| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Human Molecular Genetics Group, Institut de Génétique Humaine, CNRS UPR1142, 34396 Montpellier Cedex 5, France
ABSTRACT
In mammals, male sex determination is initiated by SRY (sex-determining region of the Y chromosome) gene expression and followed by testicular development. This study describes specific down-regulation of the human SRY gene transcription by cAMP stimulation using reverse transcription-polymerase chain reaction experiments. Using transfection experiments, conserved nuclear hormone receptor (NHR1) and Sp1 consensus binding sites were identified as essential for this cAMP transcriptional response. Steroidogenic factor-1 (SF-1), a component of the sex-determination cascade, binds specifically to the NHR1 site and activates the SRY promoter. Activation of SF-1 was abolished by cAMP pretreatment of the cells, suggesting a possible effect of cAMP on the SF-1 protein itself. Indeed, human SF-1 protein contains at least two in vitro cAMP-dependent protein kinase (PKA) phosphorylation sites, leading after phosphorylation to a modification of both DNA-binding activity and interaction with general transcription factors such as Sp1. Taken together, these data suggest that cAMP responsiveness of human SRY promoter involves both SF-1 and Sp1 sites and could act via PKA phosphorylation of the SF-1 protein itself.
cAMP, gene regulation, Sertoli cells
INTRODUCTION
In mammals, male sex determination is under the control of a Y chromosome-located gene, SRY (sex-determining region of the Y chromosome). This gene triggers the testis-determining pathway and sex cord formation [1], inducing Sertoli cell differentiation and subsequent anti-Müllerian hormone (AMH) or Müllerian-inhibiting substance production.
In mice, Sry expression is restricted between 10.5 to 12 days postcoitum (dpc) within the epithelial somatic cells of the genital ridge [2]. The Sry spatiotemporal expression and Sry mRNA expression level are both important for correct testicular development [3]. The apparent tight expression of mouse Sry led many authors to focus on SRY transcription regulation, but to date, factors controlling SRY expression remain poorly understood. The human SRY gene presents several transcription start sites [4–7], and putative regulatory elements have been highlighted only by analyzing different SRY genes between species [8, 9]. The 5'-flanking human SRY sequences revealed the presence of two Sp1-binding sites (-150, -130) and putative binding sites for NF-kB (-1000), for nuclear hormone receptors (NHRs; -890, -653, -315), for SOX proteins (-630, -400, -360, -90), for WT1 (-2430), and for a cAMP response element (CRE)-like motif (-750). This cAMP motif matches seven of eight bases of the consensus cAMP response element (TGACGTCA) [10]. As a first step toward the identification of regulatory elements controlling SRY transcription, binding of the Wilms' tumor (i.e., WT1) protein and the ubiquitous transcription factor Sp1 to sequences 5' of the initiation codon have been reported [11, 12].
A large number of CRE-binding proteins (CREB/CREM family) have been identified [13] with different structural characteristics and, consequently, diverse functions. Most cAMP-regulated genes contain a CRE sequence in their promoter. However, other regulatory sequences such as AP-1 [14], Pit-1, Sp1, or diverse NHR sites may also mediate cAMP-dependent gene regulation [15, 16]. Steroidogenic factor-1 (SF-1) has often been implicated in the basal activity and in the cAMP responsiveness of diverse steroid hydroxylase genes, including cytochrome P450 genes (CYPIIA) [17], aromatase gene (CYP19) [18], 17
-hydroxylase gene [19], high-density lipoprotein (HDL) receptor [20], and steroidogenic acute regulatory (StAR) genes [21, 22].
The orphan receptor SF-1 is a member of the NHR superfamily [23] and has been initially identified as a key regulator of endocrine functions within the hypothalamus-pituitary-gonadal axis and the adrenal cortex. This transcription factor recognizes a conserved regulatory motif in the proximal promoter regions of genes encoding the cytochrome P450 steroid hydroxylases [24–26]. These studies established that SF-1 was responsible, at least in part, for the tissue-specific expression of genes involved in steroid hormone biosynthesis. Broader roles for SF-1 emerged from genetic studies in mice, in which SF-1 knock-out mice exhibited adrenal and gonadal agenesis, leading to a male-to-female sex-reversal phenotype [27]. Furthermore, a mutation within the SF-1 gene in a sex-reversed 46,XY female was found [28]. Sertoli cell-specific genes, AMH gene, or AMH receptor II gene were shown to be directly regulated by SF-1 in vitro as well as in vivo [29–32]. Finally, early embryonic SF-1 expression within the genital ridge (9–12 dpc) and its differential regulation in male versus female gonads [24, 29] confirmed a pivotal role for SF-1.
To assess a putative correlation between cAMP regulation and SF-1 control of human SRY gene expression, two independent studies were performed. In the first, the effect of an increased cAMP concentration on SRY gene expression was investigated. In parallel, binding of SF-1 to the proximal NHR and its subsequent effect were evaluated. Finally, a cross-talk between these two pathways was demonstrated in different ways, including measurement of cAMP effect on SF-1 capacity to activate SRY expression or SF-1 posttranslational modification after cAMP-dependent protein kinase (PKA) activation. We then determined that cAMP action on SRY gene expression was in part due to SF-1 protein modification.
MATERIALS AND METHODS
Cell Culture and Transient Transfection
The human NT2/D1 cell line (N-Tera2, clone D1, a human pluripotent embryonic carcinoma cell line, ATCC n° CRL 1973) was obtained from the American Type Culture Collection (ATCC, Biovalley, France). The SRY stably transfected HeLa clone named HeLaB3 was previously described [33]. In HeLaB3, SRY was expressed under the control of the SV40 early promoter. The NT2/D1 and HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Life Technologies Inc.) containing 10% (v/v) fetal calf serum (Life Technologies Inc., Cergy-Pontoise, France), penicillin-streptomycin, and 2 mM glutamine at 37°C in a 5% CO2 atmosphere. Plasmids used for transfection were purified using the Maxiprep reagent system (Qiagen, Courtaboeuf, France), and at least three independent preparations of each plasmid were tested. Typically, cells were plated in a 24-well plate at a density of 105 cells/well 24 h before transfection. Plasmid-liposome complexes were formed using 2 µl of Fugene 6 (Roche Diagnostics, Meylan, France) for NT2/D1 cells. Next, 500 ng of SRY promoter-luciferase reporter plasmid were cotransfected with 100 ng of SF-1 expression vectors, and 10 ng of TK promoter-Renilla luciferase plasmid were used as an internal transfection control. Three hours after plasmid addition, transfection medium was replaced by DMEM medium for 48 h of incubation. Cells were lysed directly in transfection plates by 100 µl of Passive Lysis Buffer (Dual Luciferase Assay Kit; Promega, Lyon, France). Lysates were assayed for firefly and Renilla luciferase activities according to the Dual Luciferase Assay Kit manufacturer's instructions. Results are reported as the ratio between firefly and Renilla activities from at least three separate transfection experiments performed in triplicate.
Plasmid Constructions
The -916 (pGL916), -727 (pGL727), and -461 (pGL461)/-13 SRY promoter fragments were generated by polymerase chain reaction (PCR) with the VENT polymerase (Biolabs Laboratories, Ozyme, France) using the -916, -727, and -461 oligonucleotides, respectively, as 5' primer and the -13 oligonucleotide as 3' primer. For the PCR reaction, 5' and 3' primers contained, respectively, KpnI and XhoI sites at their 5' end. Numbers indicate positions from the first ATG initiation start codon as determined by Behlke et al. [4] and are positions of the primer's 5' end. The SRY promoter fragments were cloned into the SmaI-linearized pUC18 (SureClone kit; Amersham Pharmacia Biotech Europe GmbH, Orsay, France). The KpnI-XhoI SRY-containing fragments were subcloned into the KpnI-XhoI pGL3 basic luciferase vector (Promega). The SF-1 mutations were introduced in the pGL916 construct by PCR using mutated 29 base pairs (bp; -327/-298) of oligonucleotide steroidogenic factor (Sf): 5'-TGACATAAAAGGTCAATGAAAAAATTGGC to 5'-TGACATAAAATTTCAATGAAAAAATTGGC (mutation m1) and 5'-TGACATAAAAGGCAGATGAAAAAATTGGC (mutation m4), and the PfuI polymerase according to the QuickChange Site-Directed Mutagenesis Kit (Stratagene, Tebu, France). The Sf oligonucleotide corresponds to the -327 to -298 region of the human SRY promoter, overlapping the SF-1 element (-315). Mutation of site Sp1 was introduced using mutated Sp1 oligonucleotide (5'-GGTAGGCTGGTTGGGCGGGGTTGAG to 5'-GGGTAGGCTGGTTTCAAGGGGTTGAG). Automated sequencing (Perkin-Elmer, Courtaboeuf, France) checked the fidelity of the new plasmid constructs. Numbers on the 5'-flanking sequence and on the open-reading frame of the SRY gene are taken in relation to the first ATG codon as determined by Behlke et al. [4]. The expression vectors for human full-length (wt) deleted ligand-binding domain (
CT) and ligand-binding domain (CT) SF-1 (pcDNA and pGEX4T3) were previously described [31]. The SF-1 CT and SF-1 CT contained amino acids 1–226 and 225–461, respectively, of the SF-1 protein. Expression vector for Sp1 was kindly provided by R. Tjian.
Electrophoretic Mobility Shift Assays
Nuclear extracts from NT2/D1 and SF-1-transfected HeLa cells were prepared according to the method described by Dignam et al [34]. The GST/SF-1 fusion proteins were produced in bacteria according to the method described by de Santa Barbara et al. [31]. For gel mobility shift assays, the 29-bp double-stranded Sf oligomer was labeled by a fill-in reaction with -32P-deoxycytidine triphosphate and purified on a 5% w/v acrylamide, nondenaturating gel. Binding reactions and migration were performed as previously described [12]. For competition experiments, unlabeled oligonucleotides (100 molar excess) were incubated with proteins for 15 min before addition of the labeled probe. Oligonucleotides Sf, Sfm1, Sfm4, and unrelated oligomer (-916/-888 SRY sequence: 5'-CACTTGTTTAGTCTGGTAAACTGTGACC) were used for competition experiments. For SF-1 supershift analysis, proteins were incubated with preimmune antisera or with anti-human SF-1 polyclonal antibody [31] for 15 min before addition of the labeled oligonucleotide.
RNA Experiments
Total cell RNA was isolated according to the protocol described by Chomczynski and Sacchi [35]. The quality of RNA was verified at the ribosomal RNA level by agarose gel analysis and ethidium bromide staining. First, 15 µg of total RNA were treated by DNaseI (Life Technologies Inc.) for 15 min at room temperature and submitted to Moloney murine leukemia virus reverse transcriptase (Promega) in the presence of oligonucleotide d(T)12–18 (1 µg; Amersham Pharmacia Biotech Europe GmbH) for 1 h at 42° in a 30-µl volume. The SRY cDNA (12 µl) was then amplified by PCR using oligonucleotides spanning regions between +170/+197 (Xes8: 5'-GATAGAGTGAAGCGACCCATGAACGCA) and +387/+408 (H21.5: 5'-ATCTTCGCCTTCCGACGAGGT). Next, 2 µl of cDNA were used to amplify 200-bp glyceraldehyde phosphate dehydrogenase (GAPDH) products using oligonucleotides 5'-GTCTTCACCACCATGGAG and 5'-CAGTACCTACTGGAACCG spanning intron 2 of the GAPDH gene.
Protein Purification and In Vitro PKA Assays
The recombinant SF-1 fusion proteins GST/SF-1, GST/SF-1 CT (deleted ligand-binding domain [LBD]), GST/SF-1 CT (LBD), and the GST protein were produced in bacterial strain BL21 as previously described [31]. The purified proteins were checked by SDS-PAGE gel electrophoretic analysis and used directly for in vitro PKA assays. In these assays, 250 ng of protein were added to a standard PKA reaction mixture containing 20 mM Hepes (pH 7.5), 5 mM MgCl2, 1 mM unlabeled ATP, 1 mM 1,4-dithiothreitol, 100 mM NaCl, and 1 mM 
-32P-ATP (0.5 Ci/mmol) and then incubated with 1 mU of PKA catalytic subunit (1 U = 1 mmol phosphate/min; Promega) for 1 h at 37°C. Reactions were stopped by addition of 20 µl of 2x Laemli sample buffer, and proteins were separated by gel electrophoresis and submitted to autoradiography.
In Vivo Coimmunoprecipitation
Detection of SF-1 partners was realized in vivo in the NT2/D1 cell line treated or not treated by 8-bromo-cAMP (8-Br-cAMP) as described elsewhere [31] in TBST buffer (10 mM Tris-HCl [pH 7.5], 130 mM NaCl, 0.5% Tween, and 0.5% BSA). The rat SF-1 antibody conjugated to the protein G Sepharose was used to immunoprecipitate endogenous SF-1 protein. The final beads were washed several times in TBST buffer, and the proteins were suspended in 2x Laemli buffer and subjected to SDS-PAGE. After transfer of the proteins to nitrocellulose membrane with a Trans-Blot apparatus (Bio-Rad, Ivry-sur-Seine, France), Western blots were performed using rabbit SF-1 antibody [31] or mouse Sp1 antibody (1C6; Santa Cruz Laboratories, Tebu, France) and revealed with an enhanced chemiluminescence kit (Amersham).
RESULTS
Effect of cAMP Pathway Stimulation on the Human SRY Gene Transcription
Because of the presence of a nearly consensus CRE (TGACTTCA, -750) in the 5'-flanking sequence of the human SRY gene, cAMP regulation of SRY transcription was examined. Previous treatment of cultured Sertoli cells with cAMP analogues completely abolished Sry mRNA expression [36]. Total cellular RNA from NT2/D1 cells, a cell line positive for the expression of different genes involved in the sex-determination pathway [5, 31], was subjected to reverse transcription (RT)-PCR experiments. Using 40 amplification cycles, SRY transcripts were visualized (Fig. 1). Under the same assay conditions, with 8-Br-cAMP-treated NT2/D1 cell RNA, no SRY transcript was detected, confirming that cAMP may down-regulate SRY gene expression at the transcriptional level. As a positive control for the RT experiment, SRY mRNA from SRY stably transfected HeLaB3 cells was used (Fig. 1). Amplification of GAPDH cDNA and visualization of RNA on agarose gel were used to check the homogenous quality and the quantity of NT2/D1 and HeLa B3 RNAs (Fig. 1).
|
Involvement of the Proximal NHR1 and Sp1 Sites in the SRY Gene cAMP Regulation
To determine the genomic elements involved in cAMP regulation of the human SRY gene, nested deletions (-916/-13, -727/-13, and -461/-13) were created by PCR. The resulting fragments were ligated upstream of the luciferase reporter gene of the basic pGL3 vector, and the constructs were transfected into NT2/D1 cells (Fig. 2A). For all these constructs, incubation of transfected cells with 8-Br-cAMP resulted in a 40%–50% decrease in basal transcriptional activity. These data suggest that the previously described CRE element (-750) (Fig. 2A) is not implicated in the cAMP response of the SRY promoter, and that the CRE is contained within the proximal 461-bp fragment. This fragment is sufficient to direct expression of the SRY gene in the NT2/D1 cell line (18-fold compared to that of the empty, basic pGL3 reporter vector) (Fig. 2A). Using pGL916 and pGL727 constructs, reporter gene activities are stimulated by 37- and 52-fold, respectively, in NT2/D1 cells, indicating other positive elements in the -916/-416 region.
|
The ubiquitous Sp1 transcription factor was previously shown to bind this 416-bp proximal region [12]. This region also contains a putative binding site for NHR1 (AGGTCA, -315). Mutation of NHR1 (-315) site (pGL916NHRm1) reduces promoter activity by 70% (Fig. 2B), whereas additional mutation of Sp1 sites reduces basal promoter activity of pGL916 construct by nearly 90%. These data suggest that these proximal NHR1 and Sp1 sites are essential for transcriptional activity of the human SRY gene under these conditions. The same mutations on the SRY promoter modify the cAMP response (Fig. 2B), indicating that binding of a nuclear receptor and Sp1 to their sites is required for the cAMP transcriptional down-regulation of the human SRY gene, and that other cofactors might be involved in this cAMP response. These NHR1 and Sp1 sites are conserved among SRY regulatory sequences of different species [9]. As Sp1- and SF-1-binding sites have been directly involved in the cAMP-dependent regulation of steroid hydroxylase gene expression [15–17], this hypothesis was tested for the cAMP regulation of the SRY gene.
Binding of SF-1 to the Human SRY Promoter
Binding to the putative NHR1-binding site (-315) by the SF-1 protein, an orphan NHR involved in the sex-determination cascade, was tested next. The synthetic -327/-298 oligonucleotide was used in gel mobility shift assays with NT2/D1 nuclear extracts, a cell line positive for SF-1 protein expression [31]. As shown in Figure 3A, the -327/-298 probe forms a major slow-migration complex in the presence of NT2/D1 extracts (lane 2). This is a multiprotein complex containing other proteins such as SOX9 [31]. Partial retardation of the complex was observed after incubation of the extracts with a specific human SF-1 antibody (lane 3). Formation of this complex was specifically abolished by adding the unlabeled -327/-298 sequence (lane 5). This complex was not abolished when incubated with the m1 or m4 sequences (lanes 6 and 7) or with the unrelated oligonucleotide (lane 8). These data indicate that the five bases (GGTCA) of the core site are important for nuclear binding protein and consistent with the general binding requirement of most NHRs, even though additional A-T sequences upstream of this site may contribute to binding [37, 38]. These data also indicate that SF-1 protein specifically binds to the SRY promoter, and that the five bases (GGTCA) of the site are involved in this binding.
|
The SF-1 binding to the -327/-298 sequence was then confirmed using electrophoretic mobility shift assay (EMSA) experiments and GST-SF-1 fusion protein or SF-1-transfected HeLa cell nuclear extracts (Fig. 3B). A major complex was obtained (lanes 2 and 6) and specifically supershifted after incubation with the SF-1 antibody (lanes 3 and 5). Addition of preimmune antiserum did not modify DNA-SF-1 complex mobility (lanes 4 and 7). These results show that SF-1 protein can bind to the -327/-298 SRY element. It is noteworthy that, using SF-1-transfected HeLa cell extracts, the SF-1 complex migration rate was identical to the rate obtained with the heavier GST/SF-1 fusion protein. This result was reproducible and has been previously described using an AMH probe [31]. This effect might reflect differential conformation of the SF-1 protein when coupled to the GST protein and/or posttranslational modification of the SF-1 protein within HeLa cells.
Effect of cAMP Treatment on SF-1 Activation
To determine the ability of SF-1 to regulate the SRY promoter, cotransfection assays were performed with SRY-pGL3 luciferase constructs and a SF-1 expression vector in NT2/D1. Overexpression of SF-1 protein resulted in a 2–2.5-fold induction of SRY constructs in relation to that of the basic pGL3 empty vector (value of 1). This activation depended on the LBD of the SF-1 protein being present (Fig. 4A). The proximal SF-1 site (-315) was sufficient for the SF-1 response, because mutation of the site abolished SF-1 activation of the pGL916SFm1 construct. In NT2/D1 cells, using the SF-1 expression vector and pGL916 construct, SF-1 activation was abolished after 8-Br-cAMP treatment (Fig. 4B). This cAMP effect was not caused by modification of SF-1 protein expression and cellular localization in the NT2/D1 cells (data not shown). In summary, these data suggest that activation of the SRY promoter by SF-1 requires the proximal SF-1-binding site and presence of the LBD of SF-1. This activation was specifically abolished by cAMP treatment of the cells, suggesting that posttranslational modification of SF-1 may be involved in this mechanism.
|
SF-1 PKA Phosphorylation Inhibits Its Binding to the SRY Promoter
To study the mechanism by which SF-1 could mediate cAMP decrease of the human SRY gene transcription, we tested the effect of posttranslational modification of SF-1 on SRY activation. Rat SF-1 protein has been shown to be phosphorylated in vitro by PKA [39]. Even though no PKA phosphorylated sites have been mapped on the protein sequence, a possible site has been highlighted by Honda et al. [40] on amino acids 427–431. This site has recently been implicated in SF-1-dependent cAMP regulation of the HDL receptor [20]. Using in vitro PKA assays, we showed that human SF-1 protein is a PKA phosphoprotein (Fig. 5A). Full-length GST/SF-1 wt, LBD deleted-GST/SF-1 CT, and GST/SF-1-LBD CT fusion proteins were phosphorylated by PKA, whereas the GST protein alone was not (data not shown). These data indicated that human SF-1 protein contained at least two PKA sites: one on the LBD (as previously suggested), and the other on the N-terminal part of the protein.
|
The effect of SF-1 phosphorylation on its binding to the SRY proximal promoter was further investigated. Purified recombinant SF-1 fusion protein was phosphorylated in vitro with unlabeled ATP and incubated with the labeled SF-1-site oligonucleotide probe previously used in the EMSA experiments. Gel mobility shift assay showed that increasing PKA catalytic subunit concentrations induced a decrease in GST/SF-1 wt fusion protein binding to the SRY promoter (Fig. 5B, lanes 2–4). This effect was not caused by different SF-1 protein amounts, as confirmed by Western blotting (Fig. 5B). Previous phosphorylation of GST/SF-1 CT fusion protein did not modify its binding activity (Fig. 5B, lanes 5–7), indicating that phosphorylation on the LBD domain was responsible for modification of the SF-1-binding activity.
Effects of SF-1 Phosphorylation on Its Interaction with Sp1 Transcription Factor
Finally, the role of SF-1 phosphorylation in respect to its interaction with other regulatory proteins was investigated. As described previously for the CYP11A gene [17], interaction with the Sp1 transcription factor was studied. Coimmunoprecipitation assays revealed interaction between human SF-1 protein and Sp1 in the NT2/D1 cell line (Fig. 6). The putative role of SF-1 phosphorylation on these interactions was estimated by the same experiments in the presence of 8-Br-cAMP. As shown in Figure 6, SF-1/Sp1 interaction was greatly reduced after cAMP treatment, suggesting that cAMP stimulation inhibits interaction of SF-1 with Sp1.
|
DISCUSSION
Tight spatiotemporal expression of the human SRY gene during early embryogenesis requires the integration of extracellular signals. In the present study, we have shown that SRY gene expression is specifically down-regulated by cAMP treatment in the NT2/D1 cell line. Using RT-PCR experiments, negative regulation of SRY transcription by cAMP signaling pathway was observed. These data confirm preliminary observations of the down-regulation of Sry mRNA expression in cultured Sertoli cells after treatment with cAMP analogues [36]. The cAMP-responsive region was localized in the 416-bp proximal sequences using progressive 5' deletions of the SRY flanking region. This sequence excludes the previously described CRE sequence localized at position -750. In this case, as for most of the steroid hydroxylase genes, regulation of SRY transcription by cAMP signaling does not appear to rely on transcription factors traditionally associated with the cAMP pathway, such as the CREB/CREM protein family [13]. Within these sequences reside two binding sites: one for the ubiquitous transcription factor Sp1 [12], and a hemisite for NHR1. Mutations of either site inverted the cAMP response of SRY transcription, indicating that cAMP down-regulation requires both NHR1- (i.e., SF-1-) and Sp1-binding sites. Similar requirements were described for regulation of the bovine CYP11A gene by cAMP [17, 41]. These data also indicate that, in the absence of SF-1 binding, one or more cofactors can induce SRY expression after cAMP treatment. This SRY proximal region has been previously delineated as a minimal promoter for SRY in the mouse Ltk- and TM4 cells [7].
The differential expression of SF-1 in male versus female [24] and its expression profile in the embryo [27] prompted us to investigate its possible role in SRY transcription regulation. Furthermore, recent data suggest that the male-specific cell proliferation observed in embryonic mouse gonad after Sry expression is restricted to SF-1-positive cells [42]. This observation reinforces the possibility of a direct control of SRY expression by SF-1. The EMSA experiments showed that SF-1 protein could bind to the NHR1 (i.e., SF-1) site of the -327/-298 5' SRY sequence, and competition assays revealed the direct involvement of the five bases GGTCA in this binding. This sequence (GGTCA) is a half-site for the binding of most steroid NHRs [37], even though numerous reports have suggested that the nucleotides 5' of the response element for SF-1 and other orphan nuclear receptors might influence binding activity [37, 43–44]. Individual mutation of the SF-1 site reduced basal activity of the human SRY promoter by 50%, showing that the SF-1 site contributes partially to expression of the SRY gene. This observation might explain the persistent Sry expression still detected in SF-1 knockout mice [45].
Cotransfection experiments in NT2/D1 cells with SRY-pGL3 constructs and SF-1 expression vector allowed a twofold transactivation of the intact endogenous SRY promoter. Even though low, this activation level is sufficient to correlate binding SF-1 and transcriptional activation. The SF-1 activation depended on the presence of LBD, suggesting that a ligand, cofactor, posttranslational modification, or some combination of these alternatives could be necessary for activation by SF-1 [29].
The SRY 5'-flanking region contains three putative binding sites for nuclear receptors at positions -315, -653, and -890. Deletion of sites at positions -693 and -890 did not affect either SF-1 activation or cAMP responsiveness of the SRY promoter, indicating that site -315 is sufficient for both responses. The distal sites might bind other nuclear receptors to activate or repress SRY promoter expression. However, this contradicts recent reports showing that, in both human and rat StAR promoters, three SF-1-binding sites are required for maximal promoter activity and for cAMP regulation [21, 22, 46–48].
The cAMP regulation of SF-1-mediated transcriptional activation may be justified by diverse means. The protein might be a target of posttranslational modification such as phosphorylation. Hammer et al. [49] recently showed that SF-1 was a target for mitogen-activated protein kinase phosphorylation and delineated the phosphorylation site on Ser 203 within the AF1 domain. Rat SF-1 protein can also be phosphorylated in vitro by PKA in a dose-dependent fashion [39]. Finally, SF-1 was also shown to be an in vivo cAMP target in the ovary, being phosphorylated in granulosa cells after FSH treatment [18]. The phosphorylation sites have been mapped on serine and threonine, but without identifying their respective positions in the primary sequence [40]. The role of Ser 431 near the AF2 domain within the SF-1 LBD domain was recently outlined for the SF-1, cAMP-dependent regulation of HDL expression [20]. In vitro, we now confirm that human SF-1 is a target for PKA, and this phosphorylation might contribute to the regulation of SF-1 activity.
Studies on diverse nuclear receptors, including SF-1 [49–51], have shown that phosphorylation can lead to changes in the protein structure and, thus, can affect DNA binding and/or protein-protein interaction as well as nuclear localization. In EMSA experiments after in vitro PKA phosphorylation of SF-1, a 3.5-fold decrease of SF-1-binding activity on the SRY promoter was observed, confirming previous experiments using the LH promoter [18]. As shown by transfection experiments, SF-1 activation of SRY transcription was abolished by such cAMP treatment. Coimmunoprecipitation assays showed that protein-protein interaction between SF-1 and Sp1 was specifically inhibited after stimulation of the cAMP pathway in NT2/D1 cells. Regulation of SF-1 interaction with other transcription factors via a cAMP mechanism has been reported in many other instances [18, 37, 41, 47].
The SF-1 and Sp1 transcription factors interact with each other and seem to be involved in SRY transcriptional regulation. Cooperativity between Sp1 and other enhancer-binding factors in regulating promoters is often described. Liu and Simpson [17] reported cooperative activation of the bovine CYP11A promoter by both SF-1 and Sp1. Cooperative effects were higher than those observed for the SRY promoter, probably resulting from use of the GAL4-VP16 mammalian two-hybrid system, and was mediated by a direct protein-protein interaction between the couple SF-1/Sp1 and the CREB-binding protein (CBP) [41]. The SF-1 and Sp1 proteins interact in the NT2/D1 cell line and can cooperate to transactivate SRY expression, but they might need other factors to fully activate the gene. Furthermore, their respective expression (earlier than SRY expression) could indicate that a cofactor and/or multiple general cofactors would join the transcriptional complex and induce the onset of SRY expression. Among them, a possible effector might be WT1, the Wilms' tumor-suppressor gene, which binds to the SRY promoter [11] and cooperates with SF-1 to activate AMH expression [52]. Interaction of SF-1 and the WT1 (-KTS, exon 9) splicing isoform has also been suggested to synergistically activate Dax-1 expression [53]. General nuclear receptor cofactors [54] such as CBP or SRC-1 might also be effectors of this expression.
Because of the residual SRY expression in SF-1 -/- mice [45], SF-1 might be essential mainly for switching off SRY gene expression. The cAMP stimulation, the primary effector of which remains to be determined, might lead to phosphorylation of the SRY transcriptional complex proteins. The resulting decrease of SF-1 binding on the SRY promoter, together with the inhibition of its interaction with Sp1 and/or other nuclear factors, might lead to the SRY extinction. Further studies are in progress to delineate different combinations that might be involved in human SRY gene regulation.
ACKNOWLEDGMENTS
We thank Patrick Atger for his expertise in the preparation of illustrations. The Sp1 expression vector was supplied by Prof. R. Tjian. We also thank Wafaa Takash for careful reading of the manuscript.
FOOTNOTES
1 Supported by the European Economic Community through BIOMED2 Grant BMH4-CT96-0790 and Ligue Nationale contre le Cancer and Association pour la Recherche sur le Cancer (ARC). P.D.S.B. was supported by a postdoctoral fellowship from ARC (Association pour la Recherche contre le Cancer). 
2 Correspondence: Philippe Berta, IGH-CNRS UPR 1142, 141 rue de la Cardonille, 34396 Montpellier Cedex 5, France. FAX: 33 4 99 61 99 01; berta{at}igh.cnrs.fr
3 Current address: Department of Pathology and Pediatric Surgical Research, Massachusetts General Hospital, 32 Fruit St., Boston, MA 02114.
Accepted: October 20, 2000.
Received: April 24, 2000.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. Nikolova, J. S. Sinsheimer, E. M. Eicher, and E. Vilain The Chromosome 11 Region From Strain 129 Provides Protection From Sex Reversal in XYPOS Mice Genetics, May 1, 2008; 179(1): 419 - 427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Lewis, M. Rusten, E. A. Hoivik, E. L. Vikse, M. L. Hansson, A. E. Wallberg, and M. Bakke Phosphorylation of Steroidogenic Factor 1 Is Mediated by Cyclin-Dependent Kinase 7 Mol. Endocrinol., January 1, 2008; 22(1): 91 - 104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Wilhelm, S. Palmer, and P. Koopman Sex Determination and Gonadal Development in Mammals Physiol Rev, January 1, 2007; 87(1): 1 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kanai, R. Hiramatsu, S. Matoba, and T. Kidokoro From SRY to SOX9: Mammalian Testis Differentiation J. Biochem., July 1, 2005; 138(1): 13 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. R. Harley, M. J. Clarkson, and A. Argentaro The Molecular Action and Regulation of the Testis-Determining Factors, SRY (Sex-Determining Region on the Y Chromosome) and SOX9 [SRY-Related High-Mobility Group (HMG) Box 9] Endocr. Rev., August 1, 2003; 24(4): 466 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Pilon, I. Daneau, V. Paradis, F. Hamel, J. G. Lussier, R. S. Viger, and D. W. Silversides Porcine SRY Promoter Is a Target for Steroidogenic Factor 1 Biol Reprod, April 1, 2003; 68(4): 1098 - 1106. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
