SOX22 is a new member of the SOX gene family, mainly expressed in human nervous tissue
SOX22 is a new member of the SOX gene family, mainly expressed in human nervous tissuePhilippe Jay1, Iman, Sahly2, Catherine Gozé1, Sylvie Taviaux1, Francis Poulat1, Gérard Couly3, Marc Abitbol2 and Philippe Berta1,*
1Centre de Recherches de Biochimie Macromoléculaire, CNRS ERS155, route de Mende, BP5051, 34033 Montpellier cedex, France, 2Centre de Recherche Thérapeutique en Ophtalmologie, Faculté de Médecine Necker-Enfants Malades, 156 rue de Vaugirard, 75015 Paris, France and 3Institut d'Embryologie Cellulaire et Moléculaire du CNRS et du Collège de France, 94000 Nogent sur Marne, France
Received February 3, 1997;Revised and Accepted April 23, 1997
DDBJ/EMBL/GenBank accession no. U35612
SOX (SRY box-containing) genes share a particular DNA-binding domain, called HMG, with the mammalian testis-determining gene SRY. Several SOX genes have already been shown to be transcription factors involved in the decision of important cell fates during development. Here we report the cloning of a new human member of the SOX gene family, SOX22. The corresponding protein contains several domains that are also present in other paralogous SOX proteins. The SOX22 gene maps to chromosome 20 on band p13 and does not appear to be clustered with any other SOX gene mapped to date. SOX22 mRNA is expressed in various fetal and adult organs and tissues, suggesting that this gene plays roles in both differentiation and maintenance of several cell types.
DNA-binding proteins are typically implicated in the developmental control of gene expression. High mobility group (HMG) proteins contain a DNA-binding motif, called the HMG domain, that was first identified in the UBF RNA polymerase I-specific transcription factor by analogy with the DNA-binding domain of HMG-1 and HMG-2 proteins (1 ). These have been proposed to act either as target-specific transcription factors or as chromatin structure regulatory elements, or both together. Since then, >100 HMG box-containing proteins have been reported and classified into two distinct sub-groups according to the sequence specificity of the DNA binding, the number of HMG DNA-binding domains and phylogenetic arguments (2 ). The first sub-group comprises proteins that are all potential transcription factors believed to control gene expression during development. They contain only one DNA-binding domain and they bind to DNA in a sequence-specific fashion. The second sub-group consists of all other HMG box-containing proteins, most of which contain more than one DNA-binding domain and can bind to DNA in a non-sequence-specific manner.
The testis-determining gene located on chromosome Y (SRY) belongs to the transcription factor sub-group. Its cloning led to the discovery of a new family of both autosomal and X-linked genes that have been called SOX (SRY-box related) because of the strong homology of their DNA-binding domain with the HMG box of SRY (3 ). Since then, degenerate PCR-based methods have enabled the cloning and sequencing of a great number of new SRY-related box sequences in both vertebrates and invertebrates (3 -12 ), but few of them have been characterized further.
The precise function of most SOX genes is still not clear, and most of the available data are based on expression patterns. Several Sox genes are expressed in the developing nervous system (13 -15 ), which has led to the hypothesis that they may play interactive roles during the development of the nervous system (13 ). In addition, some SOX genes display a tightly regulated spatio-temporal expression pattern (14 ) whereas others display more complex expression patterns, suggesting that they could play a role in the development of several distinct tissues or organs (16 ).
Sox genes are believed to encode transcription regulators. In vitro, the HMG domain of SOX proteins is characterized by its ability to bind DNA in the minor groove (17 ) through AT-rich sequences, and to induce a strong bending of DNA upon binding (18 ,19 ). Because the binding of SOX proteins to DNA results in significant alterations of chromatin structure, it was proposed that they also act as architectural elements of chromatin. The recent report of a Sox gene (Sox70D) being identical to the DrosophilaDichaete gene confirmed this hypothesis, since its absence only led to a decrease in the expression of the target segmentation genes even-skipped, hairy and runt that were still expressed approximately at the correct antero-posterior position (20 ).
The capacity to regulate transcription of target genes was demonstrated for several Sox proteins. Murine Sox4, Sox17 and Sox18, as well as human SOX9, can activate the transcription of a reporter gene placed under the control of an enhancer element containing polymerized Sox DNA target sequences (21 -24 ). In addition, mouse Sox2 and human SOX4 and SOX9 activate the transcription of a reporter gene via enhancer elements present in putative in vivo target genes (25 -27 ). Sox proteins are thus considered as transcriptional regulators, the action of which involves local alteration of chromatin structure (28 ).
We recently have described the human SOX11 gene (14 ) and we assume that sequence comparison with other closely related SOX genes will allow the identification of conserved and putatively important functional domains within these proteins. With this aim, we report the cloning of SOX22, a new human member of the SOX family that has still not been identified in any other mammalian species. We also present the structure and chromosomal location of the SOX22 gene as well as its expression pattern in embryonic, fetal and adult human tissues.
Approximately 1*106 clones from the human genomic library X4Y[lambda]GEM were screened under low stringency conditions using the HMG domain of human SOX11 (10 ) as a probe. Two positive clones were detected and purified. Restriction and Southern analysis revealed that one of the phages, X4Y-A corresponded to the human SOX11 gene, whereas the second one, called X4Y-B, contained another SRY-related sequence, as yet unknown in gene databases. This putative new SOX gene will be referred to from hereafter as SOX22. A 3.5 kb HindIII fragment of X4Y-B that hybridized with the SOX11 box probe was subcloned into pBluescript KS+ phagemid (clone pKSSOX22). Sequence analysis showed that this clone contained a putative open reading frame (ORF) encompassing a typical HMG domain. This ORF was fully sequenced, and a 2 kb NotI-HindIII sub-fragment that did not contain the HMG region of SOX22 was used as a probe to screen a [lambda]ZAP fetal brain cDNA library (kindly provided by Dr C. Petit). Several overlapping cDNA clones were isolated, and the largest two, encompassing 2352 bp, were sequenced. The cDNA sequence was co-linear with the genomic sequence of X4Y-B but was truncated before the 5' end of the ORF. Rescreening of several different cDNA libraries failed to provide further 5'-located sequence. As such, the genomic sequence was used to complete the cDNA sequence. A stop codon was situated in the genomic sequence, 195 bp upstream of the first methionine of the putative ORF, and no canonical acceptor splice site (29 ) was found in these 195 bp, indicating that the SOX22 protein is likely to be encoded by a single exon. In addition, the first methionine of the ORF fulfills the Kozak criteria for efficient translation (30 ) and is followed by the amino acids VQQ, which previously have been described as the first residues in human SOX4 (31 ) and SOX11 (14 ) proteins. We consequently deduced that this methionine actually was the initiator codon. The predicted protein is 315 amino acid residues long, with a calculated mol. wt of 34.3 kDa and an isoelectric point of 7.9. Comparison of both nucleotide and amino acid sequences of SOX22 against gene and protein databases showed that human SOX22 had not been described in any mammalian species previously. The best match was found with two reptilian partial sequences called ADW5 and ADW2 that are very similar (95% identity each) to the HMG domain of human SOX22 (not shown). Among SOX proteins, SOX22 is most homologous to SOX4 and SOX11. Together, they form a sub-group of SOX proteins, the HMG DNA-binding domain of which is the most divergent from that of SRY. Conservation between these three SOX proteins is shown in Figure 1 A. Except for the SRY-related DNA-binding domain, the most striking feature in the sequence of SOX22 protein is the presence of several proline-rich regions on both sides of the HMG box and of an extremely acidic domain on the carboxy side of the SRY box. These two types of sequences have already been described in several transcription factors as responsible of transactivation properties (32 ).
The SOX22 gene was mapped to chromosome 20 at position p13 by fluorescence in situ hybridization (FISH), using a probe consisting of the 3.5 kb HindIII insert of pKSSOX22 (Fig. 2 ). For 50 metaphase preparations, specific labelling of band 20p13 was observed on two (eight cells) and one (25 cells) chromatids of chromosome 20 homologues. Twenty non-specific spots were scattered throughout the other chromosomes, and no secondary peak was detected.
Expression was first analysed by Northern blotting. The 2 kb NotI-HindIII fragment of pKSSOX22, lacking the conserved HMG domain, was used as a probe to hybridize Northern blots containing a large panel of fetal and adult tissues. SOX22 mRNAs were detected in essentially all tissues, but at variable levels (Fig. 3 ). Fetal brain and kidney and adult heart, pancreas, testis and ovary displayed the strongest signals, whereas other tissues were only weakly positive. A major transcript of 3.5 kb and a minor transcript of ~5.0 kb in length were generally visible. An additional band at ~1.5 kb was detected in skeletal and heart muscle. The presence of three distinct mRNAs could result from the use of alternative promoters and/or of different polyadenylation signals. Our data indicate that the coding region of SOX22 is monoexonic, but the existence of one or several additional(s), non-translated exon(s) cannot be excluded. The precise understanding of the structure of the three SOX22 messengers will require further investigation and screening of muscle cDNA libraries. The three messengers are not likely to represent different SOX genes for the following two reasons: first, we used a specific probe lacking the HMG box and high stringency hybridization conditions, thus minimizing the possibility of cross-hybridization; and second, no secondary peaks were detected by FISH analysis. The same blots were then stripped and hybridized with an actin probe in order to check the quantity and integrity of the mRNA immobilized in every lane (Fig. 3 ).
In order to determine the expression of SOX22 in embryonic tissues more precisely, we performed in situ hybridization with oligonucleotide probes designed from the transcribed sequence of SOX22 and located outside the SRY-related motif. Two different antisense probes were used and gave the same results.
The SOX22 gene is expressed most abundantly in the central nervous system (CNS). In 3- to 8-week-old human embryos, the whole CNS is labelled at the autoradiographic level (Fig. 4 A-D) as well as at the cellular level (Fig. 5 A-D). The most external neurons of the telencephalon are labelled more than the subventricular ones, whilst the periventricular progenitor cells of the rhombencephalon (Fig. 5 A and B) appear to be specifically more intensely labelled than the lateral cells of this structure. The choroid plexuses do express SOX22 mRNAs. The tela choroidea, a well vascularized layer of pia mater covering the thin rhombencephalon roof plate, is also clearly labelled. All the cells constituting the medulla are equally and strongly labelled (Fig. 5 C and D), whereas arachnoid and dura mater surrounding the medulla are not. The dorsal root ganglia express SOX22 mRNAs strongly (Fig. 4 B-D) and all cranial nerve nuclei are positive, with the nuclei of the Vth, VIth, VIIth, VIIIth, IXth and Xth cranial nerves being strongly labelled (Fig. 4 A). The Rathke pouch is not labelled. The optic vesicle (Fig. 4 A) and the otic vesicle are clearly labelled. The neuroblastic layer of the retina expresses SOX22 mRNAs at all of the developmental stages examined, whereas no cells of the lens are labelled.
SOX22 gene expression is not limited to the CNS. Expression of mRNA can be detected in many somitic derivatives and in many tissues originating from the mesoblast: the vertebral and the costal mesenchyme (Fig. 4 A-C) are labelled unambiguously. The developing vertebral bodies (Fig. 4 C) are labelled much more intensely than the developing vertebral arcs and, interestingly, the developing fibrous intervertebral disks are not labelled. The original core of each disk, the nucleus pulposus, which is composed of cells of notochordal origin, is not labelled. It is worthwhile mentioning that the paired mesenchymal condensations, the sternal bars, that form within the ventral body wall at the end of the sixth week of human development, do not express SOX22 mRNA.
The mesenchyme of the first, second and third branchial arcs is labelled, but much less so than the CNS. The epithelium of the pharynx stricto sensu does not express SOX22 mRNAs. The tongue, which develops mainly from pharyngeal arcs 1, 3 and 4 and from occipital somitic mesoderm, is strongly labelled. The thyroid gland primordium, which derives from an invagination of the tongue endoderm, and the thymus primordium, which arises from the third pharyngeal pouch, are clearly labelled at both the autoradiographic and cellular level. The heart and the liver are moderately labelled at the autoradiographic level. A careful analysis of the labelling at the cellular level reveals that both the myocardium or heart muscle (not shown) and the endocardium express SOX22 mRNA. The intensity of labelling of the heart is weaker than that of the CNS and is comparable with that of the branchial arcs. Many cephalic neural crest derivatives, including cartilaginous rudiments of several bones of the nose, face, middle ear and neck, do not display any detectable SOX22 gene expression. The oral ectoderm and, more generally, the surface ectoderm is not labelled. The notochord is not labelled. The musculature at the base of the tongue is strongly labelled (Fig. 4 B). The mesenchyme of the tracheal wall is labelled. The pleura and the pulmonary vasculature of the lung strongly express SOX22 mRNA, whilst the alveolar lung parenchyme and the epithelium of the bronchi do not. The haematopoietic progenitor cells of the liver exhibit notable SOX22 expression. Finally, we did not detect any SOX22 gene expression in the gut epithelium of 3- to 8-week-old human embryonic tissues.
In 25-week-old fetal brain, all neuronal structures are labelled at the autoradiographic level as well as at the cellular level (Fig. 6 A and B). No glial cells appear to be labelled. Strong labelling is observed in the cerebral and cerebellar cortices as well as in the subventricular germinative zones. The striatum and the dentate nucleus of the cerebellum are moderately labelled (Fig. 6 A and B).
Here we report the characterization of a new human SOX gene, which we have called SOX22. This gene displays strong homology with two alligator partial sequences, suggesting that an orthologous gene might exist in reptiles. Comparison of the SOX22 sequence with other genes in databases reveals significant homology of SOX22 with several paralogous SOX genes outside the SOX box, thus defining a distinct sub-group of SOX genes together with SOX4 and SOX11. The proteins encoded by these genes present several conserved stretches of amino acids in their primary structure, thus defining domains that are conserved among two or more proteins but that are distributed differently between the members of this sub-family of SOX (Fig. 1 B). One of these modules is the SRY-related DNA-binding domain, which defines the SOX family. The presence of alanine-rich regions in all three SOX proteins and the chromatin structure regulation function of SOX proteins prompt a comparison with the yeast Tup1 protein, which contains similar alanine-rich sequences and acts together with Ssn6 as a specific repressor of diverse euchromatic genes (33 ,34 ). It is possible that SOX proteins display bi-functionality, with the potential to act both as activators and as repressors of target gene expression, depending on interactions with other DNA-binding factors. Although the specific functions of each of these domains found in SOX proteins still remain to be clarified, such explicit sequence conservation among different proteins strongly suggests that they are likely to play essential roles in the function of SOX22. In addition, the analysis of these conserved sequences could refine our understanding of the phylogeny of the SOX family that, to date, has been based essentially on the HMG motif. It is also noteworthy that SOX4, SOX22 and SOX11 form a group of SOX proteins that are the most distantly related to SRY since their HMG boxes display respectively 53, 53 and 56% homology with that of SRY. Interestingly, despite their strong overall homology, these three SOX genes display markedly different expression patterns, SOX11 being expressed mainly in the nervous system (14 ) and SOX4 in the testis, brain and heart (31 ), suggesting that, after duplication, these genes adopted distinct fates.
SOX22 gene was mapped to chromosome 20 on the p13 band and is thus not clustered in the human genome either with SOX4 which maps on 6p21.3-pter (31 ) or with SOX11 which maps on 2p25 (14 ). Screening of GDB for 20p13 did not reveal any disease linked to this particular region. It is noteworthy, however, that several polymalformative syndromes were associated with deletions in the short arm of chromosome 20 (35 ). The chromosomal localization of SOX22 is close to, but probably distinct from, the region deleted in Alagille syndrome (36 ). It is also possible that alterations in the chromosomal region of SOX22 result in precocious death of embryos due to the early onset of SOX22 gene expression during development. This hypothesis is consistent with the recent report concerning disruption of murine Sox-4, which leads to death of nullisomic mouse embryos at E14 after generalized oedema caused by improper development of the semilunar valves (37 ).
Most of the SOX genes characterized so far are expressed predominantly in a specific organ or tissue where they are believed to regulate transcription of their cognate targets. Northern blotting analysis of multiple fetal and adult tissues with a SOX22-specific probe showed that the corresponding messengers were distributed in various fetal and adult tissues. This suggests first that the function of this gene is not restricted to early developmental stages and second that SOX22 regulates a set of genes that are present in several organs and that this expression pattern is representative of a unique mechanism, common to these organs. An alternative possibility is that SOX22 trans-regulates distinct pools of targets according to the tissues in which it is expressed, most likely in conjunction with additional tissue-specific factors. More detailed analysis of SOX22 expression by in situ hybridization revealed that, in embryos, SOX22 is expressed in the CNS but not in the surface ectoderm and not in many cephalic neural crest derivatives. This indicates that SOX22 might serve as a switch during the differentiation of the primitive neuro-ectoderm into the surface ectoderm, the neural tube and the neural crest. It might even play a role, yet to be explored, during the differentiation of the cephalic neural crest since SOX22 is expressed in cranial nerve ganglia which have some neurons and glial cells deriving from the cephalic neural crest. SOX22 is also expressed in many but not all mesodermic derivatives and in some endodermic derivatives. It is of outstanding importance to notice that the notochord does not express this gene at the developmental stages to which we have access. This restricted pattern of gene expression in early human embryos strongly suggests that SOX22 is involved in precocious cell fate decisions during early human embryogenesis. SOX22 expression in the CNS is likely to overlap with that of other members of the family of SOX genes. Chicken cSox2, cSox3 and cSox11 have already been shown to display overlapping expression patterns and are supposed to fulfill interacting functions in the developing nervous system (13 ). Will further studies lead to the emergence of a `Sox code' for body patterning?
The strong SOX22 gene expression in the 25-week-old human fetal brain implicates SOX22 as a candidate gene for an as yet unmapped mental retardation associated with polymalformative abnormalities affecting mainly mesenchymatous derivatives. On the other hand, the expression pattern in adult tissues indicates that SOX22 function is not restricted to the patterning of embryonic and fetal tissues and that it probably also contributes to the maintenance of several different cell types throughout adult life. Proof of such roles, however, will ultimately require a functional study in transgenic mice using the mouse homologue in the analysis of loss-of-function effects.
A total of 106 clones of a human leukocyte genomic library X4Y[lambda]GEM were screened with the HMG box of human SOX11 (14 ). Hybridizations were carried out in 6* SSC, 4.6* Denhardt's, 0.5% SDS. Filters were washed in 1* SSC, 0.1% SDS at 65oC. Positive clones were purified according to standard procedures (38 ).
SOX22 cDNAs were isolated from a human fetal brain cDNA library in [lambda]ZapII (Stratagene) with a genomic probe derived from sub-clone pKSSOX22. A total of 5*105 clones were screened, and 20 positive clones were excised according to the Stratagene protocol to obtain pBluescript plasmids. All the inserts were analysed by 3' and 5' partial sequencing. The two largest overlapping clones were then selected for further analysis and full sequencing.
Sequencing was performed with the Taq Dye Deoxy Terminator or Taq Dye Primer cycle sequencing Prism kits on an ABI 373A automatic DNA sequencer (Applied Biosystems, Inc, Foster City, CA) according to the manufacturer's instructions.
Genomic and cDNA clones were analysed using restriction fragment subcloning and gene walking strategies, and both strands were sequenced completely at least once. Further sequence analysis was performed using the GCG sequence analysis software package (University of Wisconsin, Madison, WI).
Metaphases were obtained from phytohaemagglutinin-stimulated blood lymphocytes of a healthy donor after thymidine synchronization and bromodeoxyuridine incorporation (39 ). The 3.5 kb HindIII genomic fragment was labelled with biotinyl-16 dUTP by random priming. FISH was performed as previously described (40 ).
Multiple tissue Northern blots (Clontech fetal blot and adult blots I and II) were hybridized sequentially with the radiolabelled 2 kb NotI-HindIII fragment of pKSSOX22 and human [beta]-actin cDNA probe (provided with the blots) according to the instructions of the supplier.
Morphologically normal human embryos ranging from 3 to 8 (post-ovulatory) weeks were obtained from legal abortions triggered by Mifepristone (RU 486) at Hôpital Broussais in Paris. Complete independence was respected between the medical staff of the hospital and the research group. Maternal consent was never asked for before the abortion was decided and performed. The consent was obtained after thorough explanation of the planned research, and only after medical abortion had been completed. This procedure was approved by the Ethical Committee of the Necker-Enfants malades Hôpital (Paris, France). The embryos were microdissected from the whole trophoblast under a dissecting microscope. The developmental stage of each embryo was estimated following the Carnegie classification established by O'Rahilly (41 ,42 ). The experiments were carried out on 20 embryos: one stage 11 embryo (day 23-day 26), three stage 12 embryos (day 26-day 30), two stage 13 embryos (day 28-day 32), three stage 15 embryos (day 35-day 38), five stage 16 embryos (day 37-day 42), three stage 17 (day 42-day 44), two stage 18 embryos (day 45-day 48) and one stage 20 embryo (day 51-day 53). Microdissected embryos were frozen using powdered dry ice, and stored at -80oC. Two 25-week-old fetal brains were obtained after a legally performed pregnancy interruption decided for serious medical reasons. The fetal brains were dissected and then quickly frozen using powdered ice. Cryostat sections (15 [mu]m) were processed as described (43 ). Briefly, sections were mounted onto slides previously coated with 2% 3-aminopropyl triethoxylane solution in acetone. Sections were fixed for 30 min in 2% paraformaldehyde with 0.1 M phosphate buffer (pH 7.4), rinsed once in phosphate-buffered saline (PBS), rinsed briefly in water and dehydrated with a series of ethanol concentrations (50, 75, 95%). Sections were then air-dried and finally stored at -80oC. This procedure was used to preserve the mRNAs in embryonic tissues.
The 60mer oligonucleotide probes were synthesized and purified by Genset, France. They were 3' end labelled with [[alpha]-35S]dATP (NEN) using terminal deoxyribonucleotidyl transferase (Amersham) at a specific activity of ~7*108 c.p.m./mg. The probes were purified on biospun columns (Biorad) before use. The sequence of antisense SOX22-2 was 5'-GGGAAGGAGCAGCAGGGATC TGTCCCCAGACAGACTGGTAAGGTGCCTTGGTTTTTAGT-3'. The sequence of antisense SOX22-3 was 5'-GCTATCCACTTGGG- AAGTTTCAGGCCCAAACCAACCCTTCCGAGTGCTCCTCCCATTA-3'. The sense probes were used as controls.
The hybridization cocktail contained 50% formamide, 4* SSC (standard saline citrate), 1* Denhardt's solution, 0.25 mg/ml yeast tRNA, 0.25 mg/ml sheared herring sperm DNA, 0.25 mg/ml poly(A), 10% dextran sulfate (Sigma), 100 mmol dithiothreitol (DTT) and [[alpha]-35S]dATP-labelled probe (6*106 c.p.m./ml). Then 100 ml of hybridization solution was placed on each section. The sections were then covered with a parafilm coverslip and incubated in a humidified chamber at 43oC for 20 h. After hybridization, the slides were washed twice in 1* SSC containing 10 mM DTT for 15 min each at 55oC, twice in 0.5* SSC containing 10 mM DTT for 15 min each at 55oC and finally in 0.5* SSC containing 10 mM DTT for 15 min at room temperature. The sections were then dipped in water, dehydrated with a series of graded concentrations of ethanol and exposed to Amersham Hyperfilm betamax X-ray films for 2 days and then to Kodak NTB2 photographic emulsion for 6 weeks at 4oC. The present report displays the autoradiographic labelling patterns of the SOX22 mRNA.
DNA purifications from agarose gels were performed using the Prep-A-Gene DNA purification system (Biorad). DNA templates for sequencing were prepared using plasmid isolation kits (Qiagen). All primers were purchased from Eurogentec SA (Belgium).
We thank Professor J. Demaille and Professor J.-L. Dufier for their continuous support, and May Morris for corrections to the manuscript. This work was supported by Centre National de la Recherche Scientifique, Institut National de la Santé et de la Recherche Médicale, and by grants GREG 30/95 and ACCSV1 No9501014 to PB. M.A. was supported by Association Française contre la Retinitis Pigmentosa, Association Claude Bernard, Fondation pour la Recherche Médicale and Association Valentin Haüy pour le Bien des Aveugles.
1 Jantzen, H.-M., Admon, A., Bell, S.P. and Tjian, R. (1990) Nucleolar transcription factor hUBF contains a DNA-binding motif with homology to HMG proteins. Nature, 344, 830-836.
2 Laudet, V., Stehelin, D. and Clevers, H. (1993) Ancestry and diversity of the HMG box superfamily. Nucleic Acids Res., 21, 2493-2501.MEDLINE Abstract
3 Gubbay, J., Collignon, J., Koopman, P., Capel, B., Economou, A., Münsterberg, A., Vivian, N., Goodfellow, P.N. and Lovell-Badge, R. (1990) A gene mapping to the sex-determining region of the mouse Y chromosome is a member of a novel family of embryonically expressed genes. Nature, 346, 245-346.MEDLINE Abstract
4 Denny, P., Swift, S., Brand, N., Dabhade, N., Barton, P. and Ashworth, A. (1992) A conserved family of genes related to the testis determining gene, SRY. Nucleic Acids Res., 20, 2887.MEDLINE Abstract
5 Coriat, A.-M., Müller, U., Harrry, J.L., Uwanogho, D. and Sharpe, P.T. (1993) PCR amplification of SRY-related gene sequences reveals evolutionary conservation of the SRY-box motif. PCR Methods Appl., 2, 218-222.
6 Griffiths, R. (1991) The isolation of conserved DNA sequences related to the human sex-determining region Y gene from the lesser black-backed gull (Larus fuscus). Proc. R. Soc. Lond. B., 244, 123-128.
7 Griffiths, R. and Tiwari, B. (1993) Primers for the differential amplification of the sex-determining region Y gene in a range of mammals species. Mol. Ecol., 2, 405-406.MEDLINE Abstract
8 Wright, E.M., Snopek, B. and Koopman, P. (1993) Seven new members of the Sox gene family expressed during mouse development. Nucleic Acids Res., 21, 744.MEDLINE Abstract
9 Chardard, D., Chesnel, A., Gozé, C., Dournon, C. and Berta, P. (1993) Pw Sox-1: the first member of the Sox gene family in Urodeles. Nucleic Acids Res., 21, 3576.MEDLINE Abstract
10 Gozé, C., Poulat, F. and Berta, P. (1993) Partial cloning of SOX-11 and SOX-12, two new human SOX genes. Nucleic Acids Res., 21, 2943.MEDLINE Abstract
11 van de Wetering, M. and Clevers, H. (1993) Sox 15, a novel member of the murine Sox family of HMG box transcription factors. Nucleic Acids Res., 21, 1669.MEDLINE Abstract
12 Spotila, L.D., Kaufer, N.F., Theriot, E., Ryan, K.M., Penick, D. and Spotila, J.R. (1994) Sequence analysis of the ZFY and Sox genes in the turtle Chelydra serpentina. Mol. Phylogenet. Evol., 3, 1-9.MEDLINE Abstract
13 Uwanogho, D., Rex, M., Cartwright, J., Pearl, G., Healy, C., Scotting, P.J. and Sharpe, P.T. (1995) Embryonic expression of the chicken Sox2, Sox3 and Sox11 genes suggests an interactive role in neuronal development. MD, 49, 23-36.
14 Jay, P., Gozé, C., Marsollier, C., Taviaux, S., Hardelin, J.-P., Koopman, P. and Berta, P. (1995) The human SOX11 gene: cloning, chromosomal assignment ant tissue distribution. Genomics, 29, 541-545.MEDLINE Abstract
15 Connor, F., Wright, E., Denny, P., Koopman, P. and Ashworth, A. (1995) The SRY-related HMG box-containing gene Sox6 is expressed in the adult testis and developing nervous system of the mouse. Nucleic Acids Res., 23, 3365-3372.MEDLINE Abstract
16 Wright, E., Hargrave, M.R., Christiansen, J., Cooper, L., Kun, J., Evans, T., Gandadharan, U., Greenfield, A. and Koopman, P. (1995) The Sry-related gene Sox9 is expressed during chondrogenesis in mouse embryos. Nature Genet., 9, 15-20.MEDLINE Abstract
17 van de Wetering, M. and Clevers, H. (1992) Sequence-specific interaction of the HMG box proteins TCF-1 and SRY occurs within the minor groove of a Watson-Crick double helix. EMBO J., 11, 3039-3044.MEDLINE Abstract
18 Ferrari, S., Harley, V.R., Pontiggia, A., Goodfellow, P.N., Lovell-Badge, R. and Bianchi, M.E. (1992) SRY, like HMG1, recognizes sharp angles in DNA. EMBO J., 11, 4497-4506.MEDLINE Abstract
19 Giese, K., Cox, J. and Grosschedl, R. (1992) The HMG domain of lymphoid enhancer factor 1 bends DNA and facilitates assembly of functional nucleoprotein structures. Cell, 69, 185-195.MEDLINE Abstract
20 Russell, S.R.H., Sanchez-Soriano, N., Wright, C.R. and Ashburner, M. (1996) The Dichaete gene of Drosophila melanogaster encodes a SOX-domain protein required for embryonic segmentation. Development, 122, 3669-3676.
21 van de Wetering, M., Oosterwegel, M., van Norren, K. and Clevers, H. (1993) Sox-4, an Sry-like HMG box protein, is a transcriptionnal activator in lymphocytes. EMBO J., 12, 3847-3854.MEDLINE Abstract
22 Hosking, B.M., Muscat, G.E.O., Koopman, P.A., Dowhan, D.H. and Dunn, T.L. (1995) Trans-activation and DNA-binding properties of the transcription factor, Sox-18. Nucleic Acids Res., 23, 2626-2628.MEDLINE Abstract
23 Südbeck, P., Schmitz, L., Baeuerle, P.A. and Scherer, G. (1996) Sex reversal by loss of the C-terminal transactivation domain of human SOX9. Nature Genet., 13, 230-232.MEDLINE Abstract
24 Kanai, Y., Kanai-Azuma, M., Noce, T., Saido, T.C., Shiroishi, T., Hayashi, Y. and Yazaki, K. (1996) Identification of two Sox17 messenger RNA isoforms, with and without the high mobility group box region, and their differential expression in mouse spermatogenesis. J. Cell Biol., 133, 667-681.MEDLINE Abstract
25 Wotton, D., Lake, R.A., Farr, C.J. and Owen, M.J. (1995) The high mobility group transcription factor, SOX4, transactivates the human CD2 enhancer. J. Biol. Chem., 270, 7515-7522.MEDLINE Abstract
26 Lefebvre, V., Huang, W., Harley, V.R., Goodfellow, P.N. and de Crombrugghe, B. (1997) SOX9 is a potent activator of the chondrocyte-specific enhancer of the proa1(II) collagen gene. Mol. Cell. Biol., 17, 2336-2346.MEDLINE Abstract
27 Yuan, H., Corbi, N., Basilico, C. and Dailey, L. (1995) Developmental-specific activity of the FGF-4 enhancer requires the synergetic action of Sox2 and Oct-3. Genes Dev., 9, 2635-2645.MEDLINE Abstract
28 Pontiggia, A., Rimini, R., Harley, V.R., Goodfellow, P.N., Lovell-Badge, R. and Bianchi, M.E. (1994) Sex-reversing mutations affects the architecture of SRY-DNA complexes. EMBO J., 13, 6115-6124.MEDLINE Abstract
29 Mount, S.M. (1982) A catalogue of splice junction sequences. Nucleic Acids Res., 10, 459-472.MEDLINE Abstract
30 Kozak, M. (1987) An analysis of 5'-noncoding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res., 15, 8125-8148.MEDLINE Abstract
31 Farr, C.J., Easty, D.J., Ragoussis, J., Collignon, J., Lovell-Badge, R. and Goodfellow, P.N. (1993) Characterization and mapping of the human SOX4 gene. Mamm. Genome, 4, 577-584.MEDLINE Abstract
32 Triezenberg, S.J. (1995) Structure and function of transcriptional activation domains. Curr. Opin. Genet. Dev., 5, 190-196.MEDLINE Abstract
33 Williams, F.E. and Trumbly, R.J. (1990) Characterization of Tup1, a mediator of glucose repression in Saccharomyces cerevisiae. Mol. Cell. Biol., 10, 6500-6511.MEDLINE Abstract
35 Anad, F., Burn, J., Matthews, D., Cross, I., Davison, B.C., Mueller, R., Sands, M., Lillington, D.M. and Eastham, E. (1990) Alagilles syndrome and deletion of 20p. J. Med. Genet., 27, 729-737.MEDLINE Abstract
36 Rand, E.B., Spinner, N.B., Piccoli, D.A., Whitington, P.F. and Taub, R. (1995) Molecular analysis of 24 Alagilles syndrome families identifies a single submicroscopic deletion and further localizes the Alagilles region within 20p12. Am. J. Hum. Genet., 57, 1068-1073.MEDLINE Abstract
37 Schilham, M.W., Oosterwegel, M.A., Moerer, P., Ya, J., de Boer, P.A.J., van de Wetering, M., Verbeek, S., Lamers, W.H., Kruisbeek, A.M., Cumano, A. and Clevers, H. (1996) Defects in cardiac outflow tract formation and pro-B-lymphocyte expansion in mice lacking Sox-4. Nature, 380, 711-714.MEDLINE Abstract
38 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
39 Viegas-Péquignot , E. and Dutrillaux , B. (1978) Une méthode simple pour obtenir des prométaphases. Ann. Genet. (Paris), 21, 121-125.
40 Taviaux, S.A. and Demaille, J.G. (1993) Localization of human cell cycle regulatory genes CDC25C to 5q31 and WEE1 to 11p15.3-11p15.1 by fluorescence in situ hybridization. Genomics, 15, 194-196.
41 O'Rahilly, R. and Müller, F. (1987) Developmental stages in human embryos including a revision of Streeter's `Horizons', and a survey of the Carnegie collection. Carnegie Institution of Washington, publication no. 637.
42 O'Rahilly, R. and Müller, F. (1994) The Embryonic Human Brain, An Atlas of Developmental Stages. Wiley-Liss, Inc., New York.
43 Abitbol, M., Menini, C., Delezoide, A.-L., Rhyner, T., Vekemans, M. and Mallet, J. (1993) Nucleus basalis magnocellularis and hippocampus are the major sites of FMR-1 expression in the human fetal brain. Nature Genet., 4, 147-153.MEDLINE Abstract
44 Kyte, J. and Doolittle, R.F. (1982) A simple method for displaying the hydropathic character of a protein. J. Mol. Biol., 157, 105-132.MEDLINE Abstract
*To whom correspondence should be addressed. Tel: +33 467 61 3349; Fax: +33 467 52 1559; Email: berta{at}phindy.crbm.cnrs-mop.fr
-->
This page is maintained by OUP admin. Last updated Tue Jun 10 19:01:42 BST 1997. Part of the OUP Journals World Wide Web service.
Copyright
Oxford University Press, 1996
