Direct Regulation of the Microphthalmia Promoter by Sox10 Links Waardenburg-Shah Syndrome (WS4)-associated Hypopigmentation and Deafness to WS2

Abstract

The transcription factor Sox10 is genetically linked with Waardenburg syndrome 4 (WS4) in humans and theDominant megacolon (Dom) mouse model for this disease. The pigmentary defects observed in the Dom mouse and WS4 are reminiscent of those associated with mutations in themicrophthalmia (Mitf) gene, which encodes a transcription factor essential for the development of the melanocyte lineage. We demonstrate here that wild type Sox10 directly binds and activates transcription of the MITF promoter, whereas a mutant form of the Sox10 protein genetically linked with WS4 acts as a dominant-negative repressor of MITF expression and can reduce endogenous MITF protein levels. The ability of Sox10 to activate transcription of the MITF promoter implicates Sox10 in the regulation of melanocyte development and provides a molecular basis for the hypopigmentation and deafness associated with WS4. The transcription factor Sox10 is genetically linked with Waardenburg syndrome 4 (WS4) in humans and theDominant megacolon (Dom) mouse model for this disease. The pigmentary defects observed in the Dom mouse and WS4 are reminiscent of those associated with mutations in themicrophthalmia (Mitf) gene, which encodes a transcription factor essential for the development of the melanocyte lineage. We demonstrate here that wild type Sox10 directly binds and activates transcription of the MITF promoter, whereas a mutant form of the Sox10 protein genetically linked with WS4 acts as a dominant-negative repressor of MITF expression and can reduce endogenous MITF protein levels. The ability of Sox10 to activate transcription of the MITF promoter implicates Sox10 in the regulation of melanocyte development and provides a molecular basis for the hypopigmentation and deafness associated with WS4. Waardenburg syndrome high mobility group Dominant megacolon Microphthalmia-associated transcription factor polymerase chain reaction cytomegalovirus enhanced chemiluminescence glyceraldehyde-3-phosphate dehydrogenase green fluorescent protein wild type mutant in vitrotranscribed/translated cyclic AMP response element lymphoid enhancer factor 1 T-cell factor polymerase chain reaction glutathione S-transferase Waardenburg syndrome (WS1; deafness and pigmentary disorders) and Hirschsprung's disease (aganglionic megacolon) are inherited disorders that arise due to dysfunction of the embryonic neural crest. The symptoms of these two disorders are observed in combination in individuals with Waardenburg-Hirschsprung's disease or Waardenburg-Shah syndrome (WS4). WS4 in humans has been genetically linked to mutations in either the endothelin B receptor (1Hosoda K. Hammer R.E. Richardson J.A. Baynash A.G. Cheung J.C. Giaid A. Yanagisawa M. Cell. 1994; 79: 1267-1276Abstract Full Text PDF PubMed Scopus (905) Google Scholar, 2Attie T. Till M. Pelet A. Amiel J. Edery P. Boutrand L. Munnich A. Lyonnet S. Hum. Mol. Genet. 1995; 4: 2407-2409Crossref PubMed Scopus (190) Google Scholar), endothelin 3 (3Baynash A.G. Hosoda K. Giaid A. Richardson J.A. Emoto N. Hammer R.E. Yanagisawa M. Cell. 1994; 79: 1277-1285Abstract Full Text PDF PubMed Scopus (838) Google Scholar, 4Edery P. Attie T. Amiel J. Pelet A. Eng C. Hofstra R.M.W. Martelli H. Bidaud C. Munnich A. Lyonnet S. Nat. Genet. 1996; 12: 442-444Crossref PubMed Scopus (386) Google Scholar, 5Hofstra R.M.W. Osinga J. Tan-Sindhunata G. Wu Y. Kamsteeg E.-J. Stulp R.P. van Ravenswaaij-Arts C. Majoor-Krakauer D. Angrist M. Chakravarti A. Meijers C. Buys C.H.C.M. Nat. Genet. 1996; 12: 445-447Crossref PubMed Scopus (270) Google Scholar), or the HMG box transcription factor SOX10 (6Pingault V. Bondurand N. Kuhlbrodt K. Goerich D.E. Préhu M.-O. Puliti A. Herbarth B. Hermans-Borgmeyer I. Legius E. Matthijs G. Amiel J. Lyonnet S. Ceccherini I. Romeo G. Clayton-Smith J. Read A.P. Wegner M. Goossens M. Nat. Genet. 1998; 18: 171-173Crossref PubMed Scopus (706) Google Scholar, 7Southard-Smith E.M. Kos L. Pavan W. Nat. Genet. 1998; 18: 60-64Crossref PubMed Scopus (642) Google Scholar, 8Herbarth B. Pingault V. Bondurand N. Kuhlbrodt K. Hermans- Borgmeyer I. Puliti A. Lemort N. Goossens M. Wegner M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5161-5165Crossref PubMed Scopus (326) Google Scholar). In the Dom Hirschsprung mouse model for WS4,Sox10 Dom /Sox10 Dom homozygous animals are embryonic lethal, and the embryos fail to produce melanoblasts, whereas Sox10 Dom /+heterozygous animals have intestinal aganglionosis and spotted pigmentation (6Pingault V. Bondurand N. Kuhlbrodt K. Goerich D.E. Préhu M.-O. Puliti A. Herbarth B. Hermans-Borgmeyer I. Legius E. Matthijs G. Amiel J. Lyonnet S. Ceccherini I. Romeo G. Clayton-Smith J. Read A.P. Wegner M. Goossens M. Nat. Genet. 1998; 18: 171-173Crossref PubMed Scopus (706) Google Scholar, 7Southard-Smith E.M. Kos L. Pavan W. Nat. Genet. 1998; 18: 60-64Crossref PubMed Scopus (642) Google Scholar, 8Herbarth B. Pingault V. Bondurand N. Kuhlbrodt K. Hermans- Borgmeyer I. Puliti A. Lemort N. Goossens M. Wegner M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 5161-5165Crossref PubMed Scopus (326) Google Scholar). Taken together, the mutations of theSOX10 gene associated with WS4 in humans and the phenotype of the Dom Hirschsprung mouse indicate that one function ofSOX10 is to play a critical role early in the development of the melanocyte lineage. The identification of SOX10 target genes is crucial if the molecular basis for WS4 is to be fully understood, yet to date only one candidate target gene has been identified (9Liu Q. Melnikova I.N. Hu M. Gardner P.D. J. Neurosci. 1999; 19: 9747-9755Crossref PubMed Google Scholar). Although the genes likely to be regulated by Sox10 in the development of enteric neurons remain elusive, the genetics associated with the melanocyte system provide some clues for understanding the role of Sox10 in melanocyte development. The pigmentary defects observed in theDom mouse and WS4 are similar to those seen in mice or humans with mutations of the microphthalmia(Mitf) gene. Mitf encodes a basic helix-loop-helix-leucine zipper transcription factor essential for the development of the melanocyte lineage (10Hughes M.J. Lingrel J.B. Krakowsky J.M. Anderson K.P. J. Biol. Chem. 1993; 268: 20687-20690Abstract Full Text PDF PubMed Google Scholar, 11Hodgkinson C.A. Moore K.J. Nakayama A. Steingrimsson E. Copeland N.G. Jenkins N.A. Arnheiter H. Cell. 1993; 74: 395-404Abstract Full Text PDF PubMed Scopus (959) Google Scholar). Mutations in theMITF gene in humans are associated with Waardenberg syndrome type 2 (12Tassabehji M. Newton V.E. Read A.P. Nat. Genet. 1994; 8: 251-255Crossref PubMed Scopus (580) Google Scholar). The importance of Mitf in controlling the program of gene expression underlying the genesis of the melanocyte lineage is emphasized by the fact that ectopic Mitf expression in fibroblasts conferred on the recipient cells characteristics of melanocytes (13Tachibana M. Takeda K. Nobukuni Y. Urabe K. Long J.E. Meyers K.A. Aaronson S.A. Miki T. Nat. Genet. 1996; 14: 50-54Crossref PubMed Scopus (233) Google Scholar) and can induce transdifferentiation of neuroretina into melanocytes (14Planque N. Turque N. Opdecamp K. Bailly M. Martin P. Saule S. Cell Growth Differ. 1999; 10: 525-536PubMed Google Scholar). Similarly, misexpression of zebrafish Mitf results in ectopic melanized cells (15Lister J.A. Robertson C.P. Lepage T. Johnson S.L. Raible D.W. Development. 1999; 126: 3757-3767PubMed Google Scholar). Thus, it would appear that in some cells at least the expression of Mitf is sufficient to establish a program of melanocyte-specific gene expression. In the wild type mouse Sox10 transcripts are observed in the neural crest between 8.5 and 10.5 days post-coitum (16Kuhlbrodt K. Herbarth B. Sock E. Hermans-Borgmeyer I. Wegner M. J. Neurosci. 1998; 18: 237-250Crossref PubMed Google Scholar), whereasMitf transcripts are first detected in neural crest-derived melanoblast precursor cells at embryonic day 10.5 (17Opdecamp K. Nakayama A. Nguyen M.T. Hodgkinson C.A. Pavan W.J. Arnheiter H. Development. 1997; 124: 2377-2386Crossref PubMed Google Scholar). Taken together, the similar phenotypes of the Sox10 and Mitfmutant mice and the expression pattern of these two genes during development led us to hypothesize that Sox10 might regulate expression of the Mitf gene in the neural crest-derived melanocyte lineage. In this paper we show that the melanocyte-specific MITFpromoter contains functional binding sites for the Sox10 transcription factor. We demonstrate further that wild type Sox10 can activate transcription of the MITF promoter dependent on the ability of Sox10 to bind DNA. Significantly, a mutant form of the Sox10 protein genetically linked with inheritance of human WS4 acts as a dominant-negative repressor of transcription from the MITFpromoter and can reduce endogenous MITF expression. The ability of Sox10 to regulate the MITF promoter provides a molecular basis for the hypopigmentation and deafness associated with WS4. The human MITF promoter from −395 to +128 was isolated by PCR using appropriate primers that generated 5′ SstI and 3′ HinDIII sites and was cloned between the SstI and HinDIII sites of the pGL3 (Promega) firefly luciferase reporter plasmid. The Pax3-binding site within the MITF promoter was mutated from the wild type sequence ATTAAACTACTGGAACT to the mutant sequence AGCGATACTACTCGAGGT using a PCR-based strategy, generating the plasmid pMITF.Pax3mut. The Sox10-binding sites A, B, C + D, B + C/D, and F in the MITFpromoter were mutated using a PCR-based strategy, generating plasmids pMITF-Amut, pMITF-Bmut, pMITF-C/Dmut, pMITF-B+C/Dmut, and pMITF-Fmut, respectively (details of all sub-cloning is available on request). The sequence of the wild type Sox10-binding sites and the mutations introduced are described below. The pCMV-Sox10 and pGEX-KG.Sox10HMG expression plasmids were a gift from Dr Michael Wegner and have been described previously (16Kuhlbrodt K. Herbarth B. Sock E. Hermans-Borgmeyer I. Wegner M. J. Neurosci. 1998; 18: 237-250Crossref PubMed Google Scholar). The pCMV-Sox10 HMG domain mutant construct (pCMV-Sox10HMGdom.mt) was generated by replacing nucleotides 482–503 inclusive with an oligonucleotide of the sequence 5′-ATGGTCTTTTTTGTGCTGCATCCGGAG-3′. The MIC mutant of SOX10(6Pingault V. Bondurand N. Kuhlbrodt K. Goerich D.E. Préhu M.-O. Puliti A. Herbarth B. Hermans-Borgmeyer I. Legius E. Matthijs G. Amiel J. Lyonnet S. Ceccherini I. Romeo G. Clayton-Smith J. Read A.P. Wegner M. Goossens M. Nat. Genet. 1998; 18: 171-173Crossref PubMed Scopus (706) Google Scholar) was generated by PCR-mediated mutagenesis and introduced into pcDNA3.1 (Invitrogen) to produce the pcDNA3.1-Sox10.MIC construct. All mutant constructs were verified by sequencing. The Sox10in vitro transcription/translation vectors were made by subcloning the Sox10 HMG domain cDNA or the full-length Sox10 cDNA downstream from a β-globin leader sequence pT7plink in which Sox10 expression is controlled by the bacteriophage T7 promoter. The C terminus of Sox10 (amino acids 249–466) was sub-cloned into the plasmid pGEX2TKP to produce pGEX.Sox10C-term, which was used to express the GST-Sox10C terminus fusion protein. The transfection control plasmid consists of the G3PDH promoter used previously by us (18Carreira S. Dexter T.J. Yavuzer U. Easty D.J. Goding C.R. Mol. Cell. Biol. 1998; 18: 5099-5108Crossref PubMed Scopus (154) Google Scholar) cloned into the Renilla luciferase reporter (Promega). The GFP expression plasmid, pEGFP-C1, was obtained fromCLONTECH. 501 melanoma cells (501mel cells) were maintained in RPMI supplemented with 10% fetal calf serum and penicillin/streptomycin, and NIH-3T3 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and penicillin/streptomycin. Cells were seeded at a density of 1.5 × 103 cells/2 cm2 in a 24-well plate the day before transfection. 25 ng of promoter-reporter construct were transfected with increasing amounts of plasmid-expressing activators using Fugene (Roche Molecular Biochemicals) according to the manufacturer's instructions. An equal total amount of DNA was maintained by compensation with empty expression vector DNA. Cells were harvested 48 h after transfection, assayed for firefly luciferase activity, and the results normalized to the Renilla luciferase activity expressed from the pRL-G3PDH plasmid to control for transfection efficiency. Cells were seeded on coverslips at a density of 3.75 × 104cells/well in a 6-well plate the day before transfection. 100 ng of the pEGFP-C1 plasmid and increasing amounts of the pcDNA3.1-Sox10.MIC construct were transfected using Fugene according to the manufacturer's instructions, with the total amount of DNA being maintained using empty expression vector DNA. 48 h after transfection, cells were fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100. MITF protein was visualized using a rabbit anti-MITF primary antibody diluted 1:100 (gift from Simon Saule, Paris, France), followed by a Texas Red-conjugated anti-rabbit secondary antibody diluted 1:100 (Sigma). Cells were mounted using Vectashield medium and by fusion protein was expressed and from the and used to rabbit Cell were by in by and to The was in 10% Triton for 1 h at and at 4 with Sox10 diluted 1 in in The was detected using a anti-rabbit antibody diluted 1 in in and using Sox10 and Sox10 HMG domain were by in using the (Promega). domain protein was expressed and from the and the protein was by with used are as with the mutated M. M. van K. H. J. 1993; 12: PubMed Scopus Google site A, site A, site B, site B, site C + D, site C + D, site site the box has been described previously P. Yavuzer U. Goding C.R. Mol. Cell. Biol. 12: PubMed Scopus Google Sox10 site B + and were with as described previously I. Goding C.R. Mol. Cell. Biol. 1998; 18: PubMed Scopus Google Scholar). were on for in a 2 4 of serum 1 of ng of and were in the reaction for on before of the and binding was to for a further on were on a at 4 and were and visualized by the importance of Mitf in the genesis of the melanocyte lineage and the of Sox10 and during mouse we to the that the hypopigmentation and deafness associated with mutations in the SOX10 gene an ability of SOX10 to regulate the expression of a first we assayed the ability of a Sox10 protein to activate transcription of a luciferase reporter by the melanocyte-specific promoter N. H. K. S. 1996; PubMed Scopus Google Scholar) from −395 to +128 using a transfection in 501 human melanoma cells that express endogenous MITF T.J. C. T. D.E. 1998; PubMed Scopus Google Scholar). The results demonstrate that transfection of a Sox10 expression vector in to of theMITF promoter. was dependent on Sox10 binding to DNA the HMG domain to DNA binding the ability of Sox10 to transcription SOX10 were to target the MITF promoter, mutations linked to WS4 would be to expression. The SOX10 MIC which has been genetically linked with due to a to at which results in the expression of a protein which is after the HMG domain (6Pingault V. Bondurand N. Kuhlbrodt K. Goerich D.E. Préhu M.-O. Puliti A. Herbarth B. Hermans-Borgmeyer I. Legius E. Matthijs G. Amiel J. Lyonnet S. Ceccherini I. Romeo G. Clayton-Smith J. Read A.P. Wegner M. Goossens M. Nat. Genet. 1998; 18: 171-173Crossref PubMed Scopus (706) Google Scholar). mutant the transcription the ability to bind DNA K. C. Sock E. Pingault V. Bondurand N. Goossens M. Wegner M. J. Biol. Chem. 1998; Full Text Full Text PDF PubMed Scopus Google Scholar), and would be to in a dominant-negative by with wild type SOX10 for binding to the MITF promoter. with expression of a Sox10 MIC mutant protein in the 501 melanoma in a in expression from reporter in to the Sox10 protein which activates the MITF promoter the Sox10 MIC mutant that is expressed from the promoter at similar 2 B and K. C. Sock E. Pingault V. Bondurand N. Goossens M. Wegner M. J. Biol. Chem. 1998; Full Text Full Text PDF PubMed Scopus Google Scholar) acts as an that the Sox10 MIC protein reduce endogenous MITF 501 melanoma cells were transfected with either an empty expression plasmid or the Sox10 MIC expression and endogenous MITF were by using an anti-MITF Cells transfected with the Sox10 MIC expression vector were identified using a GFP expression All cells of endogenous MITF protein with either cells 2 or control cells GFP 2 with Sox10 MIC endogenous MITF expression. The likely to for the of theMITF promoter by Sox10 MIC would be that expression of the mutant protein a transcription domain would with endogenous SOX10 for of the MITF promoter. was to that the cells express endogenous results using that the whereas cells SOX10 protein was we a antibody that SOX10 the transcription factor and we used this antibody in a with from 501 melanoma The a with a molecular to SOX10 in cells in cells which are with the results obtained using the MITF promoter these provide that one of the SOX10 MIC associated with WS4 would be expression. the critical for MITF in melanocyte the in MITF expression as a of mutations in the SOX10 gene would be sufficient to for the hypopigmentation and deafness associated with WS4 and the Dom Although we have that Sox10 a that the for SOX10 was of theMITF promoter and protein binding sites M. M. van K. H. J. 1993; 12: PubMed Scopus Google Scholar). of these sites were by from the MITF promoter were in a with in vitro The results that Sox10 binds to of the 3 binding sites within these were identified using of the promoter and 3 sites C and to an that have been as a element in to of sites and the binding site sites M. M. van K. H. J. 1993; 12: PubMed Scopus Google Scholar) were in the The results 3 indicate that Sox10 binds to sites B and C/D, to site and to sites and E. to sites B and was with being by mutations into the Sox10 and F and between and either B or C/D, with the results in C. was observed using an the box P. Yavuzer U. Goding C.R. Mol. Cell. Biol. 12: PubMed Scopus Google Scholar), as a the of these binding sites for of theMITF promoter by Sox10 in we introduced mutations to those used in the DNA binding into Sox10-binding site B, C/D, and and into the two binding sites B + in We the of these mutations on of the MITF promoter activity in a transfection The results demonstrate that of site on the ability of Sox10 to activate expression from the MITF promoter, whereas of sites B in combination the ability of Sox10 to Thus, whereas Sox10 activate the and site mutant between and only a was observed with the site B + The to the ability of Sox10 to regulate theMITF promoter by of either of the Sox10-binding sites B or with the observed of sites B + in to us that these sites to expression. with the Sox10 MIC mutant to transcription from an MITF promoter in which the B and sites are mutated have that promoter-reporter constructs can be by Sox10 in with or in a in cells (16Kuhlbrodt K. Herbarth B. Sock E. Hermans-Borgmeyer I. Wegner M. J. Neurosci. 1998; 18: 237-250Crossref PubMed Google Scholar). the MITF promoter we have identified a Sox10-binding site that is to the Pax3-binding site 3 We it was to Sox10 for the of theMITF promoter and Sox10 and can activate MITF expression. We first a using a the Sox10 site B and the Pax3-binding sites from the MITF promoter. The results indicate that Sox10 and can bind this we were to DNA binding the is with that Sox10 and in vitro in a protein can with that Sox10 and directly demonstrate that binding of one protein binding of the protein to this of the MITF promoter, we a similar using an of expressed that the would these a between the the domain and the protein was observed Sox10 and can bind to sites within this of the MITF promoter, we have been to for with we have observed of the MITF promoter by Sox10 and in transfection in Sox10 or can activate transcription that the ability of Sox10 to activate theMITF promoter is of we used promoter in which the Pax3-binding site has been has been previously to from binding to and the MITF promoter A. Takeda K. B. M. Nat. Genet. 1998; 18: PubMed Scopus Google Scholar). the of MITF promoter activity was by of the Pax3-binding site this on the ability of Sox10 to activate the promoter, that Sox10 of MITF expression can of provide of that the of theSOX10 gene in WS4 the the MITF promoter is by Sox10 in the Sox10 MIC mutant protein that the domain acts to expression from a transfected MITF promoter luciferase reporter and expression from the gene. the MITF promoter contains functional Sox10-binding results further that the and deafness associated with the SOX10 MIC arise as a of SOX10 MIC with SOX10 for binding to the MITF promoter. Taken with the SOX10 in the development of the melanocyte provide a into the of Mitf expression in development be Mitf a role in melanocyte development. In the of a functional Mitf mice a of neural crest-derived melanocytes (10Hughes M.J. Lingrel J.B. Krakowsky J.M. Anderson K.P. J. Biol. Chem. 1993; 268: 20687-20690Abstract Full Text PDF PubMed Google Scholar, 11Hodgkinson C.A. Moore K.J. Nakayama A. Steingrimsson E. Copeland N.G. Jenkins N.A. Arnheiter H. Cell. 1993; 74: 395-404Abstract Full Text PDF PubMed Scopus (959) Google Scholar), as as a of the to K.J. Genet. 1995; Full Text PDF PubMed Scopus Google Scholar). Mutations in theMITF gene in humans to (12Tassabehji M. Newton V.E. Read A.P. Nat. Genet. 1994; 8: 251-255Crossref PubMed Scopus (580) Google Scholar), and in the zebrafish a in an Mitf gene, results in of (15Lister J.A. Robertson C.P. Lepage T. Johnson S.L. Raible D.W. Development. 1999; 126: 3757-3767PubMed Google Scholar). the importance of Mitf for melanocyte a is the expression of Mitf is in the neural crest. in to the for Sox10 transcription have been in Mitf expression. are as the transcription factor A. Takeda K. B. M. Nat. Genet. 1998; 18: PubMed Scopus Google Scholar), with mutations in the gene to and M. Read A.P. Newton V.E. M. P. T. Nat. Genet. 1993; PubMed Scopus Google Scholar, M. Read A.P. Newton V.E. P. T. PubMed Scopus Google Scholar, T. W. A. Hum. 1994; PubMed Scopus Google transcription that have been in the zebrafish and in cells to regulation of the Mitf promoter in response to Raible D.W. 14: Google Scholar, K. S. T. H. K. S. J. Biol. Chem. Full Text Full Text PDF PubMed Scopus (270) Google and as protein that regulate Mitf expression in response to C. K. J. Cell Biol. 1996; PubMed Scopus Google Scholar, C. P. K. E. Mol. Cell. Biol. 1998; 18: PubMed Scopus Google Scholar). the and to target the are all expressed in the neural crest during is Mitf expression to a of cells for the melanocyte is that the of Mitf expression is a an of transcription to the at promoter. in those cells a combination of is and be The that can play a critical role in or has been previously D. Cell. Full Text PDF PubMed Scopus Google Scholar) and is likely to be to the described by Cell. Full Text PDF PubMed Scopus Google Scholar) in the of melanoma using Although the of Mitf expression would be dependent on those as Sox10 and that are genetically in melanocyte we can that these at this promoter ability to in an reporter in cells (16Kuhlbrodt K. Herbarth B. Sock E. Hermans-Borgmeyer I. Wegner M. J. Neurosci. 1998; 18: 237-250Crossref PubMed Google Scholar). is to that these at in that in the we have we can for an ability of protein to activate the promoter. Although we have been to demonstrate in melanocytes between and it is that between and for by that is to with J. M. L. D. W. 1996; PubMed Scopus Google Scholar, J. M. 1996; PubMed Scopus Google Scholar, M. van M. M. J. S. V. J. H. Cell. 1996; Full Text Full Text PDF PubMed Scopus Google Scholar) and with some of the P. Mol. Cell. 1999; 4: Full Text Full Text PDF PubMed Scopus Google Scholar). In Mitf expression is is to that would be a for the in and it is that the role of Sox10 might be by of the in development. the activity of factor binding to the be dependent on regulation and with or the regulation of by and the of the by one of the of to the Mitf promoter be to the from to expression as from the neural crest to in the and Sox10 is regulated and the DNA from Sox10 DNA binding to the regulation and of promoter are that to be if the on Mitf expression are to be We Michael Wegner for the Sox10 expression vectors and antibody and Simon Saule for the