Mutations in the Ligand-Binding Domain of the Kit Receptor: An Uncommon Site in Human Piebaldism

Roger A. Fleischman, Teresa Gallardo, * and Xiafang Mi Division of Hematology/Oncology, University of Kentucky and VA Medical Center. Lexington. Kentucky; and the ' University of Texas Southwes tern Medical Center, Dallas, Texas, U.S.A.

Heterozygous mutations in the gene for the Kit transmembrane receptor have been identified re­cently in human piebaldism and mouse "dominant spotting." Interestingly, all of the 14 known missense lDutations that cause depigmentation in these species Dlap to the tyrosine kinase domain of the receptor, whereas none have involved the extracellular ligand­binding domain. In an attempt to detect these un­COJn.mon mutations, we screened the nine exons encoding the extracellular portion of Kit for single­strand conformation polymorphisms (SSCP) in eight piebald subjects previously reported to be negative for kinase mutations. Four of these eight kindreds proved to carry novel mutations. The first mutation, found in two apparently unrelated pro bands with lDild piebaldism and English ancestry, substitutes an arginine for a highly conserved cysteine at codon 136. This substitution disrupts a putative disulfide bond required for formation of the second Ig-like (D2) loop

H U111an piebaldism is an autosomal dominant disor­der in melanocy te development characterized b y a white hair forelock, areas of hypo pigme ntation on the antel"ior trunk and extremities, and the abse nce of other manifestations (Ortonne, 1988). Based on

the similarity of the human phenotype to mouse " dominant spot­ting," which results from mutations in the c-kit gene (Chabot et ai, 1988; Geissler et ai, 1988), we and others recently identified heterozygous mutations of the human c-kit gene in kindreds with piebaldism (Fle ischman e/ ai, 1991; Giebel and Spritz, 1991) .

In contrast to the findings for other transmembrane tyrosine kinases such as the in sulin receptor (Flier, 1992), aJl of the 14 independently arising missense mutations that result in depigmen­tatiOl)' have mapped to the intracellular domains that catalyze tyrosine phosphorylation (Nocka c/ ai, 1990; R eith e( ai, 1990; Tan et af, 1990; Giebel and Spritz, 1991; Fleischman, 1992; Larue e( ai, 1992; Spritz et ai, 1992a, 1992b, 1993; Tsujimura e( ai, 1993; Ezoe e/ ai, 1995). The relative rarity of mutations in tlle extracellular domains suggests that the d efects resulting £i'om amino acid substi­tutions in the ligand-binding region are generally too mild to cause depigmentation . On the other hand, the iden tifi cation ofmutatiol1s

Manuscript received May 24, 1996; revi sed July 10, 1996; accepted for publication July 22, 1996.

R.eprint requests to: Dr. Roger A. Fleischman, University of Kentucky and VA Medical Center. 800 Rose Street, Lexington, KY 40536-0084.

Abbreviation: SSCP, single-strand conformation polymorphism.

of the Kit ligand-binding domain. The second muta­tion, detected in a piebald kindred characterized by unusually limited depigmentation, substitutes a threonine for an alanine at codon 178, a site just proximal to conserved cysteines at co dons 183 and 186. The third mutation, occnrring in a kindred with more extensive depigmentation, is a novel four-base insertion in exon 2 that results in a proximal frame­shift and premature termination. The data strongly suggest that piebaldism can result from missense mutations in the Kit ligand-binding domain, al­though the resulting phenotype may be milder than that observed for null or kinase mutations. The ap­parent clustering of these uncommon mutations at or near the conserved cysteines for the D2 Ig-like loop further suggests a critical role for this region in Kit receptor function. Key words: pigmetltation disOl'ders/ melatlocytes/tyl'osine kitlase. ] bwest DeJ?l/atol 107:703-706, 1996

in the extracellular portion of tlle receptor may reveal moieties that are particularly important for ligand binding or other aspects of receptor function.

In this regard. we now report three novel mutations in four subjects with pie baldism, including two missense mutations that map to Kit exon 3, a sequence that encodes the second of five Ig-like loops that form the extracellular portion of the receptor.

MATER.IALS AND METHODS

Subjects The piebald subjects, from apparently unrelated kindreds , have been described previously (Fleischman ct aI, 1991; Fleischman, 1992). Probands 1 and 2 were previously designated as subjects E and F (Fleisch­man ct ai , 1991) . These subjects are apparently unrelated Caucasian males of English ancestry with a f.1mily history of multiple generations of piebald individuals. Both kindreds display mild piebaldism with small areas of spotting on the ventral abdomen. with or without a small whitc hair forelock. For example, proband 2 exhibits only a small arca ofleukoderma in the midabdomen bm no white hair forelock, altllOugh the mother also has a typical small white fore lock. Proband 3 was previously designated as subject I (Flcisclll1lan, 1992). T his proband and one similarly affected sibling have unusually mild depigmentation, consisting only of a small wh.ite hair forelock at birth that gradually became unapparent after the first few years oflife. Proband 4 was previously designated as subject G (Fleischman el ai, 1991) and has a prominent white forelock and extensive areas of leuko­derma 011 the abdomen. arms, and legs.

Single-Stranded Conformation Polymorph isms (SSCP) For tl,e de­tection ofSSCP, we perfon1l cd polymerase chain rcaction (PCR) Witll 1 f.l.g of genomic DNA, 50 f.l.M each dNTP, 0.4 f.l.M each oligonucleotide primer.

0022-202X/96/S10 .50 • Copyright © 1996 by The Society for Investigative Dcnnatology. Inc.

703

704 FLEISCHMAN ET AL

10 mM Tris HCI (pH 8.3), 50 mM KCI , 1.5 mM MgCI2 , 0 .25 U Them.lIs aqll a'i",s DNA polymerase (Perkin-Elmer/Cetus, Norwalk, CT), and to ILCi of[a-3 2P]dCTP (3000 Ci per mmol, 10 mCi per ml) in a final volume of 25 ILl. The sequences of the primer pairs for PCR were: exon 2, 5' -TACAGGATCCTACTCAACACGATTCTGTTT-3 ' /5' -CACTGAAT­TCCTGCAGAAAGCCAAGCATTT-3'; and exon 3, 5'-TACTGATTTG­GATATGCTTC-3'/5'-CTCCAGAGGCAGATAAGAAA-3'.

PCR was performed for 40 cycles at a denaturation temperature of 94°C for 1 min, primer annealing at 55°C for 1 min, and extension at 72°C for 2 min. The PCR product (1.5 ILl) was added to 20 ILl offormamide dye (98% formamide120 mM ethylenediamine tetraacetic acid/0.05% bromophenol blue) , and immediately before electrophoresis, the samples were heated in a boiling water bath for 6 min. Aliquots (2 ILl) of each denatured sample were loaded onto a 38 cm X 20 cm X 0.4 mm 5.5% polyacrylamide gel (19:1 ratio of acrylamide to bis-acrylamide) with 10% glycerol and 2 X TBE buffer (50 mM Tris, pH 8.0/50 mM boric acid/l mM ethylenediamine tetraacetic acid). Electrophoresis was performed at 300 V at room temper­ature for 16-24 h. After electrophoresis, the gels were transferred to Whatrnan 3M paper (Maidstone, U .K.) and dried on a vacuum slab gel dryer for 2 h at 80°C. Autoradiography with Kodak X-Omat film (Roch­ester, NY) at room temperature without intensifying screens allowed detection of PCR products in 4-12 h of exposure.

DNA Cloning and Sequencing PCR was carried out as described previously, with the modification that the total volume was 100 ILl and the concentration of dNTPs was 200 ILM, with no radioactive label added. To confirm the absence of contamination in negative control samples and to ensure that the PCR product was a single band of the appropriate size, we visualized 1 ILl of the PCR. product after electrophoresis in a 3% NuSieve (FMC Bioprodllcts, Rockland, ME)-I% agarase minigel. To prepare the exon-3 DNA for ligation , 1 ILl of T4 DNA polymerase was added to the PCR. reaction mixtures and incubated for an additional 20 min at 37°C . The reaction product was then purified by electrophores.is overnight at 300 V on 10% polyacrylamide gels, visualized by ethidium bromide staining, excised, and cloned into M13 mp18 by ligation at the BamHl and EcoRI sites with T4 DNA ligase. M13 single-stranded DNA was sequenced with modified T7 DNA polymerase (Sequenase, Version 2.0; United States Biochemical, C leveland, OH) using [a-35S1dATP (1000 Ci/mmol) according to the protocol supplied by the manufacturer. Exon-2 fragments were cloned into the PCR II vector by TA cloning according to the manufacturer's protocol (Invitrogen, San Diego, CAl and were sequ'enced by automated bidirec­tional· dye terminator sequencing.

Restriction Digest and Analysis Kit exon-3 PCR fragments were independently reamplified from the piebald subjects with mutations in exon 3 and from a norma'! control, cleaved by repetitive incubation with either NspBIl for the mutation at codon 136 or HiraI for the mutation at codon 1.78, and analyzed by a 3% NuSieve-1 % agarose minigel. Kit exon-2 PCR. fragments were independently rcamplified from piebald subject 4 and a normal control, cleaved by incubation with Mspl, and analyzed on a 5% NuSieve agarose minigel.

RESULTS

Probands 1 and 2 Are Heterozygous for a Missense Substi­tution at Codon 136 Genomic PCR was u sed to amplify c-kit exons from nine piebald probands, one of whom has been reported previously to have a d e letion (Fleischman et ai, 1991). SSCP analysis of the amplified DNA fragments corresponding to exons 1-9 demonstrated a variant SSCP pattern in exon 3 for probands 1 and 2 (Fig 1. laues 2,3). To determine the basis for the altered SSCP, we independently reamplified the PCR products and cloned them into M13. Sequencing of the clones corresponding to lanes 2 and 3 demonstrated a h eterozygous transition in codon 136 (TGT ~ CGT), which sub stitutes arginin e for cysteine (Fig 2). Because this substitution e liminates a unique NspBII restriction site in the DNA sequence of exon 3, the presence of the heterozygous mutation was confirmed by digestion of independently amplifIed PCR products from both probands and normal controls with NspBII (not shown) .

Proband 3 Is Heterozygous for a Missense Substitution at Codon 178 SSCP analysis of the amplified DNA fragments corresponding to exon 3 also demonstrated a subtle variant SSCP pattern for proband 3 (Fig 1. lalle 4). To determine the basis for this variant, w e independently reamplified the PCR product and cloned it into M13 mp18. Sequencing of the clone (Fig 1. lalle 4)

THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

1 2 3 4 5 6 7 8 9

Figure 1. SSCP of c-kit exon 3 amplified from the DNA of piebald subjects. DNA was amplified by PCR., denatured, and subjected to polyacrylamide gel electrophoresis under neutral conditions. No variants were detected in lalle 1, a normal contro l, and lalles 5-9. Lat.es 2,3 (probands 1 and 2) show a variant pattern with two additional bands indicated in lalle 2 (an·olll). Lalle 4 (proband 3) also shows a distinct variant pattern with an additional faint band at the top of the gel and a broader band in the middle of the gel. The major band at the bottom in all samples corresponds to the native double-stranded PCR product.

demonstrated a heterozygous transltton in codon 178 (GCC ~ ACC), which substitutes a threonine for alanine (Fig 3). Because this mutation efuninates a unique HILaI restriction site in exon 3, the presence of the h eterozygous mutation was confirmed by digestion of an independently amplified PCR product from proband 3 with HhaI. In contrast, the eight other piebald subjects and 12 unrelated normal controls exhibited the normal digestion pattern (not shown) .

A. Normal B. Mutant

GATe --- -~ - --:,. -- -Am=: = ---- -:::---

GATe - - -- -- -::::::;:: - .... --G'Z-E:

Val Arg Arg Pro Leu

Mutant 5'-GTC CGC CGT CCT CTC 3'-CAG GCG GCA GGA GAG

t Normal 5'-GTC CGC IGT CCT CTC

3'-CAG GCG ACA GGA GAG

Val Arg Cys Pro Leu

Figure 2. Partial nucleotide sequence analysis of c-kit exon-3 cDNA from proband 2. Exon 3 from proband 2 (Fig 1. lalle 3) was amplified by PCR. and cloned into M13. The nucleotide sequences were determined for eight clones obtained ITom two independent PCR. reactions. Shown are portions of the sequencing ladder spanning the mutation at nucleotide 427 for one nom1al allele (A) and one mutant allele (B). The normal allele had the same sequence as that published for human c-kit (Yarden e/ ai, 1987), and was confirmed in two other clones. The mutant allele had the same sequence as described except at codon 136, which was CGT instead of TGT, resulting in the substitution of arginine for cysteine. The sequence of tlus clone was confirmed in four other clones from twO independent PCR reactions.

VOL. 107, NO. NOVEMBER 19%

A. Normal B. Mutant

GATC GATC

-- - -; =r --=.....:C --- --

lys Arg Thr Tyr His

Mutant 5'-AAA CGC ACC TAC CAT 3'-TTT GCG IGG ATG GTA

t Nonnal 5'-AAA CGC ,GCC TAC CAT

3'-TTT GCG ,CGG ATG GTA

lys Arg Ala Tyr His

Figure 3. Partial nucle otide sequence analysis of c-kit exon-3 cDNA from proband 3. Exon 3 from proband 3 (Fig 1, laue 4) was amplified by PCR and cloncd into M13. Shown are portions of the sequencing ladder spanni ng the mutation at nucleotide 553 for one normal alle le (A) and onc mutant allele (B). l30th sequcnccs arc the same as that published for human c-ki( (Yarden et nl, 1987) cxcept at codon "178, where ACC iJ) the mutant aUcle results in thc substitu tion of threonine for ala n ine. The sequence was con fi rmed in multiple other clones from two indepen­dent PCR rcactions.

Proband 4 Is Heterozygous for an Insertion in Exon 2 SSCP analysis of the amplified DNA fragments corresponding to exon 2 demonstrated a variant SSCP pattern fo r proband 4 (not shown). To determine the basis fo r this variant, we independen tly reamplified and cloned the PCR product. Automated sequencing of 11 clones demonstrated five clones with a four-base insertion and six clones with the normal seq uence (Fig 4). The inserted se-

I quence, TCCA, occurs at three tandemly repeated TCCA bases (nucleotides 133-144) in Kit exon 2 . T his frameshift mutation results in a 14-mnino acid nonsense peptide term inating at a novel in-frame TAG at codons 54 - 55. Beca use this insertion increases the size of a Mspl res triction digest fi'agment £i'om 30 to 34 bases, the presence of the heterozygous mutation was confirmed after diges­tion of an independently ampLi flCd PCR product from proband 4 and a normal contro l by electrophoresis on a 5% NuSieve agarose minigel (not shown) .

DISCUSSION

This article repons missense mutations in the extracell ular ligand­binding domain of the Kit receptor that are associated with a piebald phenotype. T he loss of tbe highly conserved cysteine at codon 136 is almost certainly the cause of depigmentation in probands 1 and 2. Indeed, this cysteine is tho ught to participate in a critical disulfide bond required fo r formation of the D2 loop (Hanks et ai, 1988), a domain recently shown to be essential for ligand binding (Blechman et (/1, 1993a, 1993b; Lev e/ ai , 1993). Th us, the loss of the D2 loop appears li kely to impair ligand binding. Although these studies cannot rule out the alternate possibility that abnormal fo lding of the mutant protein prevents insertion into the cell m embrane, rep lacement of the homologous cysteines in the 19-Like loops of a rela ted tyros ine kinase, fi broblast growth factor receptor-1 , abolished ligand binding bu t did not affec t membrane insertion (Hou e/ (/1, 1992; Z immer el (/1, 1993).

The pathologic significance of the codon-178 mutation is less certain. As expected for a region involved in the specificity of ligand binding, this res idue is not conserved in other members of the Kit fam il y, such as the receptors for macrophage colony-

MUTATIONS IN THE KIT LIGAND-BINDING DOMAIN 705

A c CAT C ,cao A T C CAT C C A 11, C C A G G A

B C &C CA T C CAT C 1<;;0 ATe C A G G A

Figurc 4. Partial nucleotide sequcncc analysis of c-kit cxon-2 cDNA from proband 4 . Sections of the sequencing chromatograms obtained by bidirectiona l dye terminator sequencing of one normal all ele (A) and one mutant all ele (B). Exon 2 was ampli fied by PCR and cloned by TA cloning into t he PCR II vector (Invitrogen). l30th sequences arc the same as th<lt published for human c-ki( (Yarden e( ai, 1987), except for a four-base TCCA insertion occurring at three tandemly rcpeated TCCA bases (nucleotides 133-144). T he mutant sequence was confi rmed in mul tiple other clones fro m two independent PCR reactions.

stimulating fuctor and platelet-del"ived groWtll factor. On tlle other hand , the alanine at codon 178 is conserved in huma.n , mouse, and rat Kit. T he absence of this mu tation in 20 unrelated normal and piebald subjects or its detection in more than 50 other reported human samples screened by SSCP or complete DNA sequencing (Yarden e( (/1, 1987; Ezoe e/ (/1, 1995; Nakata et (/1, 1995) suggests that this is not a common po lym orphism. Presumably, a substitu tion close to the conserved cysteines at codons 183 and 186 could affect formation of either the D2 disulfide bond or a second potential disulfide bond between the cysteines at codons 183 and 151. T h e defect resulting fro m the substitution at codon 178 appears to be associated with a more subtl e defect than that resulting from loss of the cysteine at codon 136, however, as suggested by the unusually mild depigmentation observed in this proband .

T he proximal fra meshift mutation in exon 2 adds a fourth TCCA tetranucleotide to the normal sequence of three tandemly repellted un.its, consistent with a slipped-strand mispairing mechanism (Coo­per and Krawczak, 1991) . T he frameshift is predicted to result in the biosynthesis of a truncated and nonfunctional Kit polypeptide. Although it has been suggested, based on a previously reported frames hift mutation in exon 2, that haploinsufficiency is associated with a very m.i1d piebald phenotype (Spritz e/ ai, 1992b), proband 4 displays extensive leukoderma with large areas of depigmentation on the llbdomen , anns, and legs. Presumably, differences ill genetic background may account in part for the phenotypic variabiljry observed in different human kindreds with Kit haploinsufficiency, similar to previous observations on the effect of Kit mutant alleles in different mouse inbred strains (Silvers, 1979) .

Even after including the two novel missense mutations reported here, 14 of the 16 reported mouse and human Kit missense mutations that resul t in depigmentation map to tll e in trace llular

706 FLEISCHMAN cr AL

domains that encode kinase actlvlty. The low trequency of muta­tions in the c-kit extracellular domain implies that most am ino acid substitutions in this region cause too mild a defect to result in depigmentation. One explan ation for this finding is suggested by recent experiments indicating that monovalent binding of li gand is sufficient to induce Kit receptor dimerizatioll . On the other hand, both chains of the resulting dimer must have kinase activity to catalyze subsequent autopbosphorylation (Lev ct ai, 1992). Th us, up to 75'Y., of the potential dimers aflected by kinase domain mutations, i.e., homod imers of the mutant and heterodimers of the mutant and normal gene products, would be defective, resulting in a "dominant negative" effect on receptor function (Tan et ai, 1990;

Fleischman, 1992). In contrast, only 25% of the potential dimers affected by a ligand-binding mutation, i.e., homodimers of the mutant gen e product, would be impaired for ligand-mediated climerization, presumably too mild a defect to affect pigmentation.

If the c ited model is correct, disruptio11 of the D2 Ig- like loop by mutations at or near the critical cysteine residues may be unu sual in its ability to exert a dominant n egative action o n receptor het­erodimers . [n accord with this suggestion, the mutations that cause the Crouzon, Jackson-Weiss, and Pfeiffer fo rms of cran iosynostosis Oabs et ai , 1994; Reardon et ai, 1994; Oldridge ef ai, 1995; Rutland et ai, 1995; Schell ct ai, 1995; Steinberger ef ai, 1995) are similarly clustered at or near the conserved cystein es req uired for formation of one Ig-like .loop in the fibroblast growth factor receptors 1 and 2. On the other hand, at least for Kit, the mild piebaldi sm observed in the three kindreds reported here with missense mutations in exon 3 suggests that the receptor defect associated with loss of the D2 loop is still less severe than that resulting from some null or deletion mutation s and from the majority of kinase mutations.

T his st"dy "'liS SHpp"rted i" plln I!y II 1/ A Mait Award fIIlfl A"/"rica,, CIIIIC''''

Society Grn/If I N-142.

REFEREN CES

I3lcchl~l:t11 JM , Lev S, Brizzi MF. Leitner 0, Pcgonlro L, G ivol 0, Y~l'clcn Y: So luh le c-kit proteins ~nd nntireceptor monoclonal anribod.ies confi ne the binding sit:c o f the Stelll cell r.,etor. J Bioi C/,w, 268:4399-4406. 199301

D1echrn :1I1 JM , Lev S, Givul 0, Varden Y: Structure-function analyses of till: kit receptor ro r the steel f:l cror. SIC'" Cells 1·1 (SliPI' I 2): 12-2 1, 1993 b

C hahot B. Ste phenson OA. C hnpman VM, Bcsl11 er P, IJc l'Tlstcin A: T he proto­oncogene c-kil encoding :l transTll clllhrOin c tyrosine ki nase rccepwr IlInps [Q the rnOllse HI loclIs. Nlltllre 335:88-89. 1988

Cooper ON. Krawcz;lk M: Mcchanisms of insertional ll1utOlgcJl csis ill human genes causing: genetic diseilse. HUIII Gmt" 87:409-4 15. J 1)91

Ezoe K. Holtnes SA . Ho L. Bennett C P, BolognioJL, Brueton L. BtI"'J, Fala belb 11.. , Gotto EM , Ishii N. Moss C. PittcIkow MR, T hompson E. \Vard KA, Spritz ItA: Novel II1tJt;ltiOIlS and dclel"ions of the KIT (steel ('ac ror receptor) gem: in ImllHlIl piebaldism. A",] 1-111111 et'lIct 56:58-66, '191)5

Fle ischman n. . ..A: Hlll1'l~11 pieb~ ld trait resuiring [r0I11:1 dominant negative mutant ~1Ic1e of tllc c-kil TlICllIbra ne receptor gene.] Cfillfll l/csI 89:17 J3-1T I 7. 1992

Fleischmil ll ItA. Salt1l1all DL, Stastny V. Z ll cirner S: Deletio n of the c-kit protu­oncogene ill the Illlman develo pment;}i defect piebald tT;}il. Pmr N fIIl Acari 'ci

USA 88: 10885-10889. 199 1 Flier J S: Lill y lecturc: syndromes of insulin resistance. Di(/fJ(~(es 4 1: 1207- 12 19,1992 Geissler EN. flYilll MA, Housman DE: The dominilnt-white spottin g: (11'11) locus of rile

mo usc encodes the c-/~iI protO-on cogelle. Cell 55: 1 85- '192. '198H Giebc l LU, Spritz flA: Mur:uion of ('he K IT' (masc/s rclIl cell growth factor receptor)

pro[Qoncogclle in human piebaldism . fJrt)C NII!I Arml Sci USII H8:8696-8699. 199 1

THE JOURNAL OF INV EST IGAT IVE DErtMATOLOGY

Han ks SK, Quinn AM. Huntcr T: Th e proteill killase f~mi1 y: co nserved feature s and dedu ccd phylogeny of Ihe c.tt~l y ti c dom;lins. Scimrr 2·11 :-12-52. 1. 988

HOtl J. I(an M . Wan g- 1=, X U J. Naka h:1ra M. Mcllridc G. McKechan K, McKeehan WL: Substitution of putative half-cysti.l1c residucs in heparin-binding fibrobla t growth factor receptors. Loss of bindin g a.;l"iviry in hath two and threc loop isoionlls.) Bi,,1 C ite", 267: 17804-17808. L992

Jabs EW . Li X. Scott AF. Meyers G , Chell \V, Eccles M, MaoJ- I, Chamas Lit. Jackson CE. J ;\ye M:Jacksoll-Wciss ;lI1d C rouzon syndro mes ilrc allelic w ith l11utalio ns in fibroblast growth ractor receptor 2. Nail/fl.: Gt'l/et 8:275-279, 1994

Larue L. Dougherty N, Porter S. Mintz B: Sponta11cOlis malign:ll1t transfonni.ltion of m clan ocytes exp,lJldcd from I llf/H;f mice witl l :.1 Kil kil lase-c1ol1lain Inutation. I'mc NlI /1 I lrnd Sri USA 89:7H 16-7820, 1992

Lev S. lJIC Chlll<l1l J. Nishikawa 5-1, Givol D. YOIrden Y: In terspecics molecular chirllcras of Kit help defi ne thc binding site ot' the stc m cell f;' ctor. tHof Cell Bioi 13:2224-2234. 1993

Lev S. Vnrde l1 y , Givo l I): D imeriz:ttio n and activntio ll of the l{j t receptor by llIonOVnlell1: ;lIld biv;dcnt hindillg of til e srclIl cc ll f:lcwr.) Hiol Cil/'111 267: 15970-

15977. 1992 Noka ta V. IGmun> A. I<aroh O. Kawoishi K. H)'odo H. Abc K, Kuramoto A, SatowY:

c-kir point 1l1utation of extr;lcel lul"r dOlllain in padclli s with lI1yeloprolifcr~ltive disorder<. Br) /-IlIC>//(/(ol 9 1 :6(tI -663, '1995

Nocka 1<. To n J C, C hiu E. Chu TV, Ray P. Traktl1lan 1'. Besme r P: Molecular basis of duminnnr lIegative <1no loss of function llIutarion s at I hc murine c-kitlwhite spotting locus: W ''' , W ", W ·tt • and HI. liMBO ) 9: 1805-1 813. 1990

O lcIridgc M, Wilkie AO. Slaney SF. Poole MD . Pulle),11 LJ . R utland P. Hockle)' AD. Wake MJ. Goldin JH . W imer ItM: Mllt:I Cio lls ill the third illllllullogiobu.lin domain of the fIbroblast grow th f:lcror rcccptor-2 genc ill C ro uzo l1 syndrome. /-I"", At .. 1 Gm e( 4: I 077- 1 082. 1995

O rtonne J-P: PiehaldisHl, Waardcnb urg's syndrOllle , :ll1d related d isorders. Dcmrntol Ciill 6:205-216 . 1988

Reardoll W, Willter RM. Rutlalld 1'. Pulleyn LJ. I-Iolles 13M. Malcolm S: Mutations ill lhc tlbro blast g l'owth f:l ctor receptor 2 gelle <.:~ u se Cl'ollzon syndro me. N(/wn' Gl'lIt/ 8:98-103. 1994

Reith AD, Rotl:i.lpcl R. Giddens E, Brady C. Forreste r L. Bcrnstei n A: I Jl I11U Wnt m ice with mild or severe devcloptncll ial dcfecrs contai n distinci po ilu 1l1uta tions ill the kinase domain of rlic c-kil receptor. GC!IU'S f)c~ I' 4:390-400. 1990

Rutlalld p. Pulle)'11 q . Reardon W, Bara itser M, Ha),ward It, Jones 13, Malcom S. Wi nter Il.....M. O ldridgc M . Siallcy SF. Poole MD , \Vilkic AO: Idcnticn lmutations ill the PCPR2 c'llIse both Pfeiffer ;1IId erOtlZOIl syndrollle phenorypes . NiHUI? Get l/'/ 9: 173- 176, 1995

Schell U, Hehr A, Feldmilll GJ. Robin NH. Zackni EH, de Die-Sllluiders C, Viskochil DH. Stewart JM, WolfF G. O hashi H : M,ttations in FGPRI an d FG I' I\.2 couse f;lIni.li al and sporadic Pfeiffe r syndro llle . 1-1/111/ Alol Celli'l 4 :323-32~, 1995

Si.lvers WI<.: Tile COllI Colol'S 1!l IHirl'. Springer. New York. 1979, pp 20(,-208 Spritz itA, Giebel Ln. l-Ioll11es SA: DOl1li n ~nt ncg;lti ve and loss of fi.lllction muta tions

or the c-kir (mast/stem cell 1!1'(Hvtb rac tor receptor) proto-on cogene in human pichaidisili. Alii) 1-111111 Gellt'l 50:261-269, 19\)2:1

Spritz ItA , I-Io llnes SA. ttin p. Kustcr W: NovelnHH;ltions of the KIT (m:lsr! stCI1l ceU growth factor receptor) prOLO-oll cogellc ill human picbaldisllI . J II1"c.~ t DcnlllHol 1 II 1:22-25, 1993

Spritz rtA. Holmes SA. Ramesar R, GreenbergJ. Curtis I) , Heighlon P: MUt;ltiOLlS of the KIT (rnast/stem cell growth nlc tor receptor) proto-on cogelle accOllnt for a continuous nU lgc ofpllcnotypcs ill 1111111 ~11 piehaldism. A lII) 1-/11111 Gt:IICt 51: 1058-1065, 1992b

Srei nberger D. Mullikcn JB. Mu llcr U: Predispositio n for cysteine substitutions in the il11n1ulloglohll lin- like chain ofFGFR2 ill Crollzo lJ s), lldro lllc. HIIIII Gr/H" 96:1 13-"115. 1995

Tan Je. Nock:1 K, Ray P. Traktlllall P. 13cstn el' P: T'he dOlllinant IV-12 spottiug phenotype rcsuil s from a misscnse mutatioll in the c-ki( receptor ki lliISC. Scirnrr 247:209-2 12. 1990

Tl"ujimura T. Koshimiz u U. I<atoll 1-1. ISOZ;lki 1<, t<an akura Y. TOllo T. Adachi . Kas llga i T. T'ci H. Nislii ll1ll11 c y, NOllltlra S. I<italllurn Y: Mast ce ll Ilumber in Ihe skin of hctc rozygurcs rcflecl"s the molec ular 1I :lture of c-kir mutation. Blood 8 1 :2530-2538. 1993

YilrdclI Y. KU:ln g W-J . Ynng-Fcng: T. COlI SSC I1 S L. MUl1cll1i tsu S. D ull TJ. Chell E. Sch lessinger J. Francke U. U llrich A: 1-lull1 :lIJ prolO-o ncogene c-/':il: a new cell surface receptor tyrosin e killasc fi)r an unidcntified ligand . EIHIJO] 6:33~ 1-·3351. 1 9~7

Z inlllier V. Givol D. Yayon A: Multiple slnu.: tur:a1 elCIlIClltS deterllline liga nd binding or fibroblast growtb factor rcccplOrs . Evidcllce t1wt both Ig domain 2 :md 3 detc rmine rceeplo r specifici ty. ) Bi,,1 C lll'lIl 288:7899 -7903. 1')93