Hum Genet (2000) 107 : 1–6 Digital Object Identifier (DOI) 10.1007/s004390000328

R A P I D C O M M U N I C AT I O N

S. Brian Potterf · Minao Furumura · Karen J. Dunn · Heinz Arnheiter · William J. Pavan

Transcription factor hierarchy in Waardenburg syndrome: regulation of MITF expression by SOX10 and PAX3

Received: 27 April 2000 / Accepted: 12 May 2000 / Published online: 21 June 2000 © Springer-Verlag 2000

Abstract Waardenburg syndrome (WS) is associated with neural crest-derived melanocyte deficiency caused by mutations in either one of three transcription factors: MITF, PAX3, and SOX10. However, the hierarchical relationship of these transcription factors is largely unknown. We show that SOX10 is capable of transactivating the MITF promoter 100-fold, and that this transactivation is further stimulated by PAX3. Promoter deletion and mutational analyses indicate that SOX10 can activate MITF expression through binding to a region that is evolutionarily conserved between the mouse and human MITF promoters. A SOX10 mutant that models C-terminal truncations in WS can reduce wild-type SOX10 induction of MITF, suggesting these mutations may act in a dominant-negative fashion. Our data support a model in which the hypopigmentation in WS, of which these factors have been implicated, results from a disruption in function of the central melanocyte transcription factor MITF.

Introduction The mechanisms governing development of neural crestderived melanocytes and the way in which alterations in these pathways lead to hypopigmentation and deafness disorders are not completely understood. Waardenburg

S.B. Potterf · K.J. Dunn · W.J. Pavan (✉) Genetic Disease Research Branch, National Human Genome Research Institute, National Institutes of Health, 49 Convent Drive MSC4472, Bethesda, MD 20892-4472, USA e-mail: [email protected], Tel.: +1-301-4967584, Fax: +1-301-4022170 M. Furumura Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD, 20892, USA H. Arnheiter Laboratory of Developmental Neurogenetics, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, 20892, USA

syndrome (WS types I–IV; Holme and Steel 1999; Read and Newton 1997; Tachibana 1999) is associated with hypopigmentation and incompletely penetrant deafness resulting from a neural crest-derived melanocyte deficiency, attributable to the failure of melanoblasts to develop normally. WS is caused by mutations in either one of three transcription factors: SOX10 (WSIV OMIM 277580), MITF (WSIIa OMIM 193510), and PAX3 (WSI OMIM 193500 and WSIII OMIM 148820). SOX10 is a member of the SRY (sex-determining factor)-like, high-mobility group (HMG) DNA binding proteins that are involved in cell lineage determination (Wegner 1999). Several mutations in the human and mouse SOX10 gene have been found to result in reduction in pigment-producing cells (melanocytes; Southard-Smith et al. 1998), varying levels of hearing impairment, aganglionic megacolon (WSIV; Bondurand et al. 1999; Pingault et al. 1998; Pusch et al. 1998; Southard-Smith et al. 1999), and dysmyelination (Inoue et al. 1999). The hypopigmentation and deafness phenotypes resulting from these SOX10 mutations resemble those observed in individuals with PAX3 and MITF mutations. Several lines of evidence suggest a possible functional relationship between these transcription factors in neural crest-derived melanocyte development. Paired-domain protein PAX3 (Chalepakis et al. 1994) is first detected in the dorsal neural tube prior to neural crest emigration (Gruss and Walther 1992), and SOX10 is expressed in early emigrating neural crest cells (Herbarth et al. 1998; Pusch et al. 1998; Southard-Smith et al. 1998). Expression of the basic helix-loop-helix protein MITF is critical for the development of melanocytes (Opdecamp et al. 1997) and occurs subsequent to PAX3 and SOX10 in melanoblasts located dorsal to the neural tube and migrating along the dorsolateral pathway (Hodgkinson et al. 1998; Nakayama et al. 1998). It has been demonstrated in vitro that PAX3 may affect melanocyte development by regulating MITF expression (Watanabe et al. 1998). Furthermore, synergistic interactions between SOX10 and paired domain proteins have been demonstrated using an artificial promoter containing consensus SOX and PAX

2

binding sites (Kuhlbrodt et al. 1998a). Therefore, based on the temporal and spatial expression patterns of these transcription factors, the similarity in the phenotypic malformations present in WS, and the interactions of these factors in vitro, we sought to determine whether SOX10 has a role in the control or activation of MITF expression.

• SOX consensus probe (van de Wetering et al. 1993): 5′ ctagaGATCCGCGCCTTTGTTCTCCCCAT • MITF promoter wild-type SX2/3: 5′ cgcgTATTAACCTATTGCTGAAAGAGAAATACCATTGTCT • MITF promoter wild-type SX2/mutant SX3: 5′ cgcgTATTAACCTATTGCTGAAAGAGAAATACTGGCAGCT. None of the putative PAX3 binding sequences are present in these probes. Samples were run on a 6% acrylamide gel at 4°C, dried, and bands were visualized by audioradiography.

Materials and methods Sox10Dom/+ mice were obtained from the Jackson laboratory and were maintained on a C57BL6/C3HeB/FeJLe-a/a background. Mitfmi/+ mice were obtained from the Jackson Laboratory on a C57BL6/C3Fe-a/a background. Genotypes of hypopigmented progeny were determined by PCR (Southard-Smith et al. 1998) and by test-crossing to Mitfmi/+ mice. NIH guidelines for animal care were followed. Sox10 and Sox10 deletion expression vectors were constructed by placing murine SOX10 cDNA (Southard-Smith et al. 1998) under the control of the CMV promoter in pcDNA3.1 (Invitrogen), resulting in pcDNA3.1-SOX10 and pcDNA3.1-SOX10del. The Sox10 deletion variant (C190X) of pcDNA3.1-SOX10 truncates the SOX10 protein sequence at residue 190 by introducing a C to Y change and subsequent addition of 8 extraneous residues (LDPRSFPW). The V5 epitope HIS-tagged Sox10 expression vector was generated by a creating a carboxyl-terminal deletion of 21 amino acids (codons 443–463) followed by in frame insertion of Sox10 into pcDNA6/V5-HIS (Invitrogen). The pCEV27-PAX3 expression vector and 2.2 kb and deleted pMITF luciferase reporter vectors (pMITF-2256 and pMITF-382, 263, and 46) have been published previously (Watanabe et al. 1998) and were obtained from M. Tachibana. The pCEV27-PAX3 contains the PAX3c splice form (Barber et al. 1999), generated by RT-PCR from 624-mel cells (Watanabe et al. 1998). Targeted and deletion mutants of the MITF promoter were generated by PCR-based methods. All mutations were verified by DNA sequencing. HeLa cells were maintained and grown in DMEM supplemented with 10% FBS and 10 U/ml penicillin/streptomycin (growth medium) and were kept in 5% CO2. Cells were routinely passaged every 3–4 days and discarded after 15 passages. HeLa cells were transiently transfected using SuperFect (Qiagen). Twenty-four hours before transfection, cells were seeded at an approximate density of 1×104/well in 24-well tissue culture plates (Costar). Each transfection mixture contained a total of 1.5 µg DNA with 0.5 µg comprised by the reporter construct. Renilla luciferase (Promega) DNA served as internal control. DNA/SuperFect mixtures were incubated in serum- and antibiotic-free DMEM at room temperature for 10 min. Growth medium was then added to the mixtures and the combined solutions were put onto cells for 2 h before removal by aspiration. Cells were washed with PBS and grown 24 h in growth medium before harvesting. Reporter assays were performed using the dual luciferase assay system (Promega). Luciferase activities were determined using a Lumat LB 9507 luminometer (EG&G Berthold). All experiments were conducted at least three times and performed in triplicate on different days using different batches of cells. SOX10 and SOX10HIS-tagged proteins were prepared by in vitro transcription/translation (IVTT) reactions using the TNT T7 coupled Reticulocyte Lysate System kit (Promega). Binding reactions were carried out in 10 mM HEPES pH 7.9, 50 mM NaCl, 5 mM MgCl2, 0.5 mM EDTA, 0.1 mM DTT, 0.1 µg poly dI.dC (Pharmacia), 5% glycerol, 1 mg/ml BSA to which 3×104 cpm [32P]dCTP (Amersham) labeled probe and 2 µl lysate were added. Binding was carried out on ice for 30 min. For competition reactions nonradiolabeled competitor DNA (25 and 125 pmol) was incubated with IVTT SOX10 on ice for 30 min before addition of radiolabeled probe. For supershifts involving SOX10His-tagged protein, binding reactions were carried out as above except that 1 µg anti-V5 antibody (Invitrogen) was included and preincubated on ice for 30 min before addition of radiolabeled probe:

Results and discussion In vivo genetic interaction between Sox10 and Mitf loci To investigate the relationship between SOX10 and MITF, crosses between mice with spontaneous mutations in these two genes (Sox10Dom/+ and Mitfmi/+) were used to determine whether synergistic interactions between these loci occur in vivo. Mice with mutations in either SOX10 or MITF (Sox10Dom/+ and Mitfmi/+) alone are almost completely pigmented with a variable white ventral spot. Double heterozygotes (Sox10Dom/+; Mitfmi/+), however, demonstrate extensive ventral and dorsal hypopigmentation (Fig. 1). No overt craniofacial or ocular defects were observed in these mice, which are typically found in PAX3 and MITF mutations, respectively. This finding is consistent with the lack of SOX10 expression in these tissues (Bondurand et al. 1998; Herbarth et al. 1998;

Fig. 1 Synergistic interactions between the SOX10 and MITF loci in vivo. On their respective backgrounds, Sox10Dom/+ mice exhibit a variable white ventral spot and white feet; Mitfmi/+ mice occasionally had a small white ventral spot. Intercrosses between these two types of mice produced progeny with extensive hypopigmentation, which genotypically were double heterozygotes

3 Fig. 2A,B Human-mouse sequence conservation in the MITF promoter. A The 2256bp sequence containing the melanocyte isoform of the human MITF promoter sequence (gi 2935699, x-axis) was compared to the corresponding murine MITF promoter sequence available from BAC RP23-271N22 (gi 6751633, y-axis) using Pustell DNA Matrix analysis (MacVector). Numbers on the axes represent base pair positions relative to the MITF transcription start site (basepair +1). Each black dot A 30-bp sequence of DNA that has greater than 65% sequence identity between the mouse and human regions. Extensive similarity was observed in two regions, 1.1 kb upstream and 382 bp upstream of the transcriptional start site. B Potential binding sites in the MITF promoter. The –314 to –159 portion of the MITF promoter containing potential PAX3, SX (i.e., SOX-like) and cAMP responsive element (CRE; Price et al. 1998) binding sites is shown

Southard-Smith et al. 1998). These results demonstrate an in vivo synergistic relationship between SOX10 and MITF that is restricted to effects on neural crest-derived melanocyte development. Given the synergistic interaction between these two loci in vivo, we sought to determine whether SOX10 could activate the MITF promoter. Comparative sequence analysis of the MITF promoter To identify potentially important MITF promoter regions, we first compared the genomic sequences upstream of the melanocyte transcriptional start site in both human (Hu gi2935699) and mouse (Mu gi6751633) MITF promoters. This analysis revealed extensive sequence similarities between position –1069 and –1214 of the human promoter and within a 382-bp region proximal to the transcription start site (Fig. 2A). Within the proximal conserved region, we identified at least four potential SOX (“SX”) binding sites (Wegner 1999; van de Wetering et al. 1993), three of which contain 100% sequence homology between human and mouse melanocyte-specific MITF promoters (SX2– SX4, Fig. 2B), and flank the 100% conserved PAX3 binding region. Because of these similarities and the previous characterization of the human MITF promoter (Fuse et al. 1996; Watanabe et al. 1998) we tested the role of SOX10 and PAX3 on a 2.2-kb human promoter fragment linked to a luciferase reporter gene (pMITF-2256).

Sox10 and Pax3 activate MITF promoter activity in vitro Cotransfection with a SOX10 expression vector resulted in a 100-fold increase in MITF promoter activity (Fig. 3A). Cotransfection with PAX3 resulted in a fivefold increase, consistent with previous results (Watanabe et al. 1998). When SOX10 and PAX3 were transfected together, greater than additive stimulation was obtained, indicating that SOX10 and PAX3 act synergistically on this promoter fragment. Furthermore, a SOX10 deletion mutant (SOX10del), lacking the transactivation domain, failed to activate the MITF promoter, and synergistic actions with PAX3 were lost (Fig. 3A). The SOX10 stimulation of MITF promoter activity was also observed in other non-neural crest-derived cell lines (NIH3T3, 293T, not shown) and in a murine melanocyte line, melan-a (Bennett et al. 1987; data not shown). Since several potential SOX10 and PAX3 binding sites were localized within the first 382-bp of the MITF promoter, we also tested this smaller promoter fragment. In fact, this fragment was still significantly transactivated by SOX10, and the synergistic action between SOX10 and PAX3 was retained (Fig. 3B). These results indicate that sequences important for the SOX10 induction are located within the proximal 382-bp conserved segment of the MITF promoter. However, based on the reduction in activity obtained with the 382-bp fragment (Fig. 3B), it is likely that an additional SOX10 responsive element(s) is present in the region deleted from the 2.2-kb human promoter fragment.

4

ity was seen with deletions just 5′ to the PAX3 binding site. The removal of all sequences 5′ to SX4 and the cAMP responsive element (Price et al. 1998) virtually abolished stimulation by SOX10 (Fig. 4A). To further determine the relevance of each of the four potential SOX10 binding sites base substitutions were introduced separately into each site. Significant decreases in activity were associated with mutations in SX2 and SX3; however, mutations in SX1 had a minimal effect and mutations in SX4 failed to reduce activity (Fig. 4B). Sox10 can bind an evolutionarily conserved site in the MITF promoter Consequently we assessed the ability of SOX10 to bind SX2 and SX3. Electrophoretic mobility shift assays using Fig. 3A, B Transient transfection of HeLa cells. A Equal amounts of pMITF-2256 (MITF promoter) luciferase reporter and pcDNA3.1-MSOX10, pCEV27-HPAX3, or pcDNA3.1MSOX10del expression vectors (0.5 µg each) were used in each experiment. Results represent the mean of triplicate samples and are expressed as fold increase relative to a promoterless pGL2basic reporter plasmid. Error bars One standard deviation from the mean. Cotransfection with a human SOX10 expression vector resulted in greater than 10- and 30-fold increases in MITF activity in a murine melanocyte cell line, melan-a (Bennett et al. 1987), and HeLa cells, respectively (data not shown). B Transient transfections with pMITF-382 luciferase reporter and pcDNA3.1MSOX10 and/or pCEV27-HPAX3

To determine which of the four potential SOX10 binding sites was functionally important in the 382-bp conserved segment of the MITF promoter we first generated 5′ deletion mutants and tested them for response to SOX10 (Fig. 4A). The greatest relative decrease in activFig. 4A–D SOX10 induction of MITF promoter deletions. A Reporter vector names denote the base site termination position (e.g., pMITF-258 contains promoter sequence up to –258 of the MITF transcriptional start site). Reporter activities were expressed as a percentage of control SOX10 induction of wild-type 382 bp MITF promoter (set at 100%). B Mutational analysis of potential SOX10 binding sites. Lowercase Base substitutions in MITF promoter targeted mutations. SX1 TCCAAAG to SX1mut TggAatc (–314– 308); SX2 TGAAAGAGAAA to SX2mut TGgcgGccgcA (–282– 272); SX3 CATTGTC to SX3mut tggcagC (–268–262); SX4 CTTGATC to SX4mut CggccgC (–221–215). Induction values by SOX10 are expressed as a percentage of wild-type pMITF-382 luciferase reporter activity upon induction by SOX10 (100%). C EMS assays demonstrating SOX10 binding to MITF promoter sequences. Lane 1 Lysate (untreated) with SOX consensus probe; lanes 2–4 SOX consensus probe plus IVTT SOX10 and increasing amounts of unlabeled probe; lane 5 lysate (untreated) plus wildtype SX2/3 probe; lanes 6–8 wild-type MITF SX2/3 probe plus IVTT SOX10 with increasing amounts of unlabeled probe; lane 9 lysate (untreated) with MITF SX2/mutSX3 probe (SX3 site mutated); lane 10 lysate containing MITF SX2/mutSX3 probe plus IVTT SOX10. D EMS assay with SOX10HIS-tagged protein containing the V5 epitope. All lanes contain wild-type MITF SX2/3 probe. Lane 1 Plus IVTT SOXHIS-tagged protein; lane 2 same as lane 1, with anti-V5 antibody addition; lane 3 lysate (untreated) plus anti-V5 antibody alone

5

appropriate wild-type and mutated probes mixed with IVTT SOX10 showed that SOX10 bound a consensus SOX sequence (van de Wetering et al. 1993) (Fig. 4C, lane 2) and a probe containing the SX2 and SX3 sequences (wt SX2/3) from pMITF (Fig. 4C, lane 6). This binding was competed with unlabeled corresponding DNA (Fig. 4C, lanes 3 and 4 and lanes 7–8, respectively). Figure 4C (lane 10) shows that IVTT SOX10 failed to bind a probe containing wild-type SX2 and mutated SX3 (SX2/mutSX3). Correspondingly, unlabeled SX2/mutSX3 competitor failed to strongly compete for SOX10 binding with labeled wild-type SX2/3 probe (data not shown). That SOX10 was in fact part of the protein-SX2/3 complex was confirmed by supershift using an antibody to the V5 epitope of a SOX10 HIS-tagged protein (Fig. 4D, lane 2). Taken together, these results indicate that SOX10 transactivates the melanocyte-specific MITF promoter by binding at least the SX3 site. Gene expression and transgenic analyses are needed to test the relevance of this hypothesis to melanocyte development in vivo. C-terminal SOX10 mutations can act as dominant negative proteins Nine dominant mutations in SOX10 have been associated with melanocyte defects in humans and mouse (Bondurand et al. 1999; Inoue et al. 1999; Pingault et al. 1998; Southard-Smith et al. 1999; Fig. 5A). Three mutations (Y83X, S135T, and LR ins) disrupt DNA binding or translocation to the nucleus and are thus likely loss of function mutations that result in developmental defects due to SOX10 haploinsufficiency (Kuhlbrodt et al. 1998b; Pingault et al. 1998). The remaining six alleles, however, could encode dominant-negative proteins. Five of these [E189X, Y201X, Dom (mouse), Q377X, and 1076delGA] result in truncation of varying portions of the transactivation domain (Fig. 5A). The sixth mutation (1400del12; Inoue et al. 1999) is a 12-bp deletion which results in ablation of the SOX10 translational stop codon and addition of 82 extraneous amino acids to the C-terminus of the transactivation domain. The resultant mutant protein was proposed to function in a dominant negative fashion (Inoue et al. 1999). To determine whether mutations that alter the transcription activation domain of SOX10 could act as dominant-negatives, a mutant SOX10 protein (Fig. 5A) was engineered and assessed for the ability to disrupt the wildtype SOX10 activation of the 2.2-kb MITF promoter. SOX10del (C190X) lacks the transactivation domain but leaves the amino-terminus and the HMG-DNA binding domain intact (Fig. 5A). As shown in Fig. 5B, cotransfection with SOX10del abolished induction of the MITF promoter activity by wild-type SOX10 in a concentration dependent manner, strongly suggesting this mutation functions through a dominant-negative mode of action. To examine the possibility that truncating SOX10 mutations could act as dominant negatives in vivo, further studies are needed to determine whether truncated proteins are in

Fig. 5A,B Mutation of the SOX10 transcription activation domain can act as a dominant negative. A Top Schematic representation of known SOX10 mutations in human and mouse; bottom engineered mouse SOX10 mutation deleting the entire transactivation domain. B In vitro test for loss of function or dominant negative effects with SOX10 mutants. Numbers to left Amount of expression vector DNA added (in micrograms). All experiments used MITF promoter/reporter plasmid pMITF-2256. Total DNA amounts were kept constant in all experiments by addition of empty pcDNA3.1 CMV vector or by a CMV vector expressing the β-galatosidase gene

fact expressed and stable in individuals with SOX10 mutations. The ability of subsets of SOX10 mutations to act as dominant-negatives could account for part of the variation in the phenotypic expression of the disorder in WSIV individuals, where SOX10 has been implicated. Variability may also be caused in part by inheritance of modifier loci (Southard-Smith et al. 1999). We determined that SOX10 can bind to a sequence in the MITF promoter that is immediately upstream of a previously described PAX3 binding site. The juxtaposition of these two binding sites may explain the synergistic action of SOX10 and PAX3 on MITF expression. Other examples exist of SOX and partner factor binding in close proximity on promoters of natural target genes (reviewed in Kamachi et al. 2000). In these studies the binding of both proteins may result in a more stable DNA protein complex and enhanced transcriptional activity than is observed with either factor alone. This stabilized ternary complex may result from physical interaction of the SOX protein with partner proteins and/or the DNA-bending activity of SOX which then facilitates the cooperative binding of partner proteins (De Santa Barabara et al. 1998; Ambrosetti et al. 1997). SOX10 is first expressed in migratory neural crest and is essential for the proper development of multiple lineages, including melanocytes. Our results suggest that MITF is one downstream target of SOX10 and PAX3 reg-

6

ulation, and that the hypopigmentation and deafness observed in WS individuals with mutations in these three genes are all ultimately caused by a disruption in MITF function. Acknowledgements The authors thank Masayoshi Tachibana for generously supplying a number of reporter and expression vectors. We thank members of the Pavan laboratory for critical discussions throughout this project and Pamela Schwartzburg and Leslie Biesecker for advice and careful reading of the manuscript.

References Ambrosetti DC, Basilico C, Dailey L (1997) Synergistic activation of the fibroblast growth factor 4 enhancer by Sox2 and Oct-3 depends on protein-protein interactions facilitated by a specific spatial arrangement of factor binding sites. Mol Cell Biol 17:6321–6329 Barber TD, Barber MC, Cloutier TE, Friedman TB (1999) PAX3 gene structure, alternative splicing and evolution. Gene 237: 311–319 Bennett DC, Cooper PJ, Hart IR (1987) A line of non-tumorigenic mouse melanocytes, syngeneic with the B16 melanoma and requiring a tumour promoter for growth. Int J Cancer 39:414–418 Bondurand N, Kobetz A, Pingault V, Lemort N, Encha-Razavi F, Couly G, Goerich DE, Wegner M, Abitbol M, Goossens M (1998) Expression of the SOX10 gene during human development. FEBS Lett 432:168–172 Bondurand N, Kuhlbrodt K, Pingault V, Enderich J, Sajus M, Tommerup N, Warburg M, Hennekam RC, Read AP, Wegner M, Goossens M (1999) A molecular analysis of the yemenite deaf-blind hypopigmentation syndrome: SOX10 dysfunction causes different neurocristopathies. Hum Mol Genet 8:1785– 1789 Chalepakis G, Jones FS, Edelman GM, Gruss P (1994) Pax-3 contains domains for transcription activation and transcription inhibition. Proc Natl Acad Sci U S A 91:12745–12749 De Santa Barbara P, Bonneaud N, Boizet B, Desclozeaux M, Moniot B, Sudbeck P, Scherer G, Poulat F, Berta P (1998) Direct interaction of SRY-related protein SOX9 and steroidogenic factor 1 regulates transcription of the human anti-Mullerian hormone gene Mol Cell Biol 18:6653–6665 Fuse N, Yasumoto K, Suzuki H, Takahashi K, Shibahara S (1996) Identification of a melanocyte-type promoter of the microphthalmia-associated transcription factor gene. Biochem Biophys Res Commun 219:702–707 Gruss P, Walther C (1992) Pax in development. Cell 69:719–722 Herbarth B, Pingault V, Bondurand N, Kuhlbrodt K, HermansBorgmeyer I, Puliti A, Lemort N, Goossens M, Wegner M (1998) Mutation of the Sry-related Sox10 gene in Dominant megacolon, a mouse model for human Hirschsprung disease. Proc Natl Acad Sci U S A 95:5161–5165 Hodgkinson CA, Nakayama A, Li H, Swenson LB, Opdecamp K, Asher JH Jr, Arnheiter H, Glaser T (1998) Mutation at the anophthalmic white locus in Syrian hamsters: haploinsufficiency in the Mitf gene mimics human Waardenburg syndrome type 2. Hum Mol Genet 7:703–708 Holme RH, Steel KP (1999) Genes involved in deafness. Curr Opin Genet Dev 9:309–314

Inoue K, Tanabe Y, Lupski JR (1999) Myelin deficiencies in both the central and the peripheral nervous systems associated with a SOX10 mutation. Ann Neurol 46:313–318 Kamachi Y, Uchikawa M, Kondoh H (2000) Pairing SOX off with partners in the regulation of embryonic development. Trends in Genetics 16:182–187 Kuhlbrodt K, Herbarth B, Sock E, Hermans-Borgmeyer I, Wegner M (1998a) Sox10, a novel transcriptional modulator in glial cells. J Neurosci 18:237–250 Kuhlbrodt K, Schmidt C, Sock E, Pingault V, Bondurand N, Goossens M, Wegner M (1998b) Functional analysis of Sox10 mutations found in human Waardenburg-Hirschsprung patients. J Biol Chem 273:23033–23038 Nakayama A, Nguyen MT, Chen CC, Opdecamp K, Hodgkinson CA, Arnheiter H (1998) Mutations in microphthalmia, the mouse homolog of the human deafness gene MITF, affect neuroepithelial and neural crest-derived melanocytes differently. Mech Dev 70:155–166 Opdecamp K, Nakayama A, Nguyen MT, Hodgkinson CA, Pavan WJ, Arnheiter H (1997) Melanocyte development in vivo and in neural crest cell cultures: crucial dependence on the Mitf basic-helix-loop-helix-zipper transcription factor. Development 124:2377–2386 Pingault V, Bondurand N, Kuhlbrodt K, Goerich DE, Prehu MO, Puliti A, Herbarth B, Hermans-Borgmeyer I, Legius E, Matthijs G, Amiel J, Lyonnet S, Ceccherini I, Romeo G, Smith JC, Read AP, Wegner M, Goossens M (1998) SOX10 mutations in patients with Waardenburg-Hirschsprung disease. Nat Genet 18:171–173 Price ER, Horstmann MA, Wells AG, Weilbaecher KN, Takemoto CM, Landis MW, Fisher DE (1998) alpha-Melanocyte-stimulating hormone signaling regulates expression of microphthalmia, a gene deficient in Waardenburg syndrome. J Biol Chem 273:33042–33047 Pusch C, Hustert E, Pfeifer D, Sudbeck P, Kist R, Roe B, Wang Z, Balling R, Blin N, Scherer G (1998) The SOX10/Sox10 gene from human and mouse: sequence, expression, and transactivation by the encoded HMG domain transcription factor. Hum Genet 103:115–123 Read AP, Newton VE (1997) Waardenburg syndrome. J Med Genet 34:656–665 Southard-Smith EM, Kos L, Pavan WJ (1998) Sox10 mutation disrupts neural crest development in Dom Hirschsprung mouse model. Nat Genet 18:60–64 Southard-Smith EM, Angrist M, Ellison JS, Agarwala R, Baxevanis AD, Chakravarti A, Pavan WJ (1999) The Sox10 (Dom) mouse: modeling the genetic variation of Waardenburg-Shah (WS4) syndrome. Genome Res 9:215–225 Tachibana M (1999) A cascade of genes related to Waardenburg syndrome. J Investig Dermatol Symp Proc 4:126–129 Watanabe A, Takeda K, Ploplis B, Tachibana M (1998) Epistatic relationship between Waardenburg syndrome genes MITF and PAX3. Nat Genet 18:283–286 Wegner M (1999) From head to toes: the multiple facets of Sox proteins. Nucleic Acids Res 27:1409–1420 Wetering M van de, Oosterwegel M, van Norren K, Clevers H (1993) Sox-4, an Sry-like HMG box protein, is a transcriptional activator in lymphocytes. EMBO J 12:3847–3854