The Transcription Factors Steroidogenic Factor-1 and SOX9 Regulate Expression of Vanin-1 during Mouse Testis Development
2004; Elsevier BV; Volume: 280; Issue: 7 Linguagem: Inglês
10.1074/jbc.m412806200
ISSN1083-351X
AutoresMegan J. Wilson, Pancharatnam Jeyasuria, Keith L. Parker, Peter Koopman,
Tópico(s)Sexual Differentiation and Disorders
ResumoWe previously showed, using differential expression screening and in situ hybridization that Vanin-1, which encodes a glycosylphosphatidylinositol-linked membrane-associated pantetheinase, is expressed in a sex-specific manner during fetal gonad development in mice (Bowles, J., Bullejos, M., and Koopman, P. (2000) Genesis 27, 124–135). In the present study we investigate in detail the expression and regulation of Vanin-1 in the fetal testis. Vanin-1 is co-expressed with the transcription factors steroidogenic factor-1 (SF-1) and SOX9 in Sertoli cells and, at a lower level, with SF-1 in Leydig cells in developing testes. SF-1 is able to activate the transcription of the Vanin-1 promoter in in vitro reporter assays, and this activation is further augmented by SOX9. We found that SF-1 is able to bind to two sites in the Vanin-1 promoter, whereas SOX9 can bind to a single interposed site defined by DNA footprinting. Mutation of the SF-1 or SOX9 sites disrupts the binding of these factors and activation of transcription. The expression of Vanin-1 was abolished in Leydig cells of a mouse mutant lacking SF-1 in that cell type. Our findings account for the sex- and cell-type-specific expression of Vanin-1 in the developing mouse gonad in vivo, which we suggest is required to provide an appropriate environment for male germ cell development. We previously showed, using differential expression screening and in situ hybridization that Vanin-1, which encodes a glycosylphosphatidylinositol-linked membrane-associated pantetheinase, is expressed in a sex-specific manner during fetal gonad development in mice (Bowles, J., Bullejos, M., and Koopman, P. (2000) Genesis 27, 124–135). In the present study we investigate in detail the expression and regulation of Vanin-1 in the fetal testis. Vanin-1 is co-expressed with the transcription factors steroidogenic factor-1 (SF-1) and SOX9 in Sertoli cells and, at a lower level, with SF-1 in Leydig cells in developing testes. SF-1 is able to activate the transcription of the Vanin-1 promoter in in vitro reporter assays, and this activation is further augmented by SOX9. We found that SF-1 is able to bind to two sites in the Vanin-1 promoter, whereas SOX9 can bind to a single interposed site defined by DNA footprinting. Mutation of the SF-1 or SOX9 sites disrupts the binding of these factors and activation of transcription. The expression of Vanin-1 was abolished in Leydig cells of a mouse mutant lacking SF-1 in that cell type. Our findings account for the sex- and cell-type-specific expression of Vanin-1 in the developing mouse gonad in vivo, which we suggest is required to provide an appropriate environment for male germ cell development. Male and female gonads, although structurally and functionally quite distinct tissues, arise in the embryo from the same tissue primordia, the genital ridges. Their developmental trajectories begin to diverge about 10.5 days post coitum (dpc) 1The abbreviations used are: dpc, days post coitum; AMH, anti-Müllerian hormone; SF-1, steroidogenic factor 1; GST, glutathione S-transferase; EMSA, electromobility shift assay; PBS, phosphate-buffered saline. in mice, when Sry expression in XY genital ridges initiates testis determination. Many genes that operate downstream of Sry have been identified as being expressed male-specifically during sex determination and gonad differentiation, but little is known about how they are regulated, interact, and function within the overall scheme of sex differentiation. SRY is the eponymous founding member of the Sox (SRY-related HMG box) gene family, encoding a group of proteins that bind to DNA in a sequence-specific manner via their HMG (high mobility group) domain (1Bowles J. Schepers G. Koopman P. Dev. Biol. 2000; 227: 239-255Crossref PubMed Scopus (760) Google Scholar). Immediately following Sry induction the expression of SOX9, another member of the SOX family, is up-regulated in male genital ridges. Transgenic mouse studies have shown that expression of SOX9 is sufficient for testis formation (2Vidal V.P. Chaboissier M.C. de Rooij D.G. Schedl A. Nat. Genet. 2001; 28: 216-217Crossref PubMed Scopus (562) Google Scholar), whereas mutations in human SOX9 can result in XY sex reversal (3Foster J.W. Dominguez-Steglich M.A. Guioli S. Kwok C. Weller P.A. Weissenbach J. Mansour S. Young I.D. Goodfellow P.N. Brook J.D. Schafer A.J. Nature. 1994; 372: 525-530Crossref PubMed Scopus (1332) Google Scholar). These observations show that SOX9 is a key transcription factor in the male sex determination pathway. SOX9 plays an essential role in regulating the gene encoding anti-Müllerian hormone (AMH), a key hormone required for the regression of the female duct system in male embryos. A homozygous mutation of the SOX9 binding site within this promoter in male mice ablates the expression of Amh, resulting in the retention of the Müllerian ducts and the generation of pseudohermaphrodites (4Arango N. Lovell-Badge R. Behringer R. Cell. 1999; 99: 409-419Abstract Full Text Full Text PDF PubMed Scopus (374) Google Scholar). In conjunction with SOX9, steroidogenic factor 1 (SF-1) is also required to modulate levels of Amh expression (4Arango N. Lovell-Badge R. Behringer R. Cell. 1999; 99: 409-419Abstract Full Text Full Text PDF PubMed Scopus (374) Google Scholar, 5de Santa Barbara P. Moniot B. Poulat F. Boizet B. Berta P. J. Biol. Chem. 1998; 273: 29654-29660Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar). SF-1 (also known as Ad4BP or NR5A1) is an orphan nuclear receptor required for gonad and adrenal gland development that is expressed in the Leydig and Sertoli cells of the testis and in the adrenal cortex (6Hatano O. Takayama K. Imai T. Waterman M.R. Takakusu A. Omura T. Morohashi K. Development. 1994; 120: 2787-2797Crossref PubMed Google Scholar, 7Ikeda Y. Shen W.H. Ingraham H.A. Parker K.L. Mol. Endocrinol. 1994; 8: 654-662Crossref PubMed Scopus (551) Google Scholar). Mice homozygous for a null mutation of SF-1 fail to develop adrenal glands and gonads, establishing SF-1 as an essential regulator required for their development (8Luo X. Ikeda Y. Parker K.L. Cell. 1994; 77: 481-490Abstract Full Text PDF PubMed Scopus (1397) Google Scholar). Many potential target genes for SF-1 have been determined, including several genes involved in steroidogenesis in the adrenal gland and Leydig cell lineage of the testis. Two of these targets encode proteins required for the rate-limiting steps in testosterone synthesis: steroidogenic acute regulatory (StAR) protein, which regulates cholesterol uptake into the mitochondria, and cholesterol side chain cleavage enzyme, which catalyzes the first cleavage reaction in steroid biosynthesis (9Reinhart A.J. Willaims S.C. Clark B.J. Stocco D.M. Mol. Endocrinol. 1999; 13: 729-741PubMed Google Scholar). In vitro studies suggest that the HMG domain of SOX9 interacts directly with the C-terminal domain of SF-1 to cooperatively activate expression from the Amh promoter in Sertoli cells (10de Santa Barbara P. Bonneaud N. Boizet B. Desclozeaux M. Moniot B. Su ̈dbeck P. Scherer G. Poulat F. Berta P. Mol. Cell. Biol. 1998; 18: 6653-6665Crossref PubMed Scopus (520) Google Scholar). Recently, we demonstrated that a related SOX factor expressed by Sertoli cells, SOX8, can synergize with SF-1 in a similar fashion (11Schepers G. Wilson M. Wilhelm D. Koopman P. J. Biol. Chem. 2003; 278: 28101-28108Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). There is further evidence to suggest that SOX9 may be responsible for the male-specific expression of SF-1, as it can bind to the proximal promoter of SF-1 and up-regulate expression in vitro (12Shen J.H.-C. Ingraham H.A. Mol. Endocrinol. 2002; 16: 529-540Crossref PubMed Scopus (80) Google Scholar). Although a large body of data indicates that SF-1 and SOX9 are critical transcription factors required for embryonic testis differentiation and are likely to cooperate in several steps of this process, only Amh has been identified as a direct target for these proteins. Vanin-1 has been identified as being expressed in a male-enriched fashion in the embryonic gonad by both subtractive hybridization (13Bowles J. Bullejos M. Koopman P. Genesis. 2000; 27: 124-135Crossref PubMed Google Scholar) and microarray screening (14Grimmond S. Van Hateren N. Siggers P. Arkell R. Larder R. Soares M. Bonaldo M. Smith L. Tymowska-Lalanne Z. Wells C. Greenfield A. Hum. Mol. Genet. 2000; 9: 1553-1560Crossref PubMed Scopus (96) Google Scholar). Whole mount in situ hybridization of XY embryonic gonads has shown that Vanin-1 is expressed immediately following Sry expression, just before the formation of the testis cords. Vanin-1 is a member of a protein family of consisting of at least three members in humans (Vanin-1, -2, and -3) and two members in mice (Vanin-1 and -3) (15Martin F. Malergue F. Pitari G. Philippe J.M. Philips S. Chabret C. Granjeaud S. Mattei M.G. Mungall A.J. Naquet P. Galland F. Immunogenetics. 2001; 53 (M. F.): 296-306Crossref PubMed Scopus (60) Google Scholar). The Vanin-1 gene product is a glycosylphosphatidylinositol-linked membrane-associated pantetheinase, a family of enzymes that catalyzes the hydrolysis of pantetheine into pantothenic acid (vitamin B5) and cysteamine, an anti-oxidant (16Maras B. Barra D. Dupre S. Pitari G. FEBS Lett. 1999; 461: 149-152Crossref PubMed Scopus (64) Google Scholar, 17Pitari G. Malergue F. Martin F. Philippe J. Massucci M. Chabret C. Maras B. Dupre ̀ S. Naquet P. Galland F. FEBS Lett. 2000; 483: 149-154Crossref PubMed Scopus (173) Google Scholar). Vanin-1 null mice develop normally, but tissues that would typically express this protein lack cysteamine and exhibit a modified stress response (17Pitari G. Malergue F. Martin F. Philippe J. Massucci M. Chabret C. Maras B. Dupre ̀ S. Naquet P. Galland F. FEBS Lett. 2000; 483: 149-154Crossref PubMed Scopus (173) Google Scholar, 18Martin F. Penet M.F. Malergue F. Lepidi H. Dessein A. Galland F. de Reggi M. Naquet P. Gharib B. J. Clin. Investig. 2004; 113: 591-597Crossref PubMed Scopus (95) Google Scholar, 19Berruyer C. Martin F.M. Castellano R. Macone A. Malergue F. Garrido-Urbani S. Millet V. Imbert J. Dupre S. Pitari G. Naquet P. Galland F. Mol. Cell. Biol. 2004; 24: 7214-7224Crossref PubMed Scopus (141) Google Scholar). Recently, it has been demonstrated that expression from the Vanin-1 proximal promoter is up-regulated in response to oxidative stress (19Berruyer C. Martin F.M. Castellano R. Macone A. Malergue F. Garrido-Urbani S. Millet V. Imbert J. Dupre S. Pitari G. Naquet P. Galland F. Mol. Cell. Biol. 2004; 24: 7214-7224Crossref PubMed Scopus (141) Google Scholar), further suggesting that Vanin-1 is likely to play an important role in tissue response to stress and inflammation. To identify proteins involved in regulating expression from the Vanin-1 promoter in the developing testis, immunofluorescence studies were carried out on cryosectioned embryonic testes. Vanin-1 is expressed by cells that are positive for SF-1 and is strongly expressed in Sertoli cells that express both SOX9 and SF-1. The proximal Vanin-1 promoter was isolated and found to contain putative binding sites for both SF-1 and SOX9. Electrophoretic mobility shift assays and cell transfection assays showed that SF-1 and SOX9 synergistically activate transcription from this promoter. PCR and Plasmid Constructions—The oligonucleotide primers 5′-GATTCCTGTCATAACCTC-3′ and 5′-CATGCTGAAGTCCAAAGA-3′ were used to amplify the Vanin-1 promoter fragment (267 bp) from mouse genomic DNA by PCR. The resulting fragment was cloned into pGEM-T Easy (Promega), sequenced, and then subcloned into pGL2-Basic (Promega) using the KpnI and SacI restrictions sites. Expression constructs for SOX9, SOX8, and SF-1 and GST fusion constructs were reported previously (11Schepers G. Wilson M. Wilhelm D. Koopman P. J. Biol. Chem. 2003; 278: 28101-28108Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). The open reading frame of Sry was amplified by PCR using the oligonucleotide pair 5′-GCGGGATCCATGGAGGGCCATGTCAAG-3′ and 5′-GCGGTCGACCTATGAGACTGCCAACCACA-3′ and cloned into pGEM-T easy. A BamHI and NotI restriction fragment from the resulting plasmid was then subcloned into the pGEX4T-3 expression vector (Amersham Biosciences) in-frame with the GST coding region and confirmed by DNA sequencing. pcDNA3-SOX9ΔTA was generated to lack the last 60 amino acids of SOX9, which includes the activation domain (21Ng L.-J. Wheatley S. Muscat G.E.O. Conway-Campbell J. Bowles J. Wright E. Bell D.M. Tam P.P.L. Cheah K.S.E. Koopman P. Dev. Biol. 1997; 183: 108-121Crossref PubMed Scopus (568) Google Scholar). Mutations were introduced into the wild-type promoter using the QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's protocol. The sequence of the top strand oligonucleotides used to generate each mutation was as follows, with the mutated nucleotides in bold, SF-M1 5′-CTGACCAGAAACTAGGGATATTCTGAAACTGTTG-3′, SF-M2 5′-TTTTTTGAAGCTACCCAGATCTGGGAACCACGTG-3′, and MutSOX 5′-CGATAAGTAACCTCTAGTCCTGTCTGCATTCCTG-3′. The presence of introduced mutations was confirmed by DNA sequencing. Electrophoretic Mobility Shift Assay—EMSA analysis was performed as described previously (11Schepers G. Wilson M. Wilhelm D. Koopman P. J. Biol. Chem. 2003; 278: 28101-28108Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). Briefly, KpnI-SacI restriction fragments from pGL2-Vnn expression constructs were purified using the Qiagen Gel Purification Kit. Following dephosphorylation using alkaline phosphatase, fragments were end-labeled with 32γ-ATP and T4 polynucleotide kinase. Probes were purified on a Nick Purification Column (Amersham Biosciences) and eluted in TE buffer (50 mm Tris-Cl, 1 mm EDTA, pH 8). DNase I Footprinting—The procedure for DNase I footprinting has been described previously (22Wilhelm D. Englert C. Genes Dev. 2002; 16: 1839-1851Crossref PubMed Scopus (231) Google Scholar). The Vanin-1 promoter fragment was labeled on either strand by PCR by using one 32P end-labeled oligonucleotide primer (for the strand to be labeled) along with the non-labeled primer. Footprinting reactions were electrophoresed in parallel with sequencing reactions to determine the sequence of the protected site. Sequencing reactions were performed using the Thermo Sequenase radiolabeled terminator cycle sequencing kit (Amersham Biosciences) using the oligonucleotide that was labeled in the PCR. Luciferase Assays—TM3 cells (23Mather J.P. Biol. Reprod. 1980; 23: 243-252Crossref PubMed Scopus (334) Google Scholar) were plated in 12-well plates and transfected with 500 ng of luciferase reporter plasmid with either empty expression vector (pcDNA3) or SF-1 (50 or 20 ng) and/or SOX9 (20 ng) expression plasmids using Lipofectamine™ 2000 (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were washed twice with PBS and assayed for luciferase activity with the Luciferase Reporter Gene assay kit (Roche Applied Science). Each transfection was done in duplicate and repeated independently at least three times; data represent the average -fold increase relative to cells co-transfected with empty expression vector (pcDNA3). Immunofluorescence—Antibody staining was performed on cryosectioned embryonic tissue with the following antibodies. Vanin-1 antibody (1:100 dilution) was purified from the rat hybridoma clone H202-407-6-3 as described previously (24Aurrand-Lions M. Galland F. Bazin H. Zakharyev V. Imhof B.A. Naquet P. Immunity. 1996; 5: 391-405Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). Rabbit anti-SF-1 antibody (1:1000 dilution) was a kind gift from K. Morohashi (National Institute for Basic Biology, Okazaki, Japan). Goat anti-AMH antibody (sc-6886) was purchased from Santa Cruz Biotechnology and used as a 1:100 dilution. Rabbit anti-SOX9 antibody has been described previously (25Kent J. Wheatley S.C. Andrews J.E. Sinclair A.H. Koopman P. Development. 1996; 122: 2813-2822Crossref PubMed Google Scholar). Secondary antibodies (anti-rat Alexa 488, anti-rabbit Alexa 594, and anti-goat Alexa 594) were purchased from Molecular Probes and used at a 1:200 dilution. Tissue samples were fixed with 4% paraformaldehyde in PBS for 1 h, washed with PBS, and then incubated overnight in PBS containing 30% sucrose. Samples were then frozen in OCT cryo-embedding compound and sectioned at a 10-μm thickness. Slides were washed three times with PBTx (PBS containing 0.1% Triton X-100) and blocked with 10% heat-inactivated horse serum. Slides were incubated overnight at 4 °C with the primary antibody diluted in blocking solution as indicated above. After three washes with PBTx, slides were incubated with the appropriate secondary antibody, washed, mounted, and imaged using a Bio-Rad Radiance 2000 MP confocal microscope. Gonad-specific SF-1 null mice were bred and genotyped as described previously (26Jeyasuria P. Ikeda Y. Jamin S.P. Zhao L. de Rooij D.G. Themmen A.P.N. Behringer R.R. Parker K.L. Mol. Endocrinol. 2004; 18: 1610-1619Crossref PubMed Scopus (219) Google Scholar). Embryos were isolated and prepared as described above for sectioning. SF-1 and SOX9 Are Co-expressed with Vanin-1 in Vivo— Previously it has been shown by in situ hybridization on whole gonads that Vanin-1 mRNA is expressed predominantly in Sertoli cells of the testis cords of mouse XY fetal gonads at 13.5 dpc (13Bowles J. Bullejos M. Koopman P. Genesis. 2000; 27: 124-135Crossref PubMed Google Scholar, 14Grimmond S. Van Hateren N. Siggers P. Arkell R. Larder R. Soares M. Bonaldo M. Smith L. Tymowska-Lalanne Z. Wells C. Greenfield A. Hum. Mol. Genet. 2000; 9: 1553-1560Crossref PubMed Scopus (96) Google Scholar). In the present study, immunofluorescence was carried out to further characterize the expression of Vanin-1 in the embryonic testis in terms of cell type and possible coexpression with transcription factors present in the Sertoli cells, SOX9 and SF-1 in vivo. Mouse fetal testes at 13.5 and 16.5 dpc were sectioned and stained with an antibody specific for Vanin-1 (24Aurrand-Lions M. Galland F. Bazin H. Zakharyev V. Imhof B.A. Naquet P. Immunity. 1996; 5: 391-405Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar), together with anti-SF-1, -SOX9, or -AMH antibodies (Fig. 1). Vanin-1 was found to be strongly expressed in the cytoplasm of Sertoli cells at these stages, as evidenced by colocalization with AMH (Fig. 1A) and SOX9 (Fig. 1C). Some staining was also observed in interstitial cells located between the cords. These interstitial cells were identified as Leydig cells by positive staining for SF-1 (6Hatano O. Takayama K. Imai T. Waterman M.R. Takakusu A. Omura T. Morohashi K. Development. 1994; 120: 2787-2797Crossref PubMed Google Scholar) (Fig. 1B). These results show that Vanin-1 protein is expressed in Leydig cells as well as Sertoli cells. SF-1 is known to be expressed strongly in the adrenal cortex at 13.5 dpc (7Ikeda Y. Shen W.H. Ingraham H.A. Parker K.L. Mol. Endocrinol. 1994; 8: 654-662Crossref PubMed Scopus (551) Google Scholar). Vanin-1 staining was found on the surface of SF-1-positive cells within the adrenal cortex (Fig. 2). Antibody staining was weaker compared with the Vanin-1 staining found in the testis cords (Fig. 2). Together these results indicate that cells positive for SF-1 also express Vanin-1 and that there may be stronger expression of Vanin-1 in cells that also express SOX9 in vivo, consistent with the possibility that SOX9 and SF-1 may together be involved in the regulation of Vanin-1 transcription. SF-1 and SOX9 Can Bind to the Vanin-1 Proximal Promoter— To investigate whether SF-1 and SOX9 can activate transcription from the proximal promoter of Vanin-1, a 267-bp fragment of the promoter was cloned by PCR using mouse genomic DNA. A sequence analysis revealed two potential SF-1 binding sites based on the consensus binding sequence previously determined for SF-1 (27Morohashi K. Honda S. Inomata Y. Handa H. Omura T. J. Biol. Chem. 1992; 267: 17913-17919Abstract Full Text PDF PubMed Google Scholar) (Fig. 3A). To test whether SF-1 is able to bind to this promoter fragment, we performed EMSA using a GST·SF-1 fusion protein. The Vanin-1 promoter probe formed DNA·protein complexes with GST·SF-1, but not with GST alone (Fig. 3B, arrowheads), confirming the presence of SF-1 binding sites within this promoter fragment. To test for binding specificity, each site was mutated and tested by EMSA for the ability to bind GST·SF-1. A mutation of either site individually produced a minor reduction of SF-1·DNA complex formation, but when both sites were mutated in the same fragment the formation of a specific SF-1·DNA complex was abolished (Fig. 3B). Because SF-1 and SOX9 cooperatively act to up-regulate Amh promoter expression, and in view of our observations that Vanin-1 expression is stronger in the testis cords where SOX9 is expressed compared with Leydig cells or the adrenal where SOX9 is not expressed (Fig. 2), the ability of SOX9 to bind to the Vanin-1 promoter was investigated. Purified GST·SOX9 fusion protein was able to bind to the isolated promoter fragment in EMSA (Fig. 3C) confirming the presence of a SOX site(s) within the promoter. In addition to SOX9, Sertoli cells also express SRY (28Albrecht K. Eicher E. Dev. Biol. 2001; 240: 92-107Crossref PubMed Scopus (307) Google Scholar) and SOX8, a protein that shares 53% identity with SOX9 (11Schepers G. Wilson M. Wilhelm D. Koopman P. J. Biol. Chem. 2003; 278: 28101-28108Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar). As with SOX9, SOX8 has also been shown to cooperate with SF-1 to activate transcription from the Amh promoter (11Schepers G. Wilson M. Wilhelm D. Koopman P. J. Biol. Chem. 2003; 278: 28101-28108Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar), although SOX8 is not crucial for testis formation because Sox8 null mice develop normally (29Sock E. Schmidt K. Hermanns-Borgmeyer I. Bo ̈sl M. Wegner M. Mol. Cell. Biol. 2001; 21: 6951-6959Crossref PubMed Scopus (137) Google Scholar). Analysis of Sox9/Sox8 double mutant mice revealed that SOX8 complements SOX9 function during testis differentiation (30Chaboissier M.C. Kobayashi A. Vidal V.I. Lutzkendorf S. van de Kant H.J. Wegner M. de Rooij D.G. Behringer R.R. Schedl A. Development. 2004; 131: 1891-1901Crossref PubMed Scopus (473) Google Scholar). Because all SOX proteins, including SRY, share the same consensus binding sequence (31Mertin S. McDowall S.G. Harley V.R. Nucleic Acids Res. 1999; 27: 1359-1364Crossref PubMed Scopus (183) Google Scholar), we next tested whether SRY and SOX8 could bind to the Vanin-1 promoter fragment. Interestingly, GST·SOX8 was able to bind to the DNA fragment; however GST·SRY failed to form a specific DNA·protein complex (Fig. 3C). To locate the putative SOX factor binding site(s) within the Vanin-1 proximal promoter fragment, DNase I footprinting analysis was carried out (Fig. 4A). The Vanin-1 promoter was labeled on one strand and incubated with either GST or GST·SOX9 fusion protein. The inclusion of purified GST·SOX9 protein resulted in a protection of bases -206 to -173 (Fig. 4A) from DNase I digestion. This protected region of 34 bases is of a similar size to that protected in a footprinting analysis carried out on the Amh promoter (10de Santa Barbara P. Bonneaud N. Boizet B. Desclozeaux M. Moniot B. Su ̈dbeck P. Scherer G. Poulat F. Berta P. Mol. Cell. Biol. 1998; 18: 6653-6665Crossref PubMed Scopus (520) Google Scholar). To confirm this was the SOX site bound by SOX9, we next used site-directed mutagenesis to induce mutations into this putative SOX binding site and tested by EMSA. A mutation of this site prevented both SOX9 and SOX8 from binding to the promoter fragment (Fig. 4B), suggesting that the correct binding site for these proteins had been identified. SF-1 and SOX9 Co-activate Transcription from the Vanin-1 Proximal Promoter—To determine whether the sites for SF-1 and SOX9 that we identified are functional in regulating Vanin-1 expression, the proximal promoter was cloned immediately upstream of the luciferase gene in the pGL2-Basic reporter construct (pGL2-Vnn). TM3 cells, derived from mouse fetal gonads (23Mather J.P. Biol. Reprod. 1980; 23: 243-252Crossref PubMed Scopus (334) Google Scholar) and retaining expression of a number of genes of the sex-determining pathway (32Beverdam A. Wilhelm D. Koopman P. Cytogenet. Genome Res. 2003; 101: 242-249Crossref PubMed Scopus (28) Google Scholar), were co-transfected with pGL2-Vnn without or with an SF-1 expression plasmid (pcDNA3-SF-1) and were assayed for luciferase activity after 48 h. Co-transfection with 20 ng of pcDNA3-SF-1 resulted in a 22-fold increase in luciferase activity compared with the empty pcDNA3 vector (Fig. 5A). This response was dose-dependent, as a transfection of 50 ng further increased the promoter activity 35-fold above basal activity. To investigate possible synergy between SF-1 and SOX9 in activating Vanin-1 transcription, TM3 cells were co-transfected with the pGL2-Vnn reporter plasmid without or together with pcDNA3-SOX9 and/or pcDNA3-SF-1 expression plasmids. No activation was found in cells transfected with pcDNA3-SOX9 alone (Fig. 5A). However, co-transfection of pcDNA3-SOX9 and pcDNA3-SF-1 resulted in a doubling in the relative levels of luciferase produced compared with pcDNA3-SF-1 alone (Fig. 5A). These results demonstrated that SOX9 required the presence of SF-1 to activate transcription and that it can act synergistically with SF-1 to mediate expression from the Vanin-1 promoter. Our EMSA experiments showed that also SOX8, but not SRY, is able to bind the Vanin-1 promoter. Expression constructs pcDNA3-SOX8 and pcDNA3-SRY were assayed to determine whether these SOX factors could stimulate activity from the Vanin-1 promoter in reporter assays. Cells transfected with pcDNA3-SOX8 showed an increase in luciferase activity compared with cells transfected with pcDNA3-SF-1 alone (Fig. 5A). This suggests that SOX8 can, like SOX9, bind to and activate transcription from the Vanin-1 promoter in combination with SF-1. pcDNA3-SRY failed to activate the transcription of pGL2-Vnn either alone or in combination with pcDNA3-SF-1. This situation is similar to that reported for the Amh promoter where SRY was also unable to activate transcription, indicating that it cannot substitute for SOX9 in vitro (10de Santa Barbara P. Bonneaud N. Boizet B. Desclozeaux M. Moniot B. Su ̈dbeck P. Scherer G. Poulat F. Berta P. Mol. Cell. Biol. 1998; 18: 6653-6665Crossref PubMed Scopus (520) Google Scholar) even though it can bind directly to SF-1. 2M. Wilson, unpublished data. SOX transcription factors bind to DNA via their HMG domains, resulting in bending of the DNA by up to 71°, and has been proposed to act as an architectural transcription factor (33Preiss S. Argentaro A. Clayton A. John A. Jans D. Ogata T. Nagai T. Barroso I. Schafer A. Harley V. J. Biol. Chem. 2001; 276: 27864-27872Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar, 34Harley V.R. Clarkson M.J. Argentaro A. Endocr. Rev. 2003; 24: 466-467Crossref PubMed Scopus (183) Google Scholar). To determine whether SOX9 synergizes with SF-1 to activate the Vanin-1 promoter through DNA bending or in a more conventional manner through its trans-activation domain, TM3 cells were co-transfected with an expression construct, pcDNA3-SOX9ΔTA, encoding a SOX9 mutant protein that lacks the C-terminal trans-activation domain but can still bind to DNA. In luciferase assays this construct failed to activate transcription from the Vanin-1 promoter over levels seen with SF-1 alone (Fig. 5A). Thus, promoter regulation depended on the SOX9 trans-activation domain. To confirm that the binding site(s) responsible for activation by SF-1 and SOX9 had been identified, reporter constructs containing mutations in SF-1 site 1 (SF-M1), site 2 (SF-M2), or both (SF-M1M2), or in the SOX site (MutSox) were co-transfected with pcDNA3-SF-1. Mutation of either SF-1 site reduced SF-1-dependent activation from the Vanin-1 promoter to ∼28% of wild-type promoter expression levels. Mutation of both SF-1 sites completely abolished promoter activity (Fig. 5B). This indicates that both binding sites are required for transcriptional activation by SF-1. Previous analysis of the Amh promoter, which also contains two SF-1 DNA binding sites, found that mutation of either site greatly reduced promoter activity demonstrating also that both sites were required for full promoter activation by SF-1 (35Watanabe K. Clarke T. Lane A. Wang X. Donahoe P. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 1624-1629Crossref PubMed Scopus (126) Google Scholar). Co-transfection of pcDNA3-SOX9 and pcDNA3-SF-1 with the Vnn1 promoter-reporter construct containing the mutated SOX site (MutSox) identified by DNase I footprinting produced luciferase levels similar to pcDNA3-SF-1 transfection alone (Fig. 5B). Thus the putative SOX binding site identified in this study is essential for activation of transcription from the Vanin-1 promoter by SOX9. Disruption of SF-1 Function Leads to Loss of Expression of Vanin-1 in Vivo—Recently, targeted excision of the SF-1 gene in fetal Leydig cells in vivo has been reported, mediated by Cre-recombinase expression under the control of the AmhR2 promoter (26Jeyasuria P. Ikeda Y. Jamin S.P. Zhao L. de Rooij D.G. Themmen A.P.N. Behringer R.R. Parker K.L. Mol. Endocrinol. 2004; 18: 1610-1619Crossref PubMed Scopus (219) Google Scholar). The developing SF-1-null testes are smaller in size, show delayed cord formation, and lack expression of cholesterol side chain cleavage enzyme and StAR, indicating that these genes are in vivo targets of SF-1. We analyzed 14.5-dpc testes from knock-out and wild-type litterma
Referência(s)