in NCI-H295A cells were cross-linked with formaldehyde, the cells were lysed, DNA was fragmented by sonication, then DNA-protein complexes were immunoprecipitated with antisera against GATA-4 or GATA-6. When DNA from the immunoprecipitated complexes was eluted and analyzed by PCR, we found that GATA-4 and GATA-6 interact with the –350/–101 region of the P450c17 promoter, which includes the Sp binding site at –196/–188 for which we found functional activity (Fig. 6FGo). We also confirmed that GATA-4 and GATA-6 interact with the –169/+45 segment, which includes the functionally inactive GATA site at –64/–59 (Fig. 6FGo). GATA-4 and GATA-6 did not interact with the –934/–671 region of the P450c17 promoter, which was chosen at random as a negative control. Thus GATA-4 and/or GATA-6 form a complex with Sp1 to transactivate transcription of P450c17 in NCI-H295A cells by interacting with the Sp1 site rather than by binding to the apparent GATA binding site.

The Genes for P450c17, GATA-4, and GATA-6 Are Methylated in JEG-3 Cells
To determine whether expressing GATA factors in placental JEG-3 cells would activate the endogenous, silent P450c17 gene, we transfected JEG-3 cells with GATA-4, SF-1, or both, prepared total RNA and performed RT-PCR for P450c17, GATA-4, GATA-6, SF-1, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). P450c17 and GATA-6 were not amplified, but GATA-4 and SF-1 were amplified, indicating that although GATA-4 and SF-1 were expressed in these transiently transfected JEG-3 cells, they were not sufficient to activate the endogenous, silent P450c17 gene (Fig. 7AGo). A possible explanation for this unresponsiveness might be that the P450c17 gene is silenced by methylation in JEG-3 cells. Previous studies with bovine adrenocortical cells, which lose the capacity to express P450c17 in culture, showed that the P450c17 gene undergoes specific changes in methylation when placed in culture (36). Therefore, we endeavored to demethylate the DNA in JEG-3 cells by treating them with 1 µM 5-aza-2-deoxycytidine (5-aza-dC) for 24 h before and 48 h after transfecting them with vectors expressing GATA-4, SF-1 or both. Treatment with 5-aza-dC derepressed expression of P450c17, GATA-4, GATA-6, and SF-1 in control cells transfected with empty vector, but cotransfection with vectors expressing GATA-4, SF-1 or both did not increase P450c17 mRNA further (Fig. 7AGo). Thus, demethylation of JEG-3 cell DNA appeared to derepress the genes for P450c17 and the genes for several factors that regulate P450c17. Because expression of GATA-4 and SF-1 from transiently transfected plasmids did not increase P450c17 mRNA further, it is likely that treatment with 5-aza-dC yielded fully effective concentrations of these factors.



View larger version (49K):
[in this window]
[in a new window]
  		Fig. 7. The Gene for P450c17 Is Methylated in JEG-3 Cells

A, Demethylation experiment using 5-aza-dC. JEG-3 cells were transfected with empty vector, GATA-4, SF-1, and GATA-4 plus SF-1. Cells were treated with (+) or without (–) 1 µM 5-aza-dC for 72 h. Total RNA was prepared and RT-PCR (30 cycles) performed on each sample for P450c17, GATA-4, GATA-6, SF-1, and GAPDH, as indicated on the right. Water (no DNA) served as negative control, and plasmid DNA served as positive control. B, Diagram of the P450c17 promoter with two methylation-sensitive NaeI sites. PCR primers (arrows) were designed to amplify DNA segments containing these two sites. C, Genomic DNA from NCI-H295A and JEG-3 cells (with and without 5-aza-dC pretreatment) was digested with Turbo NaeI for 48 h and segments A, B, AB (as indicated in panel B), and a control 943-bp fragment for the unlinked ACTH receptor (MC2R), which is expressed in both cell lines, were amplified with 30–35 cycles of PCR. D, Transfection of the –227/LUC promoter/reporter construct into JEG-3 cells, which had been pretreated with 1 µM 5-aza-dC for 48 h. Demethylation of the genes in cultured JEG-3 cells resulted in a 4- to 5-fold activation of the previously inactive –227/LUC construct.

 
To determine whether treatment with 5-aza-dC truly demethylated the P450c17 gene, we performed restriction endonuclease digests of genomic DNA from JEG-3 and NCI-H295A cells using the methylation-sensitive enzyme NaeI. NaeI recognizes and cleaves the sequence GCC/GGC when it is unmethylated but does not cleave when the cytosine residue of CpG is methylated. We examined the NaeI sites at –834 and –17 in the P450c17 promoter by cleaving genomic DNA from JEG-3 and NCI-H295A cells with NaeI, then amplifying the digestion product with primers flanking these sites (Fig. 7BGo). No product was seen from the DNA from NCI-H295A cells or JEG-3 cells treated with 5-aza-dC, indicating the NaeI sites were cleaved (Fig. 7CGo). By contrast, the fragments containing the NaeI sites were readily amplified from DNA from untreated JEG-3 cells, indicating there had been little or no digestion with NaeI (Fig. 7CGo). The expected DNA fragments were readily amplified from undigested genomic DNA from both JEG-3 and NCI-H295A cells. Also, long-term digestion with NaeI did not destroy the genomic DNA at random, as a 943-bp fragment of the ACTH receptor (MC2R), which is expressed in both cell lines, could be amplified from both JEG-3 and NCI-H295A genomic DNA (Fig. 7CGo). Thus, the P450c17 promoter appears to be methylated in JEG-3 cells, where P450c17 is not expressed, but unmethylated in NCI-H295A cells, where P450c17 is expressed.

Because transiently transfected human P450c17 promoter/reporter constructs are inactive in JEG-3 cells (Fig. 1CGo), and treatment with 5-aza-dC activates endogenous P450c17 gene transcription (Fig. 7AGo), we sought to determine whether the demethylation of genes encoding transcription factors such as GATA-4 and GATA-6 was sufficient to permit activation of a transiently transfected P450c17 promoter. Therefore, we treated JEG-3 cells with 5-aza-dC for 48 h to derepress methylated genes, then transiently transfected the cells with the –227/LUC construct. Compared with controls, treatment with 5-aza-dC resulted in a 4- to 5-fold activation of the –227/LUC construct (Fig. 7DGo). Thus, silencing of P450c17 gene transcription is due to the methylation of genes encoding factors needed for P450c17 gene transcription, as well as being due to methylation of the P450c17 gene itself.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Human P450c17 is expressed in human adrenals and in steroidogenic human adrenal NCI-H295A cells but is not expressed in the placenta or in steroidogenic human placental JEG-3 cells. Potential explanations for placental P450c17 gene silencing include a lack of needed transcriptional activators, the presence of transcriptional repressors or chromatin alteration by methylation or deacetylation. In this study, we show that P450c17 is unmethylated in NCI-H295A cells and methylated in JEG-3 cells, and that the endogenous P450c17 gene is transcribed when JEG-3 cells are treated with the demethylation drug 5-aza-dC. However, transiently transfected P450c17 promoter/reporter constructs, which are not methylated, are inactive in JEG-3 cells (13, 14). GATA-4 and GATA-6 are also not expressed in JEG-3 cells, apparently due to gene methylation, because treating JEG-3 cells with 5-aza-dC resulted in their endogenous expression and transactivation of a transiently transfected P450c17 promoter/reporter construct. Thus, methylation of P450c17 alone is insufficient to explain the lack of P450c17 gene expression in the placenta; silencing of GATA-4 and GATA-6 is also responsible. Our analysis of the transcription factors needed for P450c17 gene transcription showed a different profile of GATA factors expressed in NCI-H295A and JEG-3 cells. GATA factors play an essential role in cell-specific transcription of P450c17 promoter/reporter constructs, but they do not act through their GATA response element; rather, GATA-4 and GATA-6 activate the basal promoter/reporter construct –227/LUC through interaction with Sp1 (Fig. 8Go). These findings are consistent with established characteristics of the GATA family of transcription factors. GATA 4, 5, and 6 are important in programming cell type-specific gene expression, especially in the developing heart (33, 37). GATA factors mediate tissue specificity through cell type-specific interactions with other transcription factors, most of which are themselves expressed in semirestricted patterns. Many proteins that interact with GATA factors have been identified, including DNA binding factors such as Sp1 (34) and general transcriptional activators and repressors (33).



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 8. Diagram of the Proximal Human P450c17 Promoter

Previously described footprinted regions (15 ) are shown as open boxes below the line. Known binding elements are drawn in specific shapes. The functionally active complex of GATA-4/6 with Sp1 binds at –196/–188. GATA-4 and GATA-6 bind to the GATA site at –64/–59 although this complex is functionally inactive. SF-1 binds and activates at –58/–50; two other potential SF-1 binding elements are shown as dotted ovals.

 
The ability of GATA factors to regulate P450c17 promoter/reporter constructs but their inability to activate basal expression through the GATA-response element suggested that the GATA factors were interacting with other transcription factors. SF-1 was an obvious candidate for such a factor. SF-1 is expressed in the human adrenal, but not in the placenta; there are multiple potential SF-1 sites in the P450c17 promoter, of which the site at –58/–50 plays a crucial role in ACTH/cAMP-dependant P450c17 regulation (21, 22, 23, 24) (Fig. 8Go). GATA factors (especially GATA-4) are known partners for SF-1 in the control of SF-1 targeted genes such as Müllerian-inhibiting substance (38), steroidogenic acute regulatory protein (39), aromatase (39), and inhibin (39), all of which are related to sex steroid production. However, activation of the P450c17 promoter in JEG-3 cells did not require SF-1. Furthermore, the P450c17 promoter construct was inactive in Drosophila S2 cells that express both Ftz-F1, the Drosophila homolog of SF-1, and pannier, the Drosophila homolog of GATA-4. By contrast, cotransfecting S2 cells with the P450c17 promoter/reporter construct and with an expression vector for Sp1, which is absent in S2 cells (31), was able to activate the promoter, indicating an important role for Sp1 in P450c17 gene expression (15).

Sp1 can recruit GATA-1, a major erythroid transcription factor to a promoter, even in the absence of GATA binding sites (40); thus, Sp1 was a good candidate for interacting with GATA-4 or GATA-6. Mutating the GATA cis-acting element in the promoter construct increased rather than decreased transactivation by GATA-4 or GATA-6, suggesting that the DNA binding site and Sp1 are in competition for binding GATA-4 or GATA-6. Thus, mutating the DNA binding site may have increased the amount of GATA protein available to interact with Sp1, thereby increasing transcription of the P450c17 promoter construct. Similarly, mutation of the GATA binding site in the intestine-specific sucrase-isomaltase gene did not decrease stimulation of promoter/reporter constructs by GATA-4 interacting with Cdx2 and hepatocyte nuclear factor-1{alpha} (41). In JEG-3 cells, which express GATA-2 and GATA-3 but do not normally express P450c17, GATA-4, or GATA-6, both GATA-4 and GATA-6 could interact with Sp1 to transactivate a P450c17 promoter/reporter construct, although GATA-4 was more potent. This specificity of GATA-4 and GATA-6 is similar to the ability of GATA-4 to transactivate the IL-5 promoter in the presence of other GATA factors (42), and the ability of GATA-4 to transactivate testis-specific expression of the Dmrt1 gene (43).

The human fetal adrenal expresses both GATA-4 and GATA-6, but GATA-4 is expressed only minimally, if at all, in the adult adrenal (44). Expression of GATA-6 persists in the adult adrenal and appears to be confined to the zonae fasciculata and reticularis, where P450c17 is expressed (44). By contrast, human adrenal malignancies express abundant GATA-4 and relatively little GATA-6 (45). NCI-H295A cells are derived from a human adrenocortical malignancy (46), but unlike other human adrenal cell lines, they retain their differentiated steroidogenic phenotype having a pattern of steroid biosynthesis and gene expression that closely resembles the human fetal adrenal (47). Consistent with this, NCI-H295A cells express both GATA-4 and GATA-6. Thus, both GATA-4 and GATA-6 can regulate adrenal P450c17 expression, and a change in the profile of GATA-6 and GATA-4 expression does not affect P450c17 expression qualitatively; however, the stronger activity of GATA-4 may correlate with the tremendous expression of P450c17 in the fetal adrenal, which is required for its abundant production of DHEA, and with increased C19 steroid production in most adrenal malignancies. GATA-4 and GATA-6 also act differently on the human and rodent P450c17 promoter. Activation of rodent P450c17 specifically requires GATA-4, and GATA-6 cannot replace this action of GATA-4 (48). This may contribute to the species-specific difference in P450c17 expression in human but not rodent adrenals.

In the human ovary, P450c17 is expressed in theca cells but not granulosa cells (1). GATA-4 and GATA-6 are expressed in the human ovary (49), and a spatially and temporally regulated expression of GATA-4 and GATA-6 has been described for the porcine ovary (50). Patients with polycystic ovary syndrome (PCOS), a common endocrine disorder affecting approximately 5% of women of reproductive age, typically have hyperandrogenism of both ovarian and adrenal origin (51). Ovarian theca cells from PCOS woman have increased P450c17 expression, and microarray analysis and real-time RT-PCR show increased expression of GATA-6 in these cells (52). Thus, GATA-modulated P450c17 transcription may play important roles in both normal development and in hyperandrogenic disorders such as PCOS.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Plasmids
The P450c17 promoter reporter constructs have been described previously (15); all extend from +63 to the base number indicated in the name of the construct (e.g. –227/LUC), and base numbers are counted from the transcriptional start site. Mutant constructs were generated by PCR-based, site-directed mutagenesis (53) using the primers shown in Table 1Go. For GATA enhancer activity studies, the wild-type and the mutant GATA oligonucleotides were inserted into the SacI site of TK32/LUC as described for the various FP3-TK32/LUC constructs (15). Plasmids pcDNA3-CTF2, -CTF5, and -GATA-6 were cloned from total RNA of NCI-H295H cells. Using the primers listed in Table 1Go, the respective cDNAs were amplified from total RNA by RT-PCR and inserted into the cloning site (EcoRI/XhoI for CTF2 and CTF5, and KpnI/XhoI for GATA-6) of the mammalian expression vector pcDNA3. Plasmids containing the cDNAs for CTF1 (54), Sp1 (34), Sp3 (55), SF-1 (53), GATA-2 (42), GATA-3 (42), or GATA-4 (56) were subcloned into the mammalian (pcDNA3, pCMV5) or insect (pPac) expression vectors as needed.


View this table:
[in this window]
[in a new window]
  		Table 1. Oligonucleotide Sequences

 
Cell Cultures, Transfections, and Dual Luciferase Assays
The adherent NCI-H295A subline (13) of human adrenal NCI-H295 cells (46, 47) was grown and transfected as described (13, 15); human placenta JEG-3 cells (57) were grown and transfected as described (27); and Drosophila melanogaster Schneider S2 cells (58) were cultured as described (59). Cells were divided into 2 cm, six-well plates (Falcon 3046, BD Biosciences, Lincoln Park, NJ) 24 h before transfection at approximately 50% confluence.

For transient transfection, NCI-H295A cells were transferred to DMEM-H16 supplemented with 10% fetal bovine serum, whereas JEG-3 and S2 cells were kept in their usual growth medium. NCI-H295A and S2 cells were incubated overnight, but JEG-3 cells were incubated for only 6 h with calcium phosphate precipitates containing vectors expressing P450c17 promoter-reporter-constructs plus expression vectors for Sp1, Sp3, CTF1, CTF2, CTF5, SF-1, and GATA transcription factors in the desired combinations (total DNA content 2.5 µg/well). Six or eighteen hours after transfection, the calcium phosphate precipitates were washed off with PBS, and the cells were incubated with fresh growth medium for 24–36 h. Cells were then lysed and assayed for luciferase activity using the Dual Luciferase Reporter Assay System (Promega Corp., Madison, WI); cotransfection of 100 ng/well Renilla luciferase reporter vector (pRL-CMV, Promega Corp.) was used as a control for transfection efficiency. Data represent the mean ± SEM of three or more independent experiments, each performed in triplicate. Statistical differences were calculated by t test and significance was defined as P <= 0.05.

For demethylation experiments, freshly plated JEG-3 cells were treated with 1 µM 5-aza-dC for 24 h before transfection with expression vectors for GATA-4, SF-1 or both. The 5-aza-dC was then withheld for 6 h of transfection, and then reintroduced for another 48 h. For demethylation and transfection experiments with the –227/LUC construct, the JEG-3 cells were pretreated with 1 µM 5-aza-dC for 48 h.

RT-PCR
Total RNA was isolated from NCI-H295A and JEG-3 cells using TRI Reagent according to the manufacturer’s protocol (Molecular Research, Cincinnati, OH). Random primers (1 µg) and Superscript II reverse transcriptase (Life Technologies, Rockville, MD) were used to synthesize the first-strand cDNA from 1–2 µg of total RNA in a reaction volume of 20 µl. PCR amplification was then performed using 1–2 µl of the first-strand cDNA product and oligonucleotides as listed in Table 1Go.

Southern Blotting of RT-PCR Products
Total RNA prepared from either NCI-H295A or JEG-3 cells was reverse transcribed and 30 cycles of PCR (denaturing at 95 C for 30 sec, annealing at 60 C for 45 sec and extension at 72 C for 45 sec) was performed using the oligonucleotides listed in Table 1Go to amplify SF-1. A 10-µl aliquot of the 50-µl reaction volume was run on a 1.5% agarose gel. A 20-µl aliquot of the 50-µl reaction volume was digested with BalI, diluted and analyzed by electrophoresis in 2% agarose, transferred to a nylon membrane (Hybond N+; Amersham Pharmacia Biotech, Buckinghamshire, UK), and cross-linked with UV light. After neutralization in 0.5 M Tris-HCl (pH 7.5), 1.5 M NaCl, 5 mM EDTA (pH 8.0), the membrane was prehybridized for 4 h in 6x saline sodium citrate (SSC), 1% SDS, 10x Denhardt’s solution, 100 µl/ml denatured herring DNA at 65 C. Then, an [{alpha}-32P]deoxy-CTP-labeled probe (~50 mCi) was added and the membrane hybridized in this solution overnight at 65 C. Finally, the membrane was washed twice for 20 min in 2x SSC, 0.5% SDS at room temperature, then washed twice for 10 min in 1x SSC, 0.5% SDS at 65 C. Autoradiography was then analyzed by phosphorimager (Molecular Dynamics Storm 860, Sunnyvale, CA).

Electrophoretic Mobility Shift Assays (EMSAs)
Nuclear extracts were prepared from NCI-H295A and JEG-3 cells as described (60). Protein concentrations were determined by the Bradford method (Bio-Rad, Richmond, CA). EMSA reactions typically contained 8–12 µg of nuclear extract or 1 µl of protein produced by in vitro transcription-translation, 10,000 cpm (~0.1 ng) end-labeled double-stranded probe, and 1 µg poly(deoxyinosine-deoxycytosine) as a nonspecific competitor dissolved in 4% glycerol, 1 mM EDTA, 5 mM dithiothreitol (DTT), 10 mM Tris-HCl (pH 7.5), 0.1 mg/ml BSA, and 50 mM KCl. For competitive binding studies, 10–100 ng unlabeled specific and nonspecific oligonucleotides were premixed with the nuclear extract and the reaction buffer for 10 min before adding the probe. Reactions were then incubated 20–30 min at room temperature, electrophoresed through 8% native polyacrylamide gels in 45 mM Tris-borate (pH 8.0) and 1 mM EDTA, and analyzed by phosphorimaging. Antibody supershift assays used rabbit polyclonal antibodies to GATA-2, GATA-3, GATA-4, GATA-6 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), added at the same time as the probe.

GST-Pull-Down Experiments
The constructs pGEX-2TK, GST-Sp1, GST-Sp1/Zn, and GST-Sp3 (34) (a kind gift from J. M. Horowitz, North Carolina State University, Raleigh, NC) were used in the GST-pull-down experiments. Expression in E. coli BL21 (Stratagene, La Jolla, CA) was induced by 0.1 mM isopropthiogalactopyranoside, and bacteria were then lysed in 50 mM Tris-HCl (pH 8.0), 100 mM KCl, 0.1 mM DTT, 10 µM ZnCl2 during three freeze-thaw cycles and sonication. These cell lysates were centrifuged and the supernatants containing the soluble GST fusion proteins (and other soluble proteins) were collected and stored in 20% glycerol at –70 C until needed. To prepare GST alone (pGEX-2TK) or the GST fusion proteins, 1 ml of the prepared protein mixture was incubated with glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) for 2 h at 4 C in NETN [20 mM Tris-HCl (pH 8.0), 100 mM NaCl, 1 mM EDTA, and 0.5% Nonidet P-40 (NP-40)]. Beads were purified by washing three times with NETN, and the amount of fusion protein bound to the beads was estimated by running an aliquot on a SDS-polyacrylamide gel with known standards and staining with Coomassie blue. To test for protein-protein interaction, equal amounts (~5 µg) of GST alone or GST fusion proteins bound to the beads were then incubated for 2 h at 4 C with 20–25 µl 35S-labeled GATA-4 or GATA-6 protein, both produced in vitro using a linked transcription/translation system (TNT T7, Promega). Finally, the beads were washed four times with NETN and heated for 5 min at 95 C to release the bound, labeled proteins, which were then electrophorized on 8% or 10% SDS-polyacrylamide gels and analyzed by PhosphorImager.

Chromatin Immunoprecipitation (ChIP)
NCI-H295A cells were grown to confluence on 15-cm plates and DNA/protein complexes were then cross-linked by treating the cells with 1% formaldehyde for 15 min at room temperature. The cross-linking reaction was stopped by adding 125 mM glycine for 5 min, and the cells were lysed in 1% SDS, 20 mM Tris-HCl (pH 8.1), scraped off the plates, and collected by centrifugation. The pellet was resuspended in 0.25% SDS, 20 mM Tris-HCl (pH 8.1), and 0.5% protease inhibitor cocktail (complete, Roche Molecular Biochemicals, Indianapolis, IN), prepared according to the manufacturer’s recommendations, was added, and after 10 min incubation on ice, the samples were sonicated to obtain DNA fragments of 0.5–1.5 kb. After centrifugation at 1000 x g for 15 min, the supernatant was diluted in modified RIPA buffer [50 mM Tris-HCl (pH 7.4), 1% NP-40, 0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA], and 0.5% protease inhibitor cocktail, 2–3 µg of luciferase plasmid DNA and 1 mg BSA were added. For immunoprecipitation, 1 ml of this mixture was gently agitated overnight at 4 C with 5 µg of antibody to GATA-4, GATA-6, or control rabbit IgG. Protein A agarose beads (Invitrogen Life Technologies, Carlsbad, CA) were diluted to 50% in reaction buffer, and 50 µl added to each of the samples which were then shaken gently for another 2 h. The beads were washed once with 1 ml 0.1% SDS, 1% NP-40, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 150 mM NaCl, then once with the same buffer containing 500 mM NaCl, once with 0.25 M LiCl, 1% NP-40, 1% Na-deoxycholate, 1 mM EDTA, 10 mM Tris-HCl (pH 8.1), and three times with 10 mM Tris-HCl (pH 8.1), 1 mM EDTA. Beads were collected by centrifugation at 1000 x g for 2 min at 4 C after each washing step. Immunocomplexes were eluted from the beads with 1% SDS, 0.1 M NaHCO3, 0.2 M NaCl, 0.1 M Tris-HCl (pH 8.8) at 75 C. Samples were treated with Proteinase K (Roche) at 50 C for 1 h, and cross-links were reversed by heating to 65 C for 6 h. After adding 1 µg of poly(deoxyinosine-deoxycytosine) to each sample, DNA was extracted using QIAquick columns (QIAGEN, Valencia, CA) and 35 cycles of PCR were performed on each template using primers as listed in Table 1Go.

Coimmunoprecipitation and Western Blotting
NCI-H295A cells were grown to confluence on 15-cm plates and lysed by sonication in 5 ml PBS. Proteins were cross-linked using 0.5 mM disuccinimidyl suberate according to the manufacturer’s recommendations (www.piercenet.com; Pierce Biotechnology, Rockford, IL). One microliter of cross-linked cell lysate was then gently shaken with 5 µg of antibody against GATA-4 or GATA-6 at 4 C for 16 h, and 50 µl 50% protein A-Sepharose beads (Amersham Biosciences, Piscataway, NJ) were then added to the mixture for 4 h. The beads were washed four times with PBS, 0.1% Tween 40. The isolated protein complexes were denatured for 5 min at 95 C, and analyzed by SDS, 8% PAGE, transferred to a protein membrane (Immobilon P, Millipore Corp., Billerica, MA) followed by immunoblotting with Sp1 antibody (1:1000) and detected by enhanced chemiluminescence (Amersham Biosciences).


   ACKNOWLEDGMENTS
 
We gratefully acknowledge Dr. Amit V. Pandey for his help throughout the work and Dr. John W. M. Martens for his assistance in the preliminary phases of this work. We thank Drs. T. Yamagata (Tokyo University, Tokyo, Japan) for the GATA-2 and –3 plasmids, E. S. Pollak (Children’s Hospital, Philadelphia, PA) for the GATA-4 plasmid, Y. Dusserre (L’Ecole Polytechnique Federale, Lausanne, Switzerland) for the pPadh-CTF1 plasmid, R. Tjian (Universtiy of California-Berkeley, Berkeley, CA) for the pPac-Sp1 plasmid, G. Suske (Phillipps University, Marburg, Germany) for the pPac-Sp3 plasmid, and J. M. Horowitz (North Carolina State University, Raleigh, NC) for all the GST constructs.


   FOOTNOTES
 
This work was supported by a grant from the Swiss National Science Foundation for Medical-Biological Grants (to C.E.F), by the Foundation Eugenio Litta (to C.E.F), and by National Institutes of Health Grant HD 41958 (to W.L.M.).

Abbreviations: 5-aza-dC, 5-aza-2-deoxycytidine; ChIP, chromatin immunoprecipitation; CTF, CCAAT-binding transcription factor; DTT, dithiothreitol; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GST, glutathione-s-transferase; NF1, nuclear factor 1; NP-40, Nonidet P-40; PCOS, polycystic ovary syndrome; SDS, sodium dodecyl sulfate; SF-1, steroidogenic factor 1; SSC, saline sodium citrate.

Received for publication September 5, 2003. Accepted for publication February 17, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Voutilainen R, Tapanainen J, Chung BC, Matteson KJ, Miller WL 1986 Hormonal regulation of P450scc (20,22-desmolase) and P450c17 (17 {alpha}-hydroxylase/17,20-lyase) in cultured human granulosa cells. J Clin Endocrinol Metab 63:202–207[Abstract]
  2. Sasano H, Mason JI, Sasano N 1989 Immunohistochemical study of cytochrome P-450c17 {alpha} in human adrenocortical disorders. Hum Pathol 20:113–117[CrossRef][Medline]
  3. Yanagibashi K, Hall PF 1986 Role of electron transport in the regulation of the lyase activity of C21 side-chain cleavage P-450 from porcine adrenal and testicular microsomes. J Biol Chem 261:8429–8433[Abstract/Free Full Text]
  4. Lin D, Black SM, Nagahama Y, Miller WL 1993 Steroid 17 {alpha}-hydroxylase and 17,20-lyase activities of P450c17: contributions of serine106 and P450 reductase. Endocrinology 132:2498–2506[Abstract]
  5. Auchus RJ, Lee TC, Miller WL 1998 Cytochrome b5 augments the 17,20-lyase activity of human P450c17 without direct electron transfer. J Biol Chem 273:3158–3165[Abstract/Free Full Text]
  6. Zhang LH, Rodriguez H, Ohno S, Miller WL 1995 Serine phosphorylation of human P450c17 increases 17,20-lyase activity: implications for adrenarche and the polycystic ovary syndrome. Proc Natl Acad Sci USA 92:10619–10623[Abstract/Free Full Text]
  7. Pandey AV, Mellon SH, Miller WL 2003 Protein phosphatase 2A and phosphoprotein SET regulate androgen production by P450c17. J Biol Chem 278:2837–2844[Abstract/Free Full Text]
  8. Matteson KJ, Picado-Leonard J, Chung BC, Mohandas TK, Miller WL 1986 Assignment of the gene for adrenal P450c17 (steroid 17{alpha}-hydroxylase/17,20 lyase) to human chromosome 10. J Clin Endocrinol Metab 63:789–791[Abstract]
  9. Picado-Leonard J, Miller WL 1987 Cloning and sequence of the human gene for P450c17 (steroid 17{alpha}-hydroxylase/17,20 lyase): similarity with the gene for P450c21. DNA 6:439–448[Medline]
  10. Sparkes RS, Klisak I, Miller WL 1991 Regional mapping of genes encoding human steroidogenic enzymes: P450scc to 15q23–q24, adrenodoxin to 11q22; adrenodoxin reductase to 17q24–q25; and P450c17 to 10q24–q25. DNA Cell Biol 10:359–365[Medline]
  11. Fan YS, Sasi R, Lee C, Winter JS, Waterman MR, Lin CC 1992 Localization of the human CYP17 gene (cytochrome P450(17{alpha})) to 10q24.3 by fluorescence in situ hybridization and simultaneous chromosome banding. Genomics 14:1110–1111[CrossRef][Medline]
  12. Voutilainen R, Miller WL 1986 Developmental expression of genes for the stereoidogenic enzymes P450scc (20,22-desmolase), P450c17 (17{alpha}-hydroxylase/17,20-lyase), and P450c21 (21-hydroxylase) in the human fetus. J Clin Endocrinol Metab 63:1145–1150[Abstract]
  13. DiBlasio AM, Voutilainen R, Jaffe RB, Miller WL 1987 Hormonal regulation of mRNA for P450scc (cholesterol side chain cleavage enzyme) and P450c17 (17{alpha}-hydroxylase/17,20 lyase) in cultured human fetal adrenal cells. J Clin Endocrinol Metab 65:170–175[Abstract]
  14. Rodriguez H, Hum DW, Staels B, Miller WL 1997 Transcription of the human genes for cytochrome P450scc and P450c17 is regulated differently in human adrenal NCI-H295 cells than in mouse adrenal Y1 cells. J Clin Endocrinol Metab 82:365–371[Abstract/Free Full Text]
  15. Brentano ST, Picado-Leonard J, Mellon SH, Moore CCD, Miller WL 1990 Tissue-specific, cyclic adenosine 3',5'-monophosphate-induced, and phorbol esterrepressed transcription from the human P450c17 promoter in mouse cells. Mol Endocrinol 4:1972–1979[Abstract]
  16. Lin CJ, Martens JWM, Miller WL 2001 NF-1C, Sp1, and Sp3 are essential for transcription of the human gene for P450c17 (steroid 17{alpha}-hydroxylase/17,20 lyase) in human adrenal NCI-H295A cells. Mol Endocrinol 15:1277–1293[Abstract/Free Full Text]
  17. Givens CR, Zhang P, Bair SR, Mellon SH 1994 Transcriptional regulation of rat cytochrome P450c17 expression in mouse Leydig MA-10 and adrenal Y-1 cells: identification of a single protein that mediates both basal and cAMP-induced activities. DNA Cell Biol 13:1087–1098[Medline]
  18. Johnson DC, Sen M 1990 The cytochrome P450c17 (17{alpha}-hydroxylase/C17,20-lyase) activity of the junctional zone of the rat placenta. J Endocrinol 125:217–224[Abstract]
  19. Yamamoto T, Chapman BM, Johnson DC, Givens CR, Mellon SH, Soares MJ 1996 Cytochrome P450 17{alpha}-hydroxylase gene expression in differentiating rat trophoblast cells. J Endocrinol 150:161–168[Abstract]
  20. Durkee TJ, McLean MP, Hales DB, Payne AH, Waterman MR, Khan I, Gibori G 1992 P450(17{alpha}) and P450scc gene expression and regulation in the rat placenta. Endocrinology 130:1309–1317[Abstract]
  21. Zhang P, Mellon SH 1997 Multiple orphan nuclear receptors converge to regulate rat P450c17 gene transcription: novel mechanisms for orphan nuclear receptor action. Mol Endocrinol 11:891–904[Abstract/Free Full Text]
  22. Sewer MB, Nguyen VQ, Huang CJ, Tucker PW, Kagawa N, Waterman MR 2002 Transcriptional activation of human CYP17 in H295R adrenocortical cells depends on complex formation among p54(nrb)/NonO, proteinassociated splicing factor, and SF-1, a complex that also participates in repression of transcription. Endocrinology 143:1280–1290[Abstract/Free Full Text]
  23. Sewer MB, Waterman MR 2002 Adrenocorticotropin/cyclic adenosine 3',5'-monophosphate-mediated transcription of the human CYP17 gene in the adrenal cortex is dependent on phosphatase activity. Endocrinology 143:1769–1777[Abstract/Free Full Text]
  24. Sewer MB, Waterman MR 2002 cAMP-dependent transcription of steroidogenic genes in the human adrenal cortex requires a dual-specificity phosphatase in addition to protein kinase A. J Mol Endocrinol 29:163–174[Abstract]
  25. Sewer MB, Waterman MR 2003 cAMP-dependent protein kinase enhances CYP17 transcription via MKP-1 activation in H295R human adrenocortical cells. J Biol Chem 278:8106–8111[Abstract/Free Full Text]
  26. Moore CCD, Hum DW, Miller WL 1992 Identification of positive and negative placenta-specific basal elements and a cyclic adenosine 3',5'-monophosphate response element in the human gene for P450scc. Mol Endocrinol 6:2045–2058[Abstract]
  27. Hum DW, Aza-Blanc P, Miller WL 1995 Characterization of placental transcriptional activation of the human gene for P450scc. DNA Cell Biol 14:451–463[Medline]
  28. Huang N, Miller WL 2000 Cloning of factors related to HIV-inducible LBP proteins that regulate steroidogenic factor-1-independent human placental transcription of the cholesterol side-chain cleavage enzyme, P450scc. J Biol Chem 275:2852–2858[Abstract/Free Full Text]
  29. Jones KA, Kadonaga JT, Rosenfeld PJ, Kelly TJ, Tjian R 1987 A cellular DNA-binding protein that activates eukaryotic transcription and DNA replication. Cell 48:79–89[CrossRef][Medline]
  30. Kruse U, Sippel AE 1994 The genes for transcription factor nuclear factor I give rise to corresponding splice variants between vertebrate species. J Mol Biol 238:860–865[CrossRef][Medline]
  31. Gronostajski RM 2000 Roles of the NFI/CTF gene family in transcription and development. Gene 249:31–45[CrossRef][Medline]
  32. Suske G 2000 Transient transfection of Schneider cells in the study of transcription factors. Methods Mol Biol 130:175–187[Medline]
  33. Celeste Simon M 1995 Gotta have GATA. Nat Genet 11:9–11[CrossRef][Medline]
  34. Molkentin JD 2000 The zinc finger-containing transcription factors GATA-4, –5, and –6. Ubiquitously expressed regulators of tissue-specific gene expression. J Biol Chem 275:38949–38952[Free Full Text]
  35. Baek SJ, Horowitz JM, Eling TE 2001 Molecular cloning and characterization of human nonsteroidal anti-inflammatory drug-activated gene promoter. Basal transcription is mediated by Sp1 and Sp3. J Biol Chem 276:33384–33392[Abstract/Free Full Text]
  36. Simpson RT 1998 Chromatin structure and analysis of mechanisms of activators and repressors. Methods 15:283–294[CrossRef][Medline]
  37. Hornsby PJ, Yang L, Gunter LE 1992 Demethylation of satellite I DNA during senescence of bovine adrenocortical cells in culture. Mutat Res 275:13–19[CrossRef][Medline]
  38. Srivastava D, Olson EN 2000 A genetic blueprint for cardiac development. Nature 407:221–226[CrossRef][Medline]
  39. Tremblay JJ, Viger RS 1999 Transcription factor GATA-4 enhances Mullerian inhibiting substance gene transcription through a direct interaction with the nuclear receptor SF-1. Mol Endocrinol 13:1388–1401[Abstract/Free Full Text]
  40. Tremblay JJ, Viger RS 2001 GATA factors differentially activate multiple gonadal promoters through conserved GATA regulatory elements. Endocrinology 142:977–986[Abstract/Free Full Text]
  41. Merika M, Orkin SH 1995 Functional synergy and physical interactions of the erythroid transcription factor GATA-1 with the Kruppel family proteins Sp1 and EKLF. Mol Cell Biol 15:2437–2447[Abstract]
  42. Boudreau F, Rings EH, van Wering HM, Kim RK, Swain GP, Krasinski SD, Moffett J, Grand RJ, Suh ER, Traber PG 2002 Hepatocyte nuclear factor-1 {alpha}, GATA-4, and caudal related homeodomain protein Cdx2 interact functionally to modulate intestinal gene transcription. Implication for the developmental regulation of the sucrase-isomaltase gene. J Biol Chem 277:31909–31917[Abstract/Free Full Text]
  43. Yamagata T, Nishida J, Sakai R, Tanaka T, Honda H, Hirano N, Mano H, Yazaki Y, Hirai H 1995 Of the GATA-binding proteins, only GATA-4 selectively regulates the human interleukin-5 gene promoter in interleukin-5producing cells which express multiple GATA-binding proteins. Mol Cell Biol 15:3830–3839[Abstract]
  44. Lei N, Heckert LL 2004 Gata4 regulates testis expression of Dmrt1. Mol Cell Biol 24:377–388[Abstract/Free Full Text]
  45. Kiiveri S, Liu J, Westerholm-Ormio M, Narita N, Wilson DB, Voutilainen R, Heikinheimo M 2002 Differential expression of GATA-4 and GATA-6 in fetal and adult mouse and human adrenal tissue. Endocrinology 143:3136–3143[Abstract/Free Full Text]
  46. Kiiveri S, Siltanen S, Rahman N, Bielinska M, Lehto VP, Huhtaniemi IT, Muglia LJ, Wilson DB, Heikinheimo M 1999 Reciprocal changes in the expression of transcription factors GATA-4 and GATA-6 accompany adrenocortical tumorigenesis in mice and humans. Mol Med 5:490–501[Medline]
  47. Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, Chrousos GP, Brennan MF, Stein CA, La Rocca RV 1990 Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res 50:5488–5496[Abstract/Free Full Text]
  48. Staels B, Hum DW, Miller WL 1993 Regulation of steroidogenesis in NCI-H295 cells: a cellular model of the human fetal adrenal. Mol Endocrinol 7:423–433[Abstract]
  49. Bielinska M, Parviainen H, Porter-Tinge SB, Kiiveri S, Genova E, Rahman N, Huhtaniemi IT, Muglia LJ, Heikinheimo M, Wilson DB 2003 Mouse strain susceptibility to gonadectomy-induced adrenocortical tumor formation correlates with the expression of GATA-4 and luteinizing hormone receptor. Endocrinology 144:4123–4133[Abstract/Free Full Text]
  50. Laitinen MP, Anttonen M, Ketola I, Wilson DB, Ritvos O, Butzow R, Heikinheimo M 2000 Transcription factors GATA-4 and GATA-6 and a GATA family cofactor, FOG-2, are expressed in human ovary and sex cord-derived ovarian tumors. J Clin Endocrinol Metab 85:3476–3483[Abstract/Free Full Text]
  51. Gillio-Meina C, Hui YY, LaVoie HA 2003 GATA-4 and GATA-6 transcription factors: expression, immunohistochemical localization, and possible function in the porcine ovary. Biol Reprod 68:412–422[Abstract/Free Full Text]
  52. Rosenfield RL 1999 Ovarian and adrenal function in polycystic ovary syndrome. Endocrinol Metab Clin North Am 28:265–293[CrossRef][Medline]
  53. Wood JR, Nelson VL, Ho C, Jansen E, Wang CY, Urbanek M, McAllister JM, Mosselman S, Strauss 3rd JF 2003 The molecular phenotype of polycystic ovary syndrome (PCOS) Theca cells and new candidate PCOS genes defined by microarray analysis. J Biol Chem 278:26380–26390[Abstract/Free Full Text]
  54. Wijesuriya SD, Zhang G, Dardis A, Miller WL 1999 Transcriptional regulatory elements of the human gene for cytochrome P450c21 (steroid 21-hydroxylase) lie within intron 35 of the linked C4B gene. J Biol Chem 274:38097–38106[Abstract/Free Full Text]
  55. Nebl G, Cato AC 1995 NFI/X proteins: a class of NFI family of transcription factors with positive and negative regulatory domains. Cell Mol Biol Res 41:85–95[Medline]
  56. Hagen G, Muller S, Beato M, Suske G 1992 Cloning by recognition site screening of two novel GT box binding proteins: a family of Sp1 related genes. Nucleic Acids Res 20:5519–5525[Abstract/Free Full Text]
  57. Hung HL, Pollak ES, Kudaravalli RD, Arruda V, Chu K, High KA 2001 Regulation of human coagulation factor X gene expression by GATA-4 and the Sp family of transcription factors. Blood 97:946–951[Abstract/Free Full Text]
  58. Kohler PO, Bridson WE 1971 Isolation of hormone-producing clonal lines of human choriocarcinoma. J Clin Endocrinol Metab 32:683–687[Medline]
  59. Schneider I 1972 Cell lines derived from late embryonic stages of Drosophila melanogaster. J Embryol Exp Morphol 27:353–365[Medline]
  60. Wijesuriya SD, Bristow J, Miller WL 2002 Localization and analysis of the principal promoter for human tenascin-X. Genomics 80:443–452[CrossRef][Medline]
  61. Read M 1998 Electrophoretic mobility shift assay (EMSA). In: Docherty K, ed. Essential techniques. New York: Wiley & Sons; 7–11

NURSA Molecule Pages Link:

Nuclear Receptors:   SF-1



This article has been cited by other articles:


Home page
 Endocr Relat CancerHome page
D Bonofiglio, H Qi, S Gabriele, S Catalano, S Aquila, M Belmonte, and S Ando
Peroxisome proliferator-activated receptor {gamma} inhibits follicular and anaplastic thyroid carcinoma cells growth by upregulating p21Cip1/WAF1 gene in a Sp1-dependent manner
Endocr. Relat. Cancer, June 1, 2008; 15(2): 545 - 557.
[Abstract]