|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Male Reproductive Tract |
a Centre de recherche en reproduction animale, Department of Veterinary Biomedicine, Faculty of Veterinary Medicine, University of Montreal, St-Hyacinthe, Québec, Canada J2S 7C6
b Centre de recherche en biologie de la reproduction, Department of Obstetrics and Gynecology, Faculty of Medicine, Université Laval, Québec City, Québec, Canada G1V 4G2
| ABSTRACT |
|---|
|
|
|---|
developmental biology, embryo, gene regulation, male reproductive tract, testis
| INTRODUCTION |
|---|
|
|
|---|
In the mouse, the window for genital ridge expression of Sry is narrow, beginning at 10.5 days postcoitus (dpc), reaching a peak at 11.5 dpc and terminating after 12.5 dpc [68]. Sry expression is restricted to the somatic component of the genital ridge during development, although circular nontranslated transcripts are also detected in germ cells of the adult testis [9, 10]. Genital ridge expression of SRY is now described in several additional mammalian species. In the pig, SRY expression was detected by reverse transcriptase-polymerase chain reaction (RT-PCR) within the XY indifferent genital ridge by at least 21 dpc with a maximum at 23 dpc [11], whereas SRY transcripts were still detectable at 52 dpc [12]. The sheep SRY transcript is first detected by RT-PCR at 23 dpc, it peaks between 27 and 44 dpc, and can still be detected at the time of birth, 21 wk after conception [13]. In studies of human XY genital ridges via in situ hybridization, SRY is first detected between 41 and 44 days postovulation (dpo), it peaks at 44 dpo, and can still be detected in the testicular cords at 18 wk of gestation [14]. In addition, RT-PCR analysis of different human fetal and adult tissues have revealed that SRY expression is not as tightly regulated as mouse Sry [15], although the significance of nongonadal expression of SRY is not known.
Promoter studies have been performed in vitro for human SRY sequences and revealed several different transcriptional start sites [1518]. It was shown that the regulatory elements required for basal promoter activity lie within the first 310 base pairs (bps) upstream of the translation start site [17]. Two Sp1 sites contained within this 310 bp promoter have been functionally characterized [19], but they cannot account for the tissue-specific expression of SRY. Comparative structural studies between SRY 5' flanking sequences of different species were performed and revealed the presence of several highly conserved motifs, including potential recognition sequences for SRY itself [20, 21]. It is reported that WT-1 and steroidogenic factor-1 (SF-1) can each bind and transactivate the human SRY promoter [2224], suggesting that these factors may contribute to the tissue-specific expression profile of SRY. We recently reported that SOX9 could transactivate the pig SRY promoter via a SOX9 binding site at -205 from the ATG translational start site [25].
The mouse Sry promoter remains poorly characterized in vitro, although transgenic studies involving the mouse Sry promoter have been more informative. In the original transgenic study reported, a genomic fragment of 14.6 kilobase (kb) including about 7.5 kb of 5' flanking sequences was expressed within the genital ridge and was able to cause sex conversion of XX embryos [3]. More recent studies have shown that transgenes, including 4 kb of mouse Sry 5' flanking sequences, are active in vivo [26].
To provide comparative sequence data, we have cloned 4.5 kb of 5' flanking sequences from the porcine SRY gene locus. To initiate the characterization of these sequences in vitro, we have generated a pig genital ridge cell line that supports the expression of the SRY gene. To address the role played by SF-1 in SRY promoter activity, we have identified SF-1 potential binding sites within pig SRY promoter sequences and have shown that these sites are functional.
| MATERIAL AND METHODS |
|---|
|
|
|---|
The cloning strategy for pig SRY promoter is presented in Figure 1. In Figure 1A, a simplified view of the pig SRY locus is depicted. To initiate the cloning of a 6.6 kb genomic HindIII DNA fragment from the pig SRY locus [2], an anchored PCR procedure was used, similar to that previously reported for cloning of a 1.7 kb EcoRI genomic fragment containing the pig SRY open reading frame (ORF) [11]. Briefly, male pig genomic DNA was restricted with HindIII and size-fractionated on a 0.8% agarose gel. The bands from 6 to 8 kb were excised and ligated into HindIII-restricted pBS plasmid (Stratagene, La Jolla, CA), to generate a size-selected plasmid library. To generate 3' sequences (Fig. 1B), anchored PCR was performed using specific sense primers derived from the 3' end of the reported pig SRY locus (first primer, 5'-CACACAAACTGCTTGATTTCG-3' and nested primer, 5'-TTCCCGTGATTAGCCATTAAGTACG-3') [11] and primers derived from plasmid sequences (first primer, 5'-AAAGGGGGATGTGCTGCAAGGCG-3' and nested primer, 5'-TGGGTAACGCCAGGGTTTTCCCA-3'). A proofreading mix of thermostable polymerases (Expand High Fidelity; Roche Molecular Biochemicals, Laval PQ, Canada) was used for the amplifications. The first PCR amplification used 40 cycles of 45 sec at 95°C, 45 sec at 56°C, and 4 min at 70°C; this was followed by a nested PCR amplification using the same cycling program. This strategy proved successful for amplifying the 3' end of the genomic HindIII fragment, which was cloned into pGEM-T vector (Promega, Madison, WI), and sequenced.
|
Anchored PCR was unsuccessful for generating the 5' end of the pig HindIII fragment, so a reverse PCR strategy was adapted (Fig. 1C). Male pig genomic DNA was again restricted with HindIII, and bands from 6 to 8 kb were excised and then ligated under dilute conditions to favor self-ligation of fragments. Sense primers were designed from pig sequences at the 3' end of the genomic HindIII fragment, and antisense primers from pig sequences at the 5' end of the previously reported EcoRI genomic fragment of the pig SRY locus [11]. A first PCR was performed using the sense primer 5'-AAGCTGATGGTCTCTTGTCTCTGTA-3' and antisense primer 5'-TTCCTTTCGGCCATTAGAGCACTCA-3'; a second PCR was then performed using the nested sense primer 5'-CTTTCCAGTGCATATATTCCAAAGC-3' and antisense primer 5'-CGGATGTTATAGAGTTGAATGCTAG-3'. For each amplification, 40 cycles of 45 sec at 95°C, 45 sec at 66°C, and 4 min at 72°C were performed. An amplified band of about 3.9 kb was ligated into pGEM-T vector, and when sequenced, proved to represent 5' flanking sequences of the pig SRY locus.
Based on sequences obtained for the 5' end of the pig SRY HindIII fragment, and coding sequences of pig SRY previously reported [11], a genomic PCR amplification strategy was performed to obtain intact 5' flanking sequences for use in promoter studies. A single sense primer (5'-AAGCTTGGGGAAATCTGTTCAGTA-3') and two antisense primers, 5'-GGGGAAATCTGTTCAGTAG-3' and (nested) 5'-TTGAAAAGGGGG-AGGAAGC-3' were designed with the 5' HindIII site engineered. Male pig genomic DNA was used as a template. A first PCR amplification (as described above) was performed followed by a second, heminested PCR amplification under similar conditions. The thermostable polymerase Expand High Fidelity (Roche) was used. An amplified band of about 4.5 kb was then ligated into the plasmid vector pGEM-T, sequenced and shown to represent 4539 bp of pig SRY 5' flank up to (but not including) the ATG translational start site.
To construct the reporter transgene for in vitro and in vivo characterization, the 4.5 kb pig SRY 5' flank was placed into a unique HindIII site in front of a modified enhanced green fluorescent protein reporter sequence (pEGFP-1; Clontech, Laboratories, Palo Alto, CA) to give the transgene SRYp4.5kb-GFP (Fig. 1D).
Generation of Porcine Genital Ridge Cell Lines
Uteri from pregnant sows of known gestational length (2223 dpc) were obtained at a local abattoir and embryos were harvested. Genital ridges were then dissected and pooled together. Cells were dissociated in Dulbecco modified Eagle medium (DMEM) supplemented with ovalbumine (0.1%), DNase I (200 U/ml), collagenase (50 U/ml), and dispase 2.4 (U/ml) at 37°C, with agitation for three rounds of 20 min each. After each round, medium was taken and centrifuged at 1500 RPM for 10 min to recover the cells. By this method, approximately 2.5 x 107 cells were obtained, which were then transfected with 20 µg of pBK-CMV plasmid containing SV40 large-T antigen. Transfection was performed using 80 µl of Lipofectamine reagent according to the manufacturer's manual (Gibco-BRL, Grand Island, NY), in DMEM without serum. Cells were immediately placed onto 96-well plates (12 in total) and incubated at 37°C in a 5% CO2 atmosphere. Transfection was stopped after 5 h by adding DMEM supplemented with 20% newborn calf serum. G418 selection was initiated after 72 h of culture, and maintained for 3 wk. Experiments were performed using cells of approximately passage number 10.
RT-PCR Characterization of the Porcine Genital Ridge Cell Lines
Total RNA from the four porcine genital ridge (PGR) cell lines obtained was extracted via pelleting over a CsCl gradient [27]. RT-PCR was performed for porcine SOX9, SF-1, DAX-1, SRY, and GAPDH genes based on primers designed from GenBank sequences AF029696, U84399, AF019044, U49860, and AF017079, respectively. RT reactions were performed using the antisense primers pSOX9.3 (5'-TGTTGGAGATGACGTCGCTGCTCA-3'), pSF-1.3 (5'-CAGCTCCTTGAAGACCATGCAC-3'), pDAX-1.12 (5'-CTGCTCACAGCTCCTGTACTTGG-3'), pSRY.3'dT (5'-TTTTTTTTTT-TTTTTTTITGCACAAGGGACTG-3', where I represents inosine), and pGAPDH.1 (5'-AGGTCCACCACCCTGTTGCTGTA-3'), respectively, and RT enzyme according to the supplier's manual (Superscript II RT; Gibco). Forty cycles of PCR were then performed with 2 µl of the 30-µl total volume from the first-strand cDNA synthesis reaction using the following primers: pSOX9.A (5'-CGTATGAATCTCCTGGACCCCTT-3') and pSOX9.4 (5'-ATGTCCACGTCGCGGAAGTCGAT-3') for pSOX9, pSF-1.A (5'-GTACGACGAGGACCTGGACGA-3') and pSF-1.5'1 (5'-GCCT-GTTTCCAGCTTGAAGCCATT-3') for pSF1, pDAX-1.E (5'-GTCAAG-TACTTGCCCTGCTTCCAG-3') and pDAX-1.2 (5'-CAGCATCATATC-ATCCATGCTGAC-3') for pDAX-1, pSRY.3'A (5'-GGAG-AGAGGGCA-CAGAATTT-3') and pSRY.3'dT (as mentioned above) for pSRY, and pGAPDH.A (5'-TCCTGCACCACCAACTGCTTAGC-3') and pGAPDH.2 (5'-ATTGTCGTACCAGGAAATGAGCTTG-3') for pGAPDH. Heminested PCRs were performed with the following primers: pSOX9.B (5'-GATC-TGAAGAAGGAGAGCGAGGA-3') and pSOX9.4 (as described above) for pSOX9, pSF-1.A (as described above) and pSF1.5'2 (5'-AATCTGTGCC-TTCTTTTGCTGCT-3') for pSF-1, pDAX-1.K (5'-TGGCCCAGGACCGC-TTGAACTTT-3') and pDAX-1.2 (as described above) for pDAX-1, pSRY.3'B (5'-GGCAGTACTATGCAGCCAAG-3') and pSRY.3'dT (as described above) for pSRY, and pGAPDH.B (5'-CAAGGTCATCCATGAC-AACTTTGG-3') and pGAPDH.2 (as described above) for pGAPDH. With the exception of SRY, oligos were designed to encompass an intron, allowing the detection of contaminating amplification of genomic DNA by a longer amplified band. The cycling conditions for the PCRs were 45 sec of denaturation at 95°C, 45 sec of annealing at 64°C, and 1 min of elongation at 72°C. Fragments of known size were visualized on a 1% agarose gel with 0.1 mg/ml ethidium bromide and photographed using a gel analysis system (Fotodyne, Hartland, WI).
Primer Extension
The oligo pSRY.5'EXT.4 (5'-GGAATTCTAAATGTGAATACCTCT-G-3') was designed based on sequences approximately 50 bp downstream of a potential transcriptional start site anticipated from a Northern blot analysis of pig fetal gonads (data not shown). This primer was labeled at its 5' end with [
32-P]dATP and T4 polynucleotide kinase (Pharmacia, Baie d'Urfe, PQ, Canada) according to the supplier's instructions. The labeled primer was then purified with a Qiaquick purification column (Qiagen, Valencia, CA). Total RNA from the PGR 9E11 cell line was extracted as described above and poly(A)+ RNA was isolated using magnetic poly(dT) beads (Magnetobeads, Dynal Corp., Great Neck, NY), as previously described [28]. Labeled primer (100 000 cpm) was annealed to 2 µg of poly(A)+ RNA of the PGR 9E11 cell line in hybridization buffer (0.15 M KCl, 10 mM Tris-HCl pH 8.3, 1 mM EDTA) for 10 min at 65°C and for 90 min at 45°C. RNA from the PGR 4G10 cell line was used as a negative control. RT reaction was performed as described above and an RNase A treatment (2 mg/ml) was performed, followed by phenol-chloroform extraction and ethanol precipitation. The pellet was resuspended in 5 µl of stop solution from a T7 sequencing kit (Pharmacia) and loaded on a 6% polyacrylamide-50% urea gel along with a sequencing reaction. The sequencing reaction was performed using the pSRY.5'EXT.4 oligo (as above) as primer and pig 5' flanking SRY sequences as target DNA via the dideoxy method (T7 sequencing kit, Pharmacia).
Deletions and Mutagenesis of pSRY Promoter Sequences
From the SRYp4.5kb-GFP construct, two deletions were generated with restriction enzymes: a XhoI digest gave a 5' flanking fragment of 1426 bp (SRYp1.4kb-GFP), whereas a XhoI and SpeI digest gave a 5' flanking fragment of 906 bp (SRYp906bp-GFP). An EcoRI digest of the SRYp906bp-GFP allowed the generation of SRYp906bp
614bp-GFP where the first 614 bp from the translation start site were deleted from the SRYp900bp-GFP construct. From this SRYp906bp
614bp-GFP construct the SRYp1.4kb and SRYp4.5kb were reconstituted to allow the generation of SRYp1.4kb
614bp-GFP and SRYp4.5kb
614bp-GFP.
Mutagenesis of potential SF-1 binding sites at -1369 bp and -290 bp in the pig SRY promoter was performed via PCR amplification of specific plasmids, using one oligo carrying the mutation and the other oligo positioned head-to-head respective to their 5' phosphorylated ends. PCR amplifications of 25 cycles were performed using Expand High Fidelity DNA polymerase (Roche). The amplified fragment of interest was recovered from a 1% agarose gel and digested with DpnI restriction enzyme to remove the PCR template. The PCR product was then treated with Pfu DNA polymerase for 15 min at 72°C to generate blunt ends, which were necessary for self-ligation. The oligos used were as follows (the mutated sequences are underlined): pSRYpSF-1(-1369bp).mut.A 5'-TGAGTTTAGCTAGCGTCTGTTTTTCTCTTAATG-3' and pSRYpSF-1(-1369bp).mut.1 5'-AAGTTT-CCCGAGGTGAACATAACA-3' for the potential SF-1 binding site at -1369 bp, and pSRYpSF-1(-290bp).mut.B (5'-CCGGchTTTTCGTACGCGCAGAGCCTTCAGCAACT-3') and pSRYpSF-1(-290bp).mut.1 (5'-AAACTTTGTCATTATTAAGTTAT-3') for the potential SF-1 binding site at -290 bp.
Transfections in PGR 9E11 and CV-1 Cells
Plasmid DNA used for transfection studies was purified by ultracentrifugations on a CsCl gradient [27]. Plasmid DNA concentrations were calculated from optical density measurements and verified by comparing them to DNA standards of known concentration on a 1% agarose gel. All transient transfections were performed in 24-well culture plates using Effectene transfection reagent (Qiagen) according to the manufacturer's instructions. PGR 9E11 cells and Green monkey kidney fibroblast (CV-1) cells were grown in DMEM supplemented with 2.5% newborn calf serum.
For PGR 9E11 cells, 2.25 x 105 cells were plated per well the day before transfection. SRYp-GFP constructs (with deletions, mutations, or both) were transfected at 200 ng per well and DNA was left to remain in contact with the cells for 6 h. Media was then changed and the second morning following the day of transfection (at approximately 36 h), the cells were harvested. Quantification of green fluorescence protein (GFP) fluorescence was performed by analyzing 1 x 104 cells for the percentage of fluorescent cells within the cell population as well as for average fluorescence for the cell population, using a Coulter Epics XL FACS machine (Beckman Coulter, Krefeld, Germany) equipped with EGFP filters (Omega Optical). Background fluorescence was determined by transfection of pEGFP-1 vector without insert (200 ng), and the result was subtracted from each test measurement. Transfections were performed four separate times in duplicate. Moreover, for the 5' deletion constructs, fluorescence data were corrected for the quantity of DNA, in pmol, contained in 200 ng.
For transient transfections into CV-1 cells, 9 x 10 4 cells were plated per well prior to transfection. To observe transactivation of the pig SRY 5' flanking sequences, the SRYp4.5kb-GFP construct (100 ng) was cotransfected with increasing concentrations (0, 100, 250, 500, and 750 ng per well) of a plasmid containing the coding sequences of pig SF-1 (GenBank accession number U84399) under the control of CMVp (pBK-CMVp-SF-1 ORF). To further analyze the transactivation effect of SF-1 on the SRYp4.5kb-GFP construct, each of the SF-1 potential binding sites was mutated. In each case, 750 ng of expression vector was cotransfected with 100 ng of the SRYp4.5kb-GFP constructs. The total amount of transfected DNA was kept constant using pBK-CMV plasmid without insert. DNA was left to remain in contact with the cells for 18 h. Harvesting of the cells and quantification of GFP fluorescence was performed as described above.
DNA Binding Assays
DNA binding assays were performed using 0.5 ml of in vitro-translated mouse SF-1 protein that was prepared according to the manufacturer's instructions (TnT kit; Promega). Proteins were prepared as described in [29], except that 100 ng of dI:dC was used when working with in vitro-translated proteins. DNA binding assays were performed using a 32P-labeled double-stranded oligonucleotide corresponding to the consensus SF-1 element from the proximal MIS promoter (sense oligo, 5'-GATCCCCCAAGGTCACCTTTA-3'; antisense oligo, 5'-GATCTAAAGGTGACCTTGGGG-3'). Binding reactions and electrophoresis conditions were those as previously described [29]. A series of double-stranded, wild-type, and mutated oligonucleotides was used to confirm the specificity of SF-1 binding to the proximal (-290 bp) and distal (-1369 bp) SF-1 sites in the pig SRY promoter as follows: MISmut sense, 5'-GCCAGGCACTGTCCCCCAATTTCACCTTTGGTGTTGATAGG-3' and antisense, 5'-CCTATCAACACCAAAGGTGAAATTGGGGGACAGTGCCTGGC-3'; SRY-290bp sense, 5'-GATCCGGTTTTTTAAGGTTAGCAGAGCCTTGCATGCA-3'; and antisense, 5'-GATCTGCATGCAAGGCTCTGCTAACCTTAAAAAACCG-3'; SRY-290bpmut sense, 5'-GATCCGGTTTTTTCGTACGCGCAGAGCCTTGCATGCA-3' and antisense, 5'-GATCTGCATGCAAGGCTCTGCGCGTACGAAAAAACCG-3'; SRY-1369bp sense, 5'-GATCTTGAGTTCCAAGGTTATCTGTTT-3' and antisense, 5' GATCAAACAGATAACCTTGGAACTCAA-3'; and SRY-1369bpmut sense, 5'-GATCTTGAGTTTACGTACGGTCTGTTT-3' and antisense, 5'-GATCAAACAGACCGTACGTAAACTCAA-3'. The DNA-protein complexes were resolved on a 5% polyacrylamide gel in Tris-borate-EDTA buffer and visualized by autoradiography after drying the gel.
| RESULTS |
|---|
|
|
|---|
Comparisons of the 4.5 kb of pig SRY 5' flanking sequences with available human, mouse, and sheep sequences were performed. As shown schematically in Figure 2, comparison of pig and sheep sequences revealed numerous blocks of moderate homology varying from 53% to 80%, not always at a conserved position but displaying a conserved order. The comparison of pig and human SRY promoter sequences similarly revealed five blocks of moderate homology varying between 56% and 70%, again displaying a conserved order. To simplify the presentation, mouse sequences were compared with those of human sequences; only two small regions of homology were identified. The nucleotide sequence of pig SRY 5' flank from -1427 to the ATG is presented in Figure 3A, with potential transcriptional factor binding sites and the transcriptional initiation site marked. Restriction sites used in 5' deletion studies are underlined, including XhoI (at -1.4 kb) and SpeI (at 0.9 kb). Complete sequences for the pig 6.4 kb genomic HindIII fragment containing the SRY ORF and including 4.5 kb of 5' flanking sequences are available in GenBank (accession number U49860). Figure 3B depicts the consensus and potential binding sites for SF-1, as well as the mutated form used in subsequent transfection experiments.
|
|
Porcine Genital Ridge Cell Lines and SRY Transcriptional Initiation
To study the transcriptional activity of pig SRY 5' flanking sequences in vitro, PGR cell lines were generated. Genital ridges of pig embryos were dissected at 2223 dpc, during the time of sex determination and when SRY expression is at its maximum [11]. Four cell lines were obtained by this method and were characterized for chromosomal sex by Southern blot analysis for SRY sequences (data not shown). By RT-PCR transcriptional analysis, one XY cell line (PGR 9E11) showed expression for SRY as well as SOX9, SF-1, and DAX-1 (Fig. 4A). Expression of these genes in PGR 9E11 cells is consistent with these cells being of pre-Sertoli cell phenotype.
|
RNA of the PGR 9E11 cell line was used to perform a primer extension analysis of the 5' untranslated region (UTR) of the SRY transcript. The primer used for the primer extension was based on an estimated transcript size of 2.1 kb, as calculated from a Northern blot analysis of mRNA taken from fetal testes of 612 wk gestation (data not shown). As shown in Figure 4B, primer extension with PGR 9E11 cell line mRNA as a template allowed the identification of a transcription start site at 661 bp upstream of the translation start site (i.e., at -661 bp). A chromosomal XX PGR cell line (PGR 4G10) was used as a negative control.
Deletion Studies of the Pig SRY Promoter in PGR 9E11 Cells
Because the PGR 9E11 cell line expresses endogenous SRY, it should express the factors necessary for SRY expression. Therefore, the PGR 9E11 cell line was used to perform a 5' deletion study for transcriptional activity of the pig SRY 5' flanking sequences via transient transfections and using GFP as a reporter gene (Fig. 5A). For the sake of comparison, the -4.5 kb of SRY 5' flanking sequences were set to 100% expression. The deletion of approximately 3 kb to obtain the SRYp1.4kb construct did not compromise the level of GFP expression. However, when SRY 5' flanking sequences were reduced to -906 bp (SRYp906bp) the GFP expression decreased by approximately 40%. The deletion of the 5' UTR from the SRYp906pb (SRYp906bp
614bp), from the SRYp1.4kb (SRYp1.4kb
614bp), as well as from the SRYp4.5kb (SRYp4.5kb
614bp) constructs completely inhibited the activity of these constructs.
|
SF-1 Potential Binding Sites in the Pig SRY Promoter and Their Mutation
A computer search for potential binding sites for known transcription factors in the fully active -1.4 kb 5' flanking fragment resulted in the identification of two potential SF-1 binding sites (-1369 bp and -290 bp) as well as one potential SOX9 binding site (-205 bp) and one potential SP1 binding site (-128 bp) (Fig. 3A). The potential binding sites for SF-1 were mutated (Fig. 3B) within the full-length SRYp4.5kb-GFP construct, and results of transfections with these mutated constructs transfected into PGR 9E11 cells are shown in Figure 5B. The mutation of the SF-1 potential binding site at -290 bp resulted in a reduction of at least 65% in SRY promoter activity compared with that of the SRYp4.5kb-GFP nonmutated construct, whereas the mutation of the SF-1 potential binding site at -1369 bp resulted in a complete loss of promoter activity.
SF-1 Transactivates the Pig SRY Promoter in Mouse CV-1 Cells
Figure 5C demonstrates a transactivation effect for SF-1 in a heterologous in vitro system involving mouse CV-1 cells. In these studies, pig coding sequences were used for SF-1, and resulted in a 6-fold transactivation effect on the pig SRY -4.5 kb 5' flanking sequences. Mutated SF-1 potential binding sites within the SRYp4.5kb-GFP construct were then tested in cotransfection studies using the SF-1 coding sequences, again in CV-1 cells. As shown in Figure 5D, the SF-1 transactivation effect was reduced by approximately 33% when the SF-1 (-290 bp) potential binding site was mutated, whereas a transactivation effect was eliminated when the SF-1 (-1369 bp) potential binding site was mutated.
Pig SRY Promoter Contains a Functional SF-1 Binding Element
DNA binding assays (electrophoretic mobility shift assays) using the two SF-1 potential binding sites as probe demonstrated that SF-1 can bind the SF-1 (-1369 bp) (Fig. 6) potential binding site. Similar binding was not evident for the SF-1 (-290 bp) potential binding site. These results are consistent with the fact that the SF-1 (-1369 bp) potential binding site retains closer homology to the consensus SF-1 binding site [30] than the SF-1 (-290 bp) site (Fig. 3B).
|
| DISCUSSION |
|---|
|
|
|---|
Transient transfection studies in PGR cell lines using 5' deletions of the pig 5' flanking SRY sequences suggested that the regulatory elements required for basic SRY promoter activity are present within -1.4 kb of pig SRY 5' flanking sequences (where the ATG translational start codon is +1). We have identified a transcriptional start site at -661 bp, and furthermore shown that sequences downstream of this site (i.e., transcribed but untranslated 5' flanking sequences) are absolutely required for full promoter activity. Thus basal promoter activity was associated with sequences 766 nucleotides upstream and 661 nucleotides downstream of the identified transcriptional start site at -661 bp. As seen in Figure 3A for the pig SRY 5' flanking region, there is an additional TATA element at -746 bp from the ATG (with a CAAT element 22 bp further upstream); an additional TATA element is found immediately after the transcriptional initiation site at -661 nucleotides, whereas an additional TATA element is found at -516 bp from the translational initiation site (with a CAAT element 55 bp upstream). Multiple transcription start sites have been proposed for human SRY [1517]. Although primer extension analysis of pig SRY 5' flanking sequences using oligos to target the TATA element at -516 bp failed to reveal a functional transcription start site, it remains possible that multiple transcription start sites exist for the pig SRY promoter.
Previously, we reported a transactivation effect of SOX9 on the pig SRY -4.5 kb 5' flanking sequences in transient cotransfection assays in a heterologous in vitro system involving mouse CV-1 cells, and have postulated that at least in some species, SOX9 may be active in initiating and also maintaining SRY expression profiles [25]. We now extend these studies to consider the role of SF-1 in sex determination. SF-1 is a DNA binding protein belonging the nuclear hormone receptor family of transcriptional factors, and is required for the formation of the bipotential gonads [34]. In our current work we provide evidence for two potential binding sites for SF-1 within the pig SRY promoter, one located upstream of the site of transcriptional initiation (at -1369 bp from the site of translational initiation), and another located downstream of the site of transcriptional initiation (at -290 bp from the site of translational initiation). Structurally, the upstream site retains the highest homology with published SF-1 consensus binding sites [30]. It is interesting that mutation of the SF-1 (-1369 bp) potential binding site within the context of the -4.5 kb fragment essentially eliminated promoter activity, whereas eliminating the same site via 5' deletions resulted in a more modest reduction in transcriptional activity. This may indicate that sequences upstream of -1.4 kb are important for modulation of transcriptional activity. Structural homologies noted upstream of -1.4 kb between pig, sheep, and human 5' flanking sequences (Fig. 2) would support this suggestion.
De Santa Barbara et al. [24] have reported the presence of an SF-1 binding site at -315 of the human SRY promoter. SF-1 was able to generate a modest transactivation effect via this site. Sequence comparisons show that the -315 site in the human SRY promoter sequence is not equivalent to either the -1369 or the -205 sites within the pig SRY promoter sequence. These authors further report that the transactivation of the human SRY promoter by SF-1 depends on the phosphorylation status of SF-1, such that phosphorylated SF-1 inhibits SRY promoter activation. This inhibition requires an interaction between SF-1 and SP1 proteins. Two SP1 binding sites are reported within the human SRY promoter, at -150 and -130 [19]. Within the pig SRY promoter sequence, there is a potential SP1 binding site at -128 bp that is homologous to the -130 bp site within the human SRY promoter sequences.
Currently, using either human or pig SRY sequences, WT-1, SF-1, and SOX9 have been shown to transactivate the SRY promoter [22, 24, 25]. SOX9 and SF-1 are in turn up-regulated in a testes-specific fashion following SRY expression, and may in turn (depending on phosphorylation status, as in SF-1) contribute to the inhibition of SRY expression. Further comparative studies are necessary to test the generalities of these observations and to identify additional components of the transcriptional control of SRY expression.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Correspondence: FAX: 450 778 8103; silverdw{at}medvet.umontreal.ca ![]()
3 Current address: Institut de recherches cliniques de Montréal, Montreal, Québec, Canada H2W 1R7 ![]()
Received: 30 August 2002.
First decision: 24 September 2002.
Accepted: 11 October 2002.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Kothapalli, E. Kirkness, S. Pujar, R. Van Wormer, and V. N. Meyers-Wallen Exclusion of Candidate Genes for Canine SRY-Negative XX Sex Reversal J. Hered., November 1, 2005; 96(7): 759 - 763. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Dufresne, N. St-Pierre, R. S. Viger, L. Hermo, and D. G. Cyr Characterization of a Novel Rat Epididymal Cell Line to Study Epididymal Function Endocrinology, November 1, 2005; 146(11): 4710 - 4720. [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] |
||||
![]() |
A. Boyer, J. G. Lussier, A. H. Sinclair, P. J. McClive, and D. W. Silversides Pre-Sertoli Specific Gene Expression Profiling Reveals Differential Expression of Ppt1 and Brd3 Genes Within the Mouse Genital Ridge at the Time of Sex Determination Biol Reprod, September 1, 2004; 71(3): 820 - 827. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-C. Chaboissier, A. Kobayashi, V. I. P. Vidal, S. Lutzkendorf, H. J. G. van de Kant, M. Wegner, D. G. de Rooij, R. R. Behringer, and A. Schedl Functional analysis of Sox8 and Sox9 during sex determination in the mouse Development, May 1, 2004; 131(9): 1891 - 1901. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |