Two Independent Nuclear Localization Signals Are Present in the DNA-binding High-mobility Group Domains of SRY and SOX9
1997; Elsevier BV; Volume: 272; Issue: 44 Linguagem: Inglês
10.1074/jbc.272.44.27848
ISSN1083-351X
Autores Tópico(s)Chromosomal and Genetic Variations
ResumoSRY and SOX9, members of the family of high-mobility group (HMG) domain transcription factors, are both essential for testis formation during human embryonic development. The HMG domain is a DNA-binding and DNA-bending motif comprising about 80 amino acid residues. It has been shown that SRY and SOX9 are nuclear proteins. Using normal or mutant SRY-β-galactosidase and SOX9-β-galactosidase fusion proteins in transfection studies involving COS-7 cells, we have identified two nuclear localization signals (NLSs) within the HMG domains of both proteins that can independently direct the fusion proteins into the nucleus. Only mutational inactivation of both NLS motifs resulted in complete exclusion of the fusion proteins from the nucleus. The NLS sequences are located at the N and C termini of the HMG domain and are a bipartite NLS motif and a basic cluster NLS motif, respectively. Both NLS motifs are conserved in the HMG domains of other transcription factors. The implications of the present results are discussed regarding (a) the apparent dual function of certain basic amino acid residues in the HMG domain of SRY in both DNA binding and in nuclear localization and (b) the possible control of SOX9 in early gonadal differentiation at the level of nuclear translocation. SRY and SOX9, members of the family of high-mobility group (HMG) domain transcription factors, are both essential for testis formation during human embryonic development. The HMG domain is a DNA-binding and DNA-bending motif comprising about 80 amino acid residues. It has been shown that SRY and SOX9 are nuclear proteins. Using normal or mutant SRY-β-galactosidase and SOX9-β-galactosidase fusion proteins in transfection studies involving COS-7 cells, we have identified two nuclear localization signals (NLSs) within the HMG domains of both proteins that can independently direct the fusion proteins into the nucleus. Only mutational inactivation of both NLS motifs resulted in complete exclusion of the fusion proteins from the nucleus. The NLS sequences are located at the N and C termini of the HMG domain and are a bipartite NLS motif and a basic cluster NLS motif, respectively. Both NLS motifs are conserved in the HMG domains of other transcription factors. The implications of the present results are discussed regarding (a) the apparent dual function of certain basic amino acid residues in the HMG domain of SRY in both DNA binding and in nuclear localization and (b) the possible control of SOX9 in early gonadal differentiation at the level of nuclear translocation. Mammalian sex determination and early gonadal differentiation is a developmental process involving a cascade of regulatory gene interactions. Only a few of these genes, all encoding transcription factors, are known (reviewed in Ref. 1Wolf U. J. Mol. Med. 1995; 73: 325-331Crossref PubMed Scopus (16) Google Scholar), among them the related genesSRY and SOX9. SRY encodes the Y-chromosomal testis-determining factor as shown by XY sex reversal in human individuals mutant for SRY (2Jäger R.J. Anvret M. Hall K. Scherer G. Nature. 1990; 348: 452-454Crossref PubMed Scopus (340) Google Scholar, 3Berta P. Hawkins J.R. Sinclair A.H. Taylor A. Griffiths B.L. Goodfellow P.N. Fellous M. Nature. 1990; 348: 448-450Crossref PubMed Scopus (428) Google Scholar) and by the demonstration of testis formation in chromosomally female mice transgenic for mouse Sry (4Koopman P. Gubbay J. Vivian N. Goodfellow P. Lovell-Badge R. Nature. 1991; 351: 117-121Crossref PubMed Scopus (1707) Google Scholar). SOX9 on chromosome 17 is an autosomal gene essential for testis development as mutations in and around this gene cause XY sex reversal in association with the skeletal malformation syndrome campomelic dysplasia (5Wagner T. Wirth J. Meyer J. Zabel B. Held M. Zimmer J. Pasantes J. Bricarelli F.D. Keutel J. Hustert E. Wolf U. Tommerup N. Schempp W. Scherer G. Cell. 1994; 79: 1111-1120Abstract Full Text PDF PubMed Scopus (1267) Google Scholar, 6Foster J.W. Dominguez-Steglich M.A. Guioli S. Kwok C. Weller P.A. Stevanovic M. Weissenbach J. Mansour S. Young I.D. Goodfellow P.N. Brook J.D. Schafer A.J. Nature. 1994; 372: 525-530Crossref PubMed Scopus (1303) Google Scholar).Both SRY and SOX9 contain an 80-amino acid DNA-binding motif known as the high-mobility group (HMG) 1The abbreviations used are: HMG, high-mobility group; NLS, nuclear localization signal; biNLS, bipartite NLS sequence; bcNLS, basic cluster NLS sequence; PCR, polymerase chain reaction; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactoside. 1The abbreviations used are: HMG, high-mobility group; NLS, nuclear localization signal; biNLS, bipartite NLS sequence; bcNLS, basic cluster NLS sequence; PCR, polymerase chain reaction; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactoside. domain that characterizes a whole class of transcription factors (reviewed in Ref.7Grosschedl R. Giese K. Pagel J. Trends Genet. 1994; 10: 94-100Abstract Full Text PDF PubMed Scopus (729) Google Scholar). SRY binds to the sequence AACAAT and variants thereof (8Harley V.R. Lovell-Badge R. Goodfellow P.N. Nucleic Acids Res. 1994; 22: 1500-1501Crossref PubMed Scopus (331) Google Scholar) and induces a sharp bend in the DNA (9Ferrari S. Harley V.R. Pontiggia A. Goodfellow P.N. Lovell-Badge R. Bianchi M.E. EMBO J. 1992; 11: 4497-4506Crossref PubMed Scopus (380) Google Scholar). The three-dimensional solution structure of the SRY HMG domain complexed with its target sequence has been solved (10Werner M.H. Huth J.R. Gronenborn A.M. Clore G.M. Cell. 1995; 81: 705-714Abstract Full Text PDF PubMed Scopus (425) Google Scholar), as has a similar complex of the related factor LEF-1 (11Love J.J. Li X. Case D.A. Giese K. Grosschedl R. Wright P.E. Nature. 1995; 376: 791-795Crossref PubMed Scopus (516) Google Scholar). In cell transfection studies, some evidence for transcriptional activation of testis-specific genes by SRY has been presented (12Haqq C.M. King C.-Y. Ukiyama E. Falsafi S. Haqq T.N. Donahoe P.K. Weiss M.A. Science. 1994; 266: 1494-1500Crossref PubMed Scopus (222) Google Scholar). We have shown in similar transfection assays that SOX9 also functions as a transcription factor, contains a C-terminal transactivation domain (13.Nat. Genet. 13, 230–232Südbeck, P., Schmitz, M. L., Baeuerle, P. A., and Scherer, G. Nat. Genet., 13, 230–232.Google Scholar) and can bind via its HMG domain to the motif AACAAT (14Meyer J. Südbeck P. Held M. Wagner T. Schmitz M.L. Eggermont E. Haas O. Kobelt A. Leroy J.G. van Maldergem L. Michel E. Mitulla B. Pfeiffer R.A. Schinzel A. Schmidt H. Scherer G. Hum. Mol. Genet. 1997; 6: 91-98Crossref PubMed Scopus (153) Google Scholar). Recently, mouse Sox9 was found to be expressed in the gonadal anlage of both sexes, with expression increasing in the developing testis and decreasing in the developing ovary, consistent with a role forSOX9/Sox9 in Sertoli cell differentiation (15Morais da Silva S. Hacker A. Harley V. Goodfellow P. Swain A. Lovell-Badge R. Nat. Genet. 1996; 14: 62-68Crossref PubMed Scopus (687) Google Scholar, 16Kent J. Wheatley S.C. Andrews J.E. Sinclair A.H. Koopman P. Development (Camb.). 1996; 122: 2813-2822Crossref PubMed Google Scholar).As transcription factors, SRY and SOX9 must gain access to the nucleus. Studies on nuclear localization indicate that transport across the nuclear envelope is an active process mediated by one or more nuclear localization signal sequences (NLSs), usually present in the protein itself or in a cofactor (for review, see Refs. 17Dingwall C. Laskey R.A. Trends Biochem. Sci. 1991; 16: 478-481Abstract Full Text PDF PubMed Scopus (1708) Google Scholar and 18Garcia-Bustos J. Heitman J. Hall M.N. Biochim. Biophys. Acta. 1991; 1071: 83-101Crossref PubMed Scopus (442) Google Scholar). With some exceptions, two main types of NLS motifs exist. One is a short cluster of mainly basic amino acids (arginine and/or lysine), its prototype found in the simian virus 40 large tumor antigen (19Kalderon D. Roberts B.L. Richardson W.D. Smith A.E. Cell. 1984; 39: 499-509Abstract Full Text PDF PubMed Scopus (1854) Google Scholar). The other is a bipartite NLS motif that comprises two basic amino acids, a spacer of about 10–15 residues consisting of any amino acid, followed by generally three basic residues, as first described for nucleoplasmin (17Dingwall C. Laskey R.A. Trends Biochem. Sci. 1991; 16: 478-481Abstract Full Text PDF PubMed Scopus (1708) Google Scholar). Specialized NLS-binding transporter proteins that carry NLS-containing proteins through the nuclear pore complex into the nucleus have been identified recently (20Görlich D. Vogel F. Mills A.D. Hartmann E. Laskey R.A. Nature. 1995; 377: 246-248Crossref PubMed Scopus (409) Google Scholar).Karyophilic NLS sequences are generally identified by their ability to direct an otherwise cytoplasmic protein to the nucleus when fused to it genetically or biochemically and by the effects of deletion or point mutations on nuclear entry (18Garcia-Bustos J. Heitman J. Hall M.N. Biochim. Biophys. Acta. 1991; 1071: 83-101Crossref PubMed Scopus (442) Google Scholar). Using these approaches with β-galactosidase as a reporter protein, we have identified two independent NLS motifs within the DNA-binding HMG domains of SRY and SOX9 that are both required for complete nuclear translocation. In a previous study, only one of the two NLSs had been identified in SRY (21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar).DISCUSSIONWe have observed that SRY and SOX9, two proteins essential for normal testis formation during human embryonic development, are located exclusively in the nuclei of transfected cells, in agreement with previous findings (15Morais da Silva S. Hacker A. Harley V. Goodfellow P. Swain A. Lovell-Badge R. Nat. Genet. 1996; 14: 62-68Crossref PubMed Scopus (687) Google Scholar, 21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar). A bipartite and a basic cluster NLS motif were identified within the HMG domains of both proteins to be necessary and sufficient for effective nuclear import. Both NLS motifs are conserved in the HMG domains of other members of this family of transcription factors (Fig. 3). This is particularly evident for the subgroup of the SRY HMG-box (domain)-related SOX proteins. It is conceivable that most, if not all, HMG domain transcription factors utilize NLS motifs identical or related to those we have identified here. In fact, for the lymphocyte-specific factor LEF-1, which apparently lacks a bipartite NLS, the extended basic cluster at the C terminus of its HMG domain has been identified as its major NLS (Ref. 30Prieve M.G. Guttridge K.L. Munguia J.E. Waterman M.L. J. Biol. Chem. 1996; 271: 7654-7658Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar; see Fig. 3). Moreover, the recently identified SOX17 isoform that lacks essentially all of the HMG domain except the C-terminal basic cluster sequence was found to be localized in the nucleus and in the cytoplasm of transfected cells (31Kanai Y. Kanai-Azuma M. Noce T. Saido T.C. Shiroishi T. Hayashi Y. Yazaki K. J. Cell Biol. 1996; 133: 667-681Crossref PubMed Scopus (188) Google Scholar). By contrast, the full-length SOX17 with the complete HMG domain containing a bipartite and a basic cluster NLS motif very similar to those in SRY and SOX9 (Fig. 3) was completely nuclear (31Kanai Y. Kanai-Azuma M. Noce T. Saido T.C. Shiroishi T. Hayashi Y. Yazaki K. J. Cell Biol. 1996; 133: 667-681Crossref PubMed Scopus (188) Google Scholar). This is in line with our observation of two independent NLSs in the HMG domain being necessary for complete nuclear translocation.While this work was in progress, a similar study by Poulat et al. (21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar) appeared describing the N-terminal bipartite NLS in the HMG domain of SRY. These authors identified the same basic residues as essential components of this motif as the present study. However, in contrast to our findings, they describe nuclear import of SRY to be controlled solely by the bipartite NLS motif. By deleting the 20 N-terminal residues of the HMG domain that span this motif in full-length SRY, the little nuclear staining seen in transfected cells with an SRY antibody was interpreted as passive diffusion of the NLS-deleted protein into the nucleus. The same deletion in the context of an SRY-β-galactosidase fusion construct was reported, but not shown, to result in localization only in the cytoplasm. In contrast, in the present study, by mutating the relevant residues in both parts of the bipartite NLS but keeping the remainder of the HMG domain unchanged in construct SRY-M3, nuclear translocation was observed to be reduced but clearly not abolished. Only after additional deletion of the C-terminal basic cluster in construct SRY-D3-M3 did we observe complete nuclear exclusion (Fig. 1). Furthermore, when tested individually and outside the context of the HMG domain, not only the bipartite NLS, but also the basic cluster NLS, was able to direct the reporter protein into the nucleus, although not completely, as shown by constructs SRY-biNLS and SRY-bcNLS (Fig. 1). Corresponding results were obtained for SOX9 (Fig. 2). Our data do not allow one to decide as to which of the two NLS motifs is stronger. It is possible that under the experimental conditions used by Poulat et al. (21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar), the basic cluster NLS functioned as a slightly weaker nuclear localization signal than the bipartite NLS and therefore was not apparent in their study. A weakly functioning basic cluster NLS may also explain why a peptide corresponding to the entire SRY HMG domain but the first 16 residues failed to translocate coupled rabbit IgG to the nucleus in the study of Poulat et al. (21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar). It has been shown that weak NLS motifs require a high coupling ratio of peptide/IgG to drive the coupled protein into the nucleus (25Vandromme M. Cavadore J.-C. Bonnieu A. Froschlé A. Lamb N. Fernandez A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 4646-4650Crossref PubMed Scopus (54) Google Scholar).It is unlikely that another functional NLS, in addition to the two motifs in the HMG domain, is present in SRY. The fact that constructs SRY-D3-M1, -M2, and -M3, where these motifs are mutated, are localized only in the cytoplasm, argues against an NLS in the N-terminal 58 residues. The C-terminal 68 residues of SRY that are missing from these constructs contain only six basic residues that are not clustered. Furthermore, the HMG domain is the only conserved region in SRY sequences from different species (24Whitfield L.S. Lovell-Badge R. Goodfellow P.N. Nature. 1993; 364: 713-715Crossref PubMed Scopus (323) Google Scholar), suggesting that this domain is the only functionally relevant part. However, for SOX9, where we have only analyzed the HMG domain, we cannot fully exclude an additional NLS elsewhere in the 509-residue SOX9 protein, although a typical NLS motif is not apparent from the sequence.Two interesting features of the NLSs present in the HMG domain of SRY and SOX9 are shared by other nuclear proteins. The first relates to the phenomenon of multiple karyophilic signals. As in SRY and SOX9, other nuclear proteins contain two mutually independent NLSs that have to be inactivated together to completely abolish nuclear import (25Vandromme M. Cavadore J.-C. Bonnieu A. Froschlé A. Lamb N. Fernandez A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 4646-4650Crossref PubMed Scopus (54) Google Scholar, 32Hall M.N. Craik C. Hiraoka Y. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 6954-6958Crossref PubMed Scopus (65) Google Scholar). Such an arrangement may reduce the susceptibility of a nuclear protein to total loss of nuclear translocation by NLS-inactivating point mutations. The second feature is the location of the NLSs in a DNA-binding domain, a situation described for several other proteins, including a number of transcription factors such as MyoD (25Vandromme M. Cavadore J.-C. Bonnieu A. Froschlé A. Lamb N. Fernandez A. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 4646-4650Crossref PubMed Scopus (54) Google Scholar) or NF-κB (33Henkel T. Zabel U. van Zee K. Müller J.M. Fanning E. Baeuerle P.A. Cell. 1992; 68: 1121-1133Abstract Full Text PDF PubMed Scopus (302) Google Scholar). The apparent economic design of incorporating basic nuclear targeting sequences together with frequently basic DNA-binding residues in a single domain may be the result of an evolutionary process if, as has been suggested (17Dingwall C. Laskey R.A. Trends Biochem. Sci. 1991; 16: 478-481Abstract Full Text PDF PubMed Scopus (1708) Google Scholar), DNA-binding domains were the archetypal targeting signals during evolution of the nuclear membrane. Interestingly, in the case of SRY, the two arginine residues 62 and 75 in the bipartite NLS (residues 5 and 18 in Fig. 3) contribute to the DNA binding of the HMG domain, both by forming salt bridges to phosphate residues and by hydrophobic interactions with sugar residues, as revealed by the NMR structure of the SRY/DNA complex (10Werner M.H. Huth J.R. Gronenborn A.M. Clore G.M. Cell. 1995; 81: 705-714Abstract Full Text PDF PubMed Scopus (425) Google Scholar). Likewise, the lysine and arginine residues in LEF-1 that correspond to basic residues 61, 62, and 75 in the bipartite NLS of SRY (residues 4, 5, and 18 in Fig. 3) also interact with the DNA backbone in the LEF-1/DNA complex (11Love J.J. Li X. Case D.A. Giese K. Grosschedl R. Wright P.E. Nature. 1995; 376: 791-795Crossref PubMed Scopus (516) Google Scholar). Since the NMR analysis of the SRY-DNA complex did not clearly resolve the residues of the basic cluster NLS (10Werner M.H. Huth J.R. Gronenborn A.M. Clore G.M. Cell. 1995; 81: 705-714Abstract Full Text PDF PubMed Scopus (425) Google Scholar), it is not known if some of these residues may also make DNA contacts. This is likely, however, in view of the fact that the C-terminal basic residues in the HMG domain of LEF-1 (see Fig. 3) contact the sugar-phosphate backbone extensively in the LEF-1/DNA complex (11Love J.J. Li X. Case D.A. Giese K. Grosschedl R. Wright P.E. Nature. 1995; 376: 791-795Crossref PubMed Scopus (516) Google Scholar).It thus appears that certain basic residues in both NLS motifs of the HMG domain serve two functions, playing a role both in nuclear import and in DNA binding. Sex reversal in XY females with amino acid substitutions affecting these basic residues in SRY, or in SOX9, may then result not only from impaired DNA binding but also from impaired nuclear uptake of the mutant protein. Whereas one such published mutation in SRY, R62G (26Affara N.A. Chalmers I.J. Ferguson-Smith M.A. Hum. Mol. Genet. 1993; 2: 785-789Crossref PubMed Scopus (102) Google Scholar), showed normal nuclear translocation, another mutation tested, R76L, resulted in reduced nuclear import (Fig.1). It is likely that this or similar mutations will be described in the future.There is growing evidence that gene transcription can be controlled by regulating the nuclear import of transcription factors. This can be achieved by interacting proteins masking one or several NLSs present in these factors, thus rendering these signals nonfunctional (33Henkel T. Zabel U. van Zee K. Müller J.M. Fanning E. Baeuerle P.A. Cell. 1992; 68: 1121-1133Abstract Full Text PDF PubMed Scopus (302) Google Scholar, 34Chen C.-M.A. Kraut N. Groudine M. Weintraub H. Cell. 1996; 86: 731-741Abstract Full Text Full Text PDF PubMed Scopus (123) Google Scholar). Such a masking mechanism could account for the recent observation that SOX9 is localized predominantly in the cytoplasm of cells of the developing gonadal ridge in both sexes prior to day 11.5 of mouse embryogenesis, while it is completely nuclear at later stages of male gonadal development (15Morais da Silva S. Hacker A. Harley V. Goodfellow P. Swain A. Lovell-Badge R. Nat. Genet. 1996; 14: 62-68Crossref PubMed Scopus (687) Google Scholar). Whatever the nature of this hypothetical masking factor, the role of which may be an essential one in controlling mammalian sex determination, it is likely to interact with one or both NLS motifs in SOX9 identified in this study. Mammalian sex determination and early gonadal differentiation is a developmental process involving a cascade of regulatory gene interactions. Only a few of these genes, all encoding transcription factors, are known (reviewed in Ref. 1Wolf U. J. Mol. Med. 1995; 73: 325-331Crossref PubMed Scopus (16) Google Scholar), among them the related genesSRY and SOX9. SRY encodes the Y-chromosomal testis-determining factor as shown by XY sex reversal in human individuals mutant for SRY (2Jäger R.J. Anvret M. Hall K. Scherer G. Nature. 1990; 348: 452-454Crossref PubMed Scopus (340) Google Scholar, 3Berta P. Hawkins J.R. Sinclair A.H. Taylor A. Griffiths B.L. Goodfellow P.N. Fellous M. Nature. 1990; 348: 448-450Crossref PubMed Scopus (428) Google Scholar) and by the demonstration of testis formation in chromosomally female mice transgenic for mouse Sry (4Koopman P. Gubbay J. Vivian N. Goodfellow P. Lovell-Badge R. Nature. 1991; 351: 117-121Crossref PubMed Scopus (1707) Google Scholar). SOX9 on chromosome 17 is an autosomal gene essential for testis development as mutations in and around this gene cause XY sex reversal in association with the skeletal malformation syndrome campomelic dysplasia (5Wagner T. Wirth J. Meyer J. Zabel B. Held M. Zimmer J. Pasantes J. Bricarelli F.D. Keutel J. Hustert E. Wolf U. Tommerup N. Schempp W. Scherer G. Cell. 1994; 79: 1111-1120Abstract Full Text PDF PubMed Scopus (1267) Google Scholar, 6Foster J.W. Dominguez-Steglich M.A. Guioli S. Kwok C. Weller P.A. Stevanovic M. Weissenbach J. Mansour S. Young I.D. Goodfellow P.N. Brook J.D. Schafer A.J. Nature. 1994; 372: 525-530Crossref PubMed Scopus (1303) Google Scholar). Both SRY and SOX9 contain an 80-amino acid DNA-binding motif known as the high-mobility group (HMG) 1The abbreviations used are: HMG, high-mobility group; NLS, nuclear localization signal; biNLS, bipartite NLS sequence; bcNLS, basic cluster NLS sequence; PCR, polymerase chain reaction; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactoside. 1The abbreviations used are: HMG, high-mobility group; NLS, nuclear localization signal; biNLS, bipartite NLS sequence; bcNLS, basic cluster NLS sequence; PCR, polymerase chain reaction; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactoside. domain that characterizes a whole class of transcription factors (reviewed in Ref.7Grosschedl R. Giese K. Pagel J. Trends Genet. 1994; 10: 94-100Abstract Full Text PDF PubMed Scopus (729) Google Scholar). SRY binds to the sequence AACAAT and variants thereof (8Harley V.R. Lovell-Badge R. Goodfellow P.N. Nucleic Acids Res. 1994; 22: 1500-1501Crossref PubMed Scopus (331) Google Scholar) and induces a sharp bend in the DNA (9Ferrari S. Harley V.R. Pontiggia A. Goodfellow P.N. Lovell-Badge R. Bianchi M.E. EMBO J. 1992; 11: 4497-4506Crossref PubMed Scopus (380) Google Scholar). The three-dimensional solution structure of the SRY HMG domain complexed with its target sequence has been solved (10Werner M.H. Huth J.R. Gronenborn A.M. Clore G.M. Cell. 1995; 81: 705-714Abstract Full Text PDF PubMed Scopus (425) Google Scholar), as has a similar complex of the related factor LEF-1 (11Love J.J. Li X. Case D.A. Giese K. Grosschedl R. Wright P.E. Nature. 1995; 376: 791-795Crossref PubMed Scopus (516) Google Scholar). In cell transfection studies, some evidence for transcriptional activation of testis-specific genes by SRY has been presented (12Haqq C.M. King C.-Y. Ukiyama E. Falsafi S. Haqq T.N. Donahoe P.K. Weiss M.A. Science. 1994; 266: 1494-1500Crossref PubMed Scopus (222) Google Scholar). We have shown in similar transfection assays that SOX9 also functions as a transcription factor, contains a C-terminal transactivation domain (13.Nat. Genet. 13, 230–232Südbeck, P., Schmitz, M. L., Baeuerle, P. A., and Scherer, G. Nat. Genet., 13, 230–232.Google Scholar) and can bind via its HMG domain to the motif AACAAT (14Meyer J. Südbeck P. Held M. Wagner T. Schmitz M.L. Eggermont E. Haas O. Kobelt A. Leroy J.G. van Maldergem L. Michel E. Mitulla B. Pfeiffer R.A. Schinzel A. Schmidt H. Scherer G. Hum. Mol. Genet. 1997; 6: 91-98Crossref PubMed Scopus (153) Google Scholar). Recently, mouse Sox9 was found to be expressed in the gonadal anlage of both sexes, with expression increasing in the developing testis and decreasing in the developing ovary, consistent with a role forSOX9/Sox9 in Sertoli cell differentiation (15Morais da Silva S. Hacker A. Harley V. Goodfellow P. Swain A. Lovell-Badge R. Nat. Genet. 1996; 14: 62-68Crossref PubMed Scopus (687) Google Scholar, 16Kent J. Wheatley S.C. Andrews J.E. Sinclair A.H. Koopman P. Development (Camb.). 1996; 122: 2813-2822Crossref PubMed Google Scholar). As transcription factors, SRY and SOX9 must gain access to the nucleus. Studies on nuclear localization indicate that transport across the nuclear envelope is an active process mediated by one or more nuclear localization signal sequences (NLSs), usually present in the protein itself or in a cofactor (for review, see Refs. 17Dingwall C. Laskey R.A. Trends Biochem. Sci. 1991; 16: 478-481Abstract Full Text PDF PubMed Scopus (1708) Google Scholar and 18Garcia-Bustos J. Heitman J. Hall M.N. Biochim. Biophys. Acta. 1991; 1071: 83-101Crossref PubMed Scopus (442) Google Scholar). With some exceptions, two main types of NLS motifs exist. One is a short cluster of mainly basic amino acids (arginine and/or lysine), its prototype found in the simian virus 40 large tumor antigen (19Kalderon D. Roberts B.L. Richardson W.D. Smith A.E. Cell. 1984; 39: 499-509Abstract Full Text PDF PubMed Scopus (1854) Google Scholar). The other is a bipartite NLS motif that comprises two basic amino acids, a spacer of about 10–15 residues consisting of any amino acid, followed by generally three basic residues, as first described for nucleoplasmin (17Dingwall C. Laskey R.A. Trends Biochem. Sci. 1991; 16: 478-481Abstract Full Text PDF PubMed Scopus (1708) Google Scholar). Specialized NLS-binding transporter proteins that carry NLS-containing proteins through the nuclear pore complex into the nucleus have been identified recently (20Görlich D. Vogel F. Mills A.D. Hartmann E. Laskey R.A. Nature. 1995; 377: 246-248Crossref PubMed Scopus (409) Google Scholar). Karyophilic NLS sequences are generally identified by their ability to direct an otherwise cytoplasmic protein to the nucleus when fused to it genetically or biochemically and by the effects of deletion or point mutations on nuclear entry (18Garcia-Bustos J. Heitman J. Hall M.N. Biochim. Biophys. Acta. 1991; 1071: 83-101Crossref PubMed Scopus (442) Google Scholar). Using these approaches with β-galactosidase as a reporter protein, we have identified two independent NLS motifs within the DNA-binding HMG domains of SRY and SOX9 that are both required for complete nuclear translocation. In a previous study, only one of the two NLSs had been identified in SRY (21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar). DISCUSSIONWe have observed that SRY and SOX9, two proteins essential for normal testis formation during human embryonic development, are located exclusively in the nuclei of transfected cells, in agreement with previous findings (15Morais da Silva S. Hacker A. Harley V. Goodfellow P. Swain A. Lovell-Badge R. Nat. Genet. 1996; 14: 62-68Crossref PubMed Scopus (687) Google Scholar, 21Poulat F. Girard F. Chevron M.-P. Gozé C. Rebillard X. Calas B. Lamb N. Berta P. J. Cell Biol. 1995; 128: 737-748Crossref PubMed Scopus (100) Google Scholar). A bipartite and a basic cluster NLS motif were identified within the HMG domains of both proteins to be necessary and sufficient for effective nuclear import. Both NLS motifs are conserved in the HMG domains of other members of this family of transcription factors (Fig. 3). This is particularly evident for the subgroup of the SRY HMG-box (domain)-related SOX proteins. It is conceivable that most, if not all, HMG domain transcription factors utilize NLS motifs identical or related to those we have identified here. In fact, for the lymphocyte-specific factor LEF-1, which apparently lacks a bipartite NLS, the extended basic cluster at the C terminus of its HMG domain has been identified as its major NLS (Ref. 30Prieve M.G. Guttridge K.L. Munguia J.E. Waterman M.L. J. Biol. Chem. 1996; 271: 7654-7658Abstract Full Text Full Text PDF PubMed Scopus (92) Google Scholar; see Fig. 3). Moreover, the recently identified SOX17 isoform that lacks essentially all of the HMG domain except the C-terminal basic cluster sequence was found to be localized in the nucleus and in the cytoplasm of transfected cells (31Kanai Y. Kanai-Azuma M. Noce T. Saido T.C. Shiroishi T. Hayashi Y. Yazaki K. J. Cell Biol. 1996; 133: 667-681Crossref PubMed Scopus (188) Google Scholar). By contrast, the full-length SOX17 with the complete HMG domain containing a bipartite and a basic cluster NLS motif very similar to those in SRY and SOX9 (Fig. 3) was completely nuclear (31Kanai Y. Kanai-Azuma M. Noce T. Saido T.C. Shiroishi T. Hayashi Y. Yazaki K. J. Cell Biol. 1996; 133: 667-681Crossref PubMed Scopus (188) Google Sch
Referência(s)