gTPt2pUDbpimstacetgcae

Genomics 68, 22–29 (2000)doi:10.1006/geno.2000.6253, available online at http://www.idealibrary.com on

Refined Localization of Autosomal Recessive NonsyndromicDeafness DFNB10 Locus Using 34 Novel Microsatellite Markers,

Genomic Structure, and Exclusion of Six Known Genesin the Region

Asher Berry,* Hamish S. Scott,† ,1 Jun Kudoh,‡ Ilana Talior,* Michael Korostishevsky,*Marie Wattenhofer,† Michel Guipponi,† Christine Barras,† Colette Rossier,†

Kazunori Shibuya,‡ Jun Wang,‡ Kazuhiko Kawasaki,‡ Shuichi Asakawa,‡Shinsei Minoshima,‡ Nobuyoshi Shimizu,‡ Stylianos Antonarakis,†

and Batsheva Bonne-Tamir* ,2

*Department of Human Genetics and Molecular Medicine, Sackler School of Medicine, Tel-Aviv University, 69978 Ramat-Aviv, Israel;†Division of Medical Genetics, University of Geneva Medical School and Cantonal Hospital of Geneva, 1211 Geneva 4, Switzerland;and ‡Department of Molecular Biology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan

Received March 8, 2000; accepted May 12, 2000

An autosomal recessive nonsyndromic deafness lo-cus, DFNB10, was previously localized to a 12-cM re-

ion near the telomere of chromosome 21 (21q22.3).his locus was discovered in a large, consanguineousalestinian family. We have identified and ordered a

otal of 50 polymorphic microsatellite markers in1q22.3, comprising 16 published and 34 new markers,recisely mapped and ordered on BAC/cosmid contigs.sing these microsatellite markers, the locus forFNB10 has been refined to an area of less than 1 Mbetween markers 1016E7.CA60 and 1151C12.GT45. Sixreviously published cDNAs were mapped to this crit-

cal region, and their genomic structures were deter-ined to facilitate mutation analysis in DFNB10. All

ix genes in this region (in order from centromere toelomere: White/ABCG1, TFF3, TFF2, TFF1, PDE9A,nd NDUVF3) have been screened and eliminated asandidates for DFNB10. The new microsatellite mark-rs and single nucleotide polymorphisms identified inhis study should enable the refined mapping of otherenetic diseases that map to 21q22.3. In addition, theritical region for DFNB10 has been reduced to a sizemenable to an intensive positional cloningffort. © 2000 Academic Press

GenBank/EMBL accession numbers for the cDNAs and genomicsequences of candidate genes are as follows: White (ABCG1) cDNA,X91249; White genomic, AB038161; TFF3 cDNA, L08044; TFF2cDNA, X51698; TFF1 cDNA, X00474; TFF gene cluster, AB038162;PDE9A cDNA, AF067223; PDE9A genomic, AB017602; NDUFV3,cDNA derived from X99726/7/8, genomic, AB038163.

1 Current address: Genetics and Bioinformatics Group, Walter andEliza Hall Institute, Royal Parade, Parkville, P.O. Royal MelbourneHospital, Victoria 3050, Australia.

2 To whom correspondence should be addressed. Telephone: 972-

3-640-9318. Fax: 972-3-640-9900. E-mail: bonne@post.tau.ac.il.

220888-7543/00 $35.00Copyright © 2000 by Academic PressAll rights of reproduction in any form reserved.

INTRODUCTION

An estimated 50% of cases of congenital deafness areinherited. Autosomal recessive loci are estimated toaccount for 80% of nonsyndromic genetic deafness. Au-tosomal dominant defects are responsible in approxi-mately 18% of cases, and defects in mitochondrialgenes or the X chromosome account for the remaining2% of cases (Kalatzis and Petit, 1998). To date, 60 locihave been defined for nonsyndromic deafness (27 au-tosomal dominant, 26 autosomal recessive, 5 X-linked,and 2 mitochondrial) (http://dnalab-www.uia.ac.be/dnalab/hhh). The genes responsible have been identi-fied for a total of 21 loci, illustrating an increasinglydiverse range of proteins and genes with vital functionsin the development of the ear. In an earlier study, weidentified the autosomal recessive nonsyndromic reces-sive deafness locus DFNB10 in a large Palestinianfamily (BT117). DFNB10 was localized to a 12-cM re-gion on chromosome 21q22.3 with a maximum lodscore at D21S1260 but with homozygosity of only themost telomeric marker, D21S1259 (Bonne-Tamir et al.,1996).

Genetic, physical, and transcript maps of the21q22.3 region have been developing rapidly (Chenet al., 1996a). In particular, construction of a se-quence-ready contig and large-scale sequencing havebeen progressing rapidly (http://www.dmb.med.keio.ac.jp/seqpub/map/APECED.html, http://hgp.gsc.riken.go.jp/chr21/marker.html; http://chr21.rz-berlin.mpg.de/sequencing2.html; http://www-eri.uchsc.edu/chromosome21/frames.html; http://www.ncbi.nlm.nih.gov/genome/seq/chr.cgi?CHR521&MIN5100&SRT5ppos&ORG5Hs). Indeed, after chromosome 22

(Dunham et al., 1999), it is likely that chromosome 21

sPctbtthldtNmas

aiw(

cf

DiifidD

gaob

rrTpp

t

ctpwtpwrfs

Dm1Ttu

pDpPp

sXXcf

23REFINED LOCALIZATION OF DFNB10 TO ,1 Mb ON 21q22.3

will be the next human chromosome to have itsgenomic sequence completely determined (http://www.ncbi.nlm.nih.gov/genome/seq/page.cgi?F5HsHome.html&ORG5Hs) (See note added in proof). However,the critical region for the DFNB10 locus as previouslydefined is large and contains many genes. Thus wehave identified additional markers, using the advancedgenomic infrastructure of chromosome 21, to aid in therefinement of the DFNB10 locus.

In the current study, we were able to refine the locusto an area of less than 1 Mb with the aid of 16 pub-lished markers and 34 newly generated markers pre-cisely mapped and ordered on BAC/cosmid contigs. Sixpublished genes that map to this critical region (White,TFF3, TFF2, TFF1, PDE9A, and NDUVF3) have beeneliminated as candidates for DFNB10 by mutationanalyses.

MATERIALS AND METHODS

Sample collection and genomic DNA isolation. The DFNB10 fam-ily studied (BT117), both here and in our earlier publication (Bonne-Tamir et al., 1996), is a large, inbred Palestinian family consisting ofeveral dozen individuals who live in one small town in Israel.ersonal interviews with key figures of this kindred were used tolarify consanguineous relationships over the past seven genera-ions. More than 40 (probable nonsyndromic) deaf individuals haveeen identified within the last three generations. Hearing evalua-ions of affected and unaffected members by pure-tone audiometricests showed severe deafness in the affected individuals, without anyearing remnants, at a level of 75–80 dB. The same level of hearing

oss was evident in all affected individuals, ruling out progressiveeafness. The diagnosis of sensorineural deafness was confirmed inwo 1-week-old twin girls by a brain stem-evoked potential test.one of the deaf individuals showed any signs of vestibular involve-ent, defects in ear morphology, mental retardation, or any other

berrations that could indicate that the deafness was part of ayndrome.Peripheral blood samples were obtained from 36 normal and 16

ffected individuals after an informed consent form was signed andn accordance with institutional guidelines for human subjects. DNAas isolated from blood leukocytes by the proteinase K–SDS method

Sambrook et al., 1989).

Bacterial clone contigs and genomic sequencing. A BAC/cosmidontig spanning MX1–D21S171 was constructed using BAC clonesrom the Keio BAC library (Asakawa et al., 1997) and cosmid clones

from the KU21 “D” chromosome 21-specific cosmid library (Kudoh etal., 1997). Details of the contig are available from http://www.dmb.med.keio.ac.jp/seqpub/map/APECED.html. Genomic se-quencing was performed by the shotgun method (Kawasaki et al.,1997). Additional information to design 3 of the 34 new polymorphicmarkers and to help map the 7 markers telomeric to D21S171 wasderived from publicly available contigs (http://www.ncbi.nlm.nih.gov/genome/seq/chr.cgi?CHR521&MIN5100&SRT5ppos&ORG5Hs;http://chr21.rz-berlin.mpg.de/sequencing2.html; and http://hgp.gsc.riken.go.jp/chr21/marker.html).

Microsatellite polymorphisms. The sequence data of BAC cloneswere searched for possible dinucleotide polymorphisms. Markersfrom other genetic maps (Gyapay et al., 1994; Buetow et al., 1994a,b;

ib et al., 1996; http://http://gdbwww.gdb.org/) were precisely locatedn the BAC clones by PCR and sequence comparison. Primers flank-ng the possible polymorphisms were designed, and PCR ampli-cation was carried out on normal and affected individuals (foretails, please see our Web site: http://medgen.unige.ch/research/

FNB10.html). Primers were designed using Primer3 (http://www-

enome.wi.mit.edu/cgi-bin/primer/primer3.cgi) and the Macintoshpplication Amplify (Bill Engels, Department of Genetics, Universityf Wisconsin). One oligonucleotide primer of each marker was la-eled with [g-32P]ATP using T4 polynucleotide kinase. PCR was

performed on 90 ng of genomic DNA in a total volume of 15 ml pereaction containing 0.4 pM labeled forward primer, 2.6 pM unlabeledeverse primer, a 1.3 mM concentration of each dNTP, and 0.25 Uaq polymerase (Pharmacia) or KlenTaq (Ab Peptides, Inc.). PCRroducts were separated by electrophoresis in a 6% denaturing urea/olyacrylamide gel. Genotypes were scored after autoradiography.

Mutation analyses. The genomic structure and exon–intron junc-ions of published genes that mapped to the DFNB10 critical region

were determined by comparison of the genomic sequence to thecDNA sequences. Unless stated, amplification was carried out on50–100 ng of genomic DNA in a 30- or 40-cycle PCR, in which theinitial 5-min denaturation of template DNA at 94°C was followed bya “touchdown” program for 10 cycles of 94°C for 15–30 s, 65°C for15–30 s (21°C/cycle), 72°C for 1 min, and then 20 or 30 cycles of 94°Cfor 15–30 s, 55°C for 15–30 s, 72°C for 1 min, in a volume of 25 mlontaining 10 mM Tris–HCl, pH 8.3, 50 mM KCl, a 0.2 mM concen-ration of each dNTP, 1.5 mM MgCl2, a 0.5 mM concentration of eachrimer, and 0.5 units of Taq polymerase. PCR products were purifiedith QIAquick spin columns and directly sequenced (d-rhodamine

erminator cycle sequencing) in both directions with appropriaterimers using an ABI 377 automated sequencer. Some PCR productsere cloned into pTOPO (InVitrogen), and a minimum of four clones

epresenting the exon in question were sequenced. Chromatogramsrom normal and affected individuals were compared to the genomicequence using the editor of GAP4 (Staden, 1996).

Statistical analysis. Two-point lod score analysis between theFNB10 gene and the 23 markers, as well as analysis among thearkers, was performed using the computer program LIPED (Ott,

991). The model specifications were previously described (Bonne-amir et al., 1996). Since most of the individuals in the family wereyped, no formal protocol was required to reconstruct haplotypesnambiguously.

Electronic database information. Information regarding map-ing data was obtained from the following Web sites: The Genomeatabase, http://gdbwww.gdb.org/; Hereditary Hearing Loss homeage, http://dnalab-www.uia.ac.be/dnalab/hhh/main.html; and therimer3 program, http://www-genome.wi.mit.edu/cgi-bin/primer/rimer3.cgi.GenBank/EMBL accession numbers of the cDNAs and genomic

equences of candidate genes are as follows: White (ABCG1), cDNA,91249, genomic, AB038161; TFF3 cDNA, L08044; TFF2 cDNA,51698; TFF1 cDNA, X00474; TFF gene cluster, AB038162; PDE9A,DNA, AF067223, genomic, AB017602; NDUFV3, cDNA derivedrom X99726/7/8, genomic, AB038163.

RESULTS

DFNB10 was previously reported to be linked tomarkers located at 21q22.3, spanning an interval ofabout 12 cM between markers D21S1260 and 21qter(Bonne-Tamir et al., 1996). To aid in refining the map-ping position of DFNB10, we have identified a total of50 microsatellite markers on 21q22.3, preciselymapped and ordered on BAC/cosmid contigs, including16 published and 34 potential new microsatellitemarkers. The majority of these new microsatellitemarkers were tested on either normal or affected DNAsamples and shown to be polymorphic (http://medgen.unige.ch/research/DFNB10.html).

Twenty-seven of these microsatellite markers, in-cluding 13 published and 14 new markers derived from

BACs, were typed on DNAs of the DFNB10 family to

ark

24 BERRY ET AL.

narrow the candidate region. After the initial genotyp-ing results were obtained, additional microsatellitemarkers were determined in appropriate intervals tohelp refine the critical region. Critical genotyping re-sults are shown in the pedigree in Fig. 1.

Lod scores were calculated for 23 typed markers inthe pedigree. The results are presented in Table 1.Linkage of adjacent markers confirmed localization ofthe gene for DFNB10 to the distal part of the chromo-somal region 21q22.3 between the markers D21S1260and 274G10.CA85. No recombinants were detectedwithin the interval including the following seven mark-ers: D21S212, 834A1.CA78, White, D21S1225,169B4.GT118, 994G8.CA82, and 161B8.CA32. Themaximum lod score (11.78) was obtained with marker994G8.CA82 (Table 1).

Haplotype reconstruction in the pedigree, for six se-lected markers, is presented in Fig. 1. An informativerecombination was found in affected individual 1208(right-hand side of the pedigree). This recombinationindicated that the gene is localized centromeric to the836E9.GT85 marker. Other recombinations, whichwere found to be carried by individuals in otherbranches of the pedigree, 1108, 1097, 1093, and 1117,were not informative for the gene localization. Theserecombinations were used for estimation of recombina-tion distances between markers and confirmed themarker order.

FIG. 1. Haplotype analysis showing selected m

Affected individuals 1181, 1035, 1115, and 1104 are

all heterozygous for the two most centromeric markers,while affected individual 1111 is heterozygous forone of two telomeric markers (Fig. 1). Other centro-meric and telomeric markers not shown in Fig. 1 giveresults supporting the recombination origin of theobserved heterozygosity. The six markers between1016E7.CA60 and 1151C12.GT45 (834A1.CA78,D21S1225, hWpoly-T, 169B4.GT118, 994G8.CA82, and161B8.CA32) were found to be homozygous in all af-fected individuals (Table 1, Fig. 1). Thus, based on thisobservation, the critical region for DFNB10 can be de-fined between 1016E7.CA60 and 1151C12.GT45, an areaof ,1 cM and ,1 Mb. Attempts to refine the regionfurther with additional markers were not successful. Po-tential markers between 1016E7.CA60 and 834A1.CA78were surrounded by SINE repeats, making them difficultor impossible to use and interpret. On the telomeric side,six potential microsatellite markers between161B8.CA32 (homozygous) and 1151C12.GT45 (het-erozygous in 1111) were attempted (http://medgen.unige.ch/research/DFNB10.html). Again, these markerseither were surrounded by SINE repeats and proved dif-ficult to use and interpret or were not informative in thesamples tested.

The following six known genes were shown to bewithin the DFNB10 critical region, as defined by1016E7.CA60 and 1151C12.GT45, and their exonswere analyzed by direct sequencing for the presence of

ers in the Palestinian DFNB10 family (BT117).

mutations in affected individuals from the Palestinian

1l(td

pddsmcr1vi

1W

m

25REFINED LOCALIZATION OF DFNB10 TO ,1 Mb ON 21q22.3

DFNB10 family (BT117). They are presented in order,from centromere to telomere.

(1) White is a human homologue of the Drosophilawhite gene (also called ABCG1 or ABC8) (Chen et al.,996b). It is an integral membrane protein that be-ongs to the ATP-binding transport protein familyABC transporters), MDR subfamily. As an activeransporter, it may be considered similar to the Pen-red syndrome gene (PDS) mutated in DFNB4 (Li et

al., 1998), a chloride–iodide transporter (Scott et al.,1999). The genomic structure of White was previouslyundescribed. The gene covers 15 exons over 77 kb, andthe direction of transcription is from centromere totelomere. Exons 1 and 2 are grouped in 7 kb, and thelast 13 exons are in 25 kb (Fig. 2). The coding region ofexon 1, 42 bp (14 aa), has proven to be impossible toamplify from normal or DFNB10 DNA using manydifferent strategies, possibly because it is GC-rich,hampering primer design and PCR amplification.Thus, exon 1 of White remains to be formally excluded.Only a single nucleotide change was detected betweenthe normal sequence and the DFNB10 sequence inWhite, an A . G at position 2454 of the cDNA, in the39UTR (Table 2).

The three trefoil factor (TFF) genes are grouped to-gether within 55 kb in the DFNB10 critical region(Seib et al., 1997). All are transcribed from telomere tocentromere.

All three TFF genes are strongly expressed in the

TAB

Lod Scores of 23 Markers T

Marker

Lod score at recombinatio

0 0.05 0.1 0

D21S266 3.34 5.32 5.08 3.D21S1260 2` 2.75 2.73 2.D21S212 7.52 6.67 5.80 4.1016E7.CA60 9.69 8.61 7.53 5.834A1.CA78 10.66 9.45 8.26 5.D21S1225 9.53 8.50 7.47 5.hWpoly-T 7.75 6.91 6.07 4.169B4.GT118 10.60 9.49 8.35 6.994G8.CA82 11.78 10.52 9.25 6.161B8.CA32 11.65 10.40 9.13 6.1151C12GT45 6.90 6.96 6.18 4.274G10.CA85 2` 7.57 6.95 5.D21S1890 4.06 5.77 5.37 3.D21S1885 2` 6.52 6.25 4.953G5.CA29 1.78 4.36 4.08 2.836E9.GT85 2` 3.12 3.06 2.D21S1259 4.11 3.62 3.13 2.D21S1912 2` 4.26 4.42 3.PFKL 2` 3.68 3.83 3.D21S171 2` 2.66 2.92 2.D21S1903 2` 4.85 4.85 3.D21S1897 20.30 1.56 1.62 1.D21S1575 2` 6.49 5.83 4.

digestive system and encode secreted proteins that

may have a role in cell migration and the maintenanceof mucosal integrity. This latter function may beachieved by the protein being a structural componentsof the gastric mucus, stabilizing glycoproteins in themucus gel (Rio et al., 1988). The contents of the en-dolymphatic sac show a prominence of glycosylatedacidic proteins (Thalmann and Thalmann, 1999).COCH, mutated in DFNA9 (Robertson et al., 1998), isexpressed at high levels in the support structures andneural channels of the cochlear and vestibular sys-tems, which show acidic deposits in DFNA9 patients(Khetarpal, 1993; Khetarpal et al., 1991). The COCH

rotein shows homology to von Willebrand factor Aomains, found in a variety of molecules involved inifferent functions including extracellular matrix as-embly. Additionally, COCH shows homology to a do-ain of unknown function in factor C in the horseshoe

rab, Limulus, which, upon binding lipopolysaccha-ides, initiates a coagulation cascade (Robertson et al.,998). Thus by analogy, the TFFs could also be in-olved in the extracellular acidic glycoprotein matricesn the inner ear.

(2) TFF3 (intestinal trefoil factor, ITF (Hauser et al.,993)), is the first of the TFF genes, 14 kb telomeric tohite. It has three exons over 3.3 kb.(3) TFF2 (spasmolytic polypeptide, SP, SML1 (To-asetto et al., 1990)) is 35.5 kb telomeric of TFF3 and

has four exons over 4.5 kb.(4) TFF1 (pS2 protein, breast cancer estrogen-induc-

1

ed in the DFNB10 Family

ractionsMax lod

score

Estimatedrecombination

fraction0.3 0.4

2.34 0.87 0.051.30 0.46 2.78 0.072.41 1.00 7.52 03.31 1.40 9.69 03.53 1.47 10.66 03.38 1.45 9.53 02.75 1.23 7.75 03.77 1.60 10.60 04.08 1.71 11.78 03.98 1.65 11.65 02.51 0.90 7.24 0.023.06 1.16 7.62 0.042.39 0.90 5.78 0.042.98 1.16 6.52 0.051.67 0.65 4.36 0.051.53 0.65 3.14 0.071.32 0.56 4.11 02.26 0.87 4.44 0.081.77 0.62 3.85 0.081.48 0.60 2.92 0.12.44 1.02 4.93 0.070.78 0.32 1.64 0.082.48 0.96 6.65 0.02

LE

yp

n f

.2

8814053884413906655435119780934019590138842819

ible protein, BCE1 (Prud’homme et al., 1985)) is 11.3

Psrlcntgs

wgqa

9

T

T

WP

26 BERRY ET AL.

kb telomeric to TFF2, with three exons over 4.3 kb.Only nonpathogenic sequence variants were detectedin our mutation analyses of the TFF genes in theDFNB10 family (Table 2).

(5) A gap of ;270 kb follows the TFFs before theDE9A gene (Guipponi et al., 1998). Cyclic nucleotide-pecific phosphodiesterases (PDEs) play an essentialole in signal transduction by regulating the intracel-ular concentration of second messengers (cAMP andGMP). None of the identified genes for syndromic oronsyndromic deafness loci has a similar function, andhus PDE9A is a candidate by position. The PDE9Aene is split into 20 exons over 122 kb of genomicequence (Guipponi et al., 1998). No sequence variants

FIG. 2. A map of chromosome region 21q22.3. STSs used to constrto selected markers in the region distal to D21S171. The positions oare also shown. The position, structure, and orientation of the six gethe bottom.

TAB

Nonpathogenic Nucleotide Change

Gene Change

FF2 1 IVS11218G.A2 IVS22454G.Ca

FF1 1 2433C.Ta

2 IVS11187C.Ta

3 c336G.Aa

hite 1 c2454A.GDE9A 1 IVS127insT

2 IVS2238A.G3 IVS229insG4 IVS7247A.G5 IVS122194G.A

NDUFV3 1 IVS22144A.G

Note. GenBank accession numbers are given under Materials and M

a Observed in a control sample.

ere detected in the coding regions, and five nonpatho-enic sequence variants were detected in intronic se-uences included in the amplicons used for mutationnalyses (Table 2).(6) The gene for NADH-ubiquinone oxidoreductase

-kDa subunit (NDUFV3; de Coo et al., 1997) is ;120kb telomeric to PDE9A and comprises three exons over16 kb. The direction of transcription is from centro-mere to telomere. NDUFV3 is a subunit of the mito-chondrial inner membrane complex I, which is com-posed of around 30 different subunits. It transferselectrons from NADH to the respiratory chain. Muta-tions in mitochondrial genes are known to be involvedin both nonsyndromic and syndromic deafness. How-

the BAC/cosmid contig from MX1 to D21S171 are shown in additione BACs/cosmids and the markers mapped to or derived from themin the DFNB10 critical region excluded in this study are shown at

2

etected during Mutation Analyses

Exon/intron DFNB10 Genomic DNA

Intron 1 A GIntron 2 C GPromoter C CIntron 1 C C39UTR A G39UTR A GIntron 1 Ins T —Intron 2 G AIntron 2 Ins G —Intron 7 G AIntron 12 A GIntron 2 G A

hods. Nomenclature is according to Dunnen and Antonarakis (2000).

uctf thnes

LE

s D

et

1Am(M

nrtspattoseie

erqfbD

ap

(wqr

offnc1m

afsslsrdpg

27REFINED LOCALIZATION OF DFNB10 TO ,1 Mb ON 21q22.3

ever, known mitochondrial mutations are in mitochon-drial tRNAs, rRNAs, and/or heteroplasmic large dele-tions. No nuclear encoded mitochondrial protein hasyet to be implicated in deafness (Kalatzis and Petit,1998) http://dnalab-www.uia.ac.be/dnalab/hhh/main.html). However, disruptions of the respiratory chain,specific to the inner ear, could conceivably result indeafness. Only a single nonpathogenic sequence vari-ant was detected in intron 2 of NDUFV3 (Table 2).

DISCUSSION

All changes detected in the six genes studied areconsidered to be nonpathogenic and thus polymor-phisms. However, the precise mapping and ordering of50 microsatellite polymorphisms in 21q22.3 shouldprove useful for the analyses of other complex andmonogenic diseases that map to 21q22.3 including an-other form of nonsyndromic, recessively inherited deaf-ness (NSRD), DFNB8 (MIM 601072, (Veske et al.,996), bipolar affective disorder (Smyth et al., 1997;ita et al., 1999; Straub et al., 1994), and two develop-ental disorders (Knobloch syndrome, MIM 267750

Sertie et al., 1996), and holoprocencephaly HPE1,IM 236100 (Muenke et al., 1995)).It is not clear whether the DFNB8 and DFNB10 loci

are in fact the same. However, the Pakistani (DFNB8)and Palestinian (DFNB10) kindreds, while both hav-ing recessive nonsyndromic deafness, are phenotypi-cally different, having childhood-onset deafness andcongenital deafness, respectively. However, furtherstudies of DFNB8 are limited at this stage by a lack ofDNA from critical meioses.

NSRD is the predominant form of profound child-hood deafness (McKusick, 1992; Fraser, 1976). Previ-ously, homozygosity-mapping analysis in the Palestin-ian family (BT117) suggested that the gene forDFNB10 NSRD is localized to an ;12-cM interval onchromosome 21q22.3 (Bonne-Tamir et al., 1996). The

ew microsatellite polymorphisms, linkage-disequilib-ium analysis, and haplotype analysis have allowedhe refinement of the DFNB10 critical region to a muchmaller interval of less than 1 cM corresponding to ap-roximately 1 Mb, between the markers 1016E7.CA60nd 1151C12.GT45 (Fig. 2). Since all affected members ofhe family have the same haplotype in the interval be-ween these markers, this indicates that all the carriersf this haplotype are most probably descendants of aingle ancestor from a few generations ago. This hypoth-sis is further supported by the fact that all affectedndividuals in our pedigree are the result of consanguin-ous marriages.As we attempted to increase the density of microsat-

llite markers near the borders of the DFNB10 criticalegions, an increasing number of the flanking se-uences contained SINE repetitive elements. Samplesrom all critical meioses in this family have alreadyeen obtained. Thus, further refinement of the

FNB10 critical region may rely on identification of

dditional families and/or the use of single nucleotideolymorphisms on critical meioses.Six candidate genes in the DFNB10 critical region

White, TFF3, TFF2, TFF1, PDE9A, and NDUVF3)ere analyzed for pathogenic mutations by direct se-uencing. All these genes have been excluded as beingesponsible for this form of NSRD.The identification of candidate genes from model

rganisms has had some success in identifying genesor human deafness (e.g., Gibson et al., 1995). However,or DFNB10 none of the available mouse models witheuroepithelial or morphogenetic defects maps to theorresponding syntenic region of mouse chromosome7. ENU mouse mutagenesis may provide additionalodels (e.g., http://www.mgu.har.mrc.ac.uk/mutabase/).The critical region for DFNB10 has been reduced tosize amenable to an intensive positional cloning ef-

ort. HC21 will probably be the second human chromo-ome to be completely sequenced. Almost the entireequence of the DFNB10 region is now available, ateast in draft form. The complete genomic sequencehould greatly facilitate the final discovery of the geneesponsible for DFNB10 with identification of candi-ate genes by function from protein homologies, ex-ression patterns, and finally, mutation analyses of allenes in the defined region.

ACKNOWLEDGMENTS

The laboratory of B.B-T. is supported in part by Grant 1140115from the Applebaum Foundation. We also thank Dr. Sari Weiss andProf. Moshe Frydman of Sackler School of Medicine, and Drs. Lind-say Farrer and Clinton Baldwin of Boston University School ofMedicine for their contribution to the early linkage studies on thisfamily. The laboratory of S.E.A. is supported by Grants 31.40500.94and 31.57149.99 from the Swiss FNRS, 98-3039 from the OFES/EU,and funds from the University and Cantonal Hospital of Geneva. Thelaboratory staff of N.S. thank all members of the genomic sequencingteam in the Laboratory of Genomic Medicine, Keio University Schoolof Medicine for their contribution to this work, which was supportedin part by the Fund for the Human Genome Sequencing Project fromthe Japan Science and Technology Corp. (JST); Grants in Aid forScientific Research on Priority Areas and Scientific Research fromthe Ministry of Education, Science, Sports, and Culture of Japan;and the Fund for “Research for the Future” Program from the JapanSociety for the Promotion of Science (JSPS).

Note added in proof. Sequence of chromosome 21 announced May18, 2000 (The Chromosome 21 Mapping and Sequencing Consortium,2000).

REFERENCES

Aita, V. M., Liu, J., Knowles, J. A., Terwilliger, J. D., Baltazar, R.,Grunn, A., Loth, J. E., Kanyas, K., Lerer, B., Endicott, J., Wang, Z.,Penchaszadeh, G., Gilliam, T. C., and Baron, M. (1999). A compre-hensive linkage analysis of chromosome 21q22 supports prior ev-idence for a putative bipolar affective disorder locus. Am. J. Hum.Genet. 64: 210–217.

Asakawa, S., Abe, I., Kudoh, Y., Kishi, N., Wang, Y., Kubota, R.,Kudoh, J., Kawasaki, K., Minoshima, S., and Shimizu, N. (1997).Human BAC library: Construction and rapid screening. Gene 191:

69–79.

C

T

d

D

D

D

F

G

G

G

H

K

K

K

K

K

L

M

M

O

P

R

R

S

S

S

S

S

28 BERRY ET AL.

Bonne-Tamir, B., DeStefano, A. L., Briggs, C. E., Adair, R., Frank-lyn, B., Weiss, S., Korostishevsky, M., Frydman, M., Baldwin,C. T., and Farrer, L. A. (1996). Linkage of congenital recessivedeafness (gene DFNB10) to chromosome 21q22.3. Am. J. Hum.Genet. 58: 1254–1259.

Buetow, K. H., Ludwigsen, S., Scherpbier-Heddema, T., Quillen, J.,Murray, J. C., Sheffield, V. C., Duyk, G. M., Weber, J. L., Weis-senbach, J., and Gyapay, G. (1994a). Human genetic map. Genomemaps V. Science 265: 2055–2070. [Wall chart]

Buetow, K. H., Weber, J. L., Ludwigsen, S., Scherpbier-Heddema, T.,Duyk, G. M., Sheffield, V. C., Wang, Z., and Murray, J. C. (1994b).Integrated human genome-wide maps constructed using theCEPH reference panel. Nat. Genet. 6: 391–393.

Chen, H., Chrast, R., Rossier, C., Morris, M. A., Lalioti, M. D., andAntonarakis, S. E. (1996a). Cloning of 559 potential exons of genesof human chromosome 21 by exon trapping. Genome Res. 6: 747–760.hen, H., Rossier, C., Lalioti, M. D., Lynn, A., Chakravarti, A.,Perrin, G., and Antonarakis, S. E. (1996b). Cloning of the cDNA fora human homologue of the Drosophila white gene and mapping tochromosome 21q22.3. Am. J. Hum. Genet. 59: 66–75.

he Chromosome 21 Mapping and Sequencing Consortium (2000).The DNA sequence of human chromosome 21. Nature 405: 311–319.

e Coo, R. F., Buddiger, P., Smeets, H. J., and van Oost, B. A. (1997).Molecular cloning and characterization of the human mitochon-drial NADH:oxidoreductase 10-kDa gene (NDUFV3). Genomics45: 434–437.ib, C., Faure, S., Fizames, C., Samson, D., Drouot, N., Vignal, A.,Millasseau, P., Marc, S., Hazan, J., Seboun, E., Lathrop, M., Gya-pay, G., Morissette, J., and Weissenbach, J. (1996). A comprehen-sive genetic map of the human genome based on 5,264 microsat-ellites. Nature 380: 152–154.unham, I., Shimizu, N., Roe, B. A., Chissoe, S., Hunt, A. R., Collins,J. E., Bruskiewich, R., Beare, D. M., Clamp, M., Smink, L. J.,Ainscough, R., Almeida, J. P., Babbage, A., Bagguley, C., Bailey,J., Barlow, K., Bates, K. N., Beasley, O., Bird, C. P., Blakey, S.,Bridgeman, A. M., Buck, D., Burgess, J., Burrill, W. D., andO’Brien, K. P. (1999). The DNA sequence of human chromosome22. Nature 402: 489–495.unnen, J. T., and Antonarakis, S. E. (2000). Mutation nomencla-ture extensions and suggestions to describe complex mutations: Adiscussion. Hum. Mutat. 15: 7–12.

raser, G. R. (1976). Heredity, early identification of hearing loss,and the risk register. In “Early Identification of Hearing Loss,” (G.Mencher, Ed.), pp. 23–32, Basel, Karger.ibson, F., Walsh, J., Mburu, P., Varela, A., Brown, K. A., Antonio,M., Beisel, K. W., Steel, K. P., and Brown, S. D. (1995). A type VIImyosin encoded by the mouse deafness gene shaker-1. Nature 374:62–64.uipponi, M., Scott, H. S., Kudoh, J., Kawasaki, K., Shibuya, K.,Shintani, A., Asakawa, S., Chen, H., Lalioti, M. D., Rossier, C.,Minoshima, S., Shimizu, N., and Antonarakis, S. E. (1998). Iden-tification and characterization of a novel cyclic nucleotide phos-phodiesterase gene (PDE9A) that maps to 21q22.3: Alternativesplicing of mRNA transcripts, genomic structure and sequence.Hum. Genet. 103: 386–392.yapay, G., Morissette, J., Vignal, A., Dib, C., Fizames, C., Millas-seau, P., Marc, S., Bernardi, G., Lathrop, M., and Weissenbach, J.(1994). The 1993–94 Genethon human genetic linkage map. Nat.Genet. 7: 246–339.auser, F., Poulsom, R., Chinery, R., Rogers, L. A., Hanby, A. M.,Wright, N. A., and Hoffmann, W. (1993). hP1.B, a human P-domain peptide homologous with rat intestinal trefoil factor, isexpressed also in the ulcer-associated cell lineage and the uterus.

Proc. Natl. Acad. Sci. USA 90: 6961–6965.

alatzis, V., and Petit, C. (1998). The fundamental and medicalimpacts of recent progress in research on hereditary hearing loss.Hum. Mol. Genet. 7: 1589–1597.awasaki, K., Minoshima, S., Nakato, E., Shibuya, K., Shintani, A.,Schmeits, J. L., Wang, J., and Shimizu, N. (1997). One-megabasesequence analysis of the human immunoglobulin lambda genelocus. Genome Res. 7: 250–261.hetarpal, U. (1993). Autosomal dominant sensorineural hearingloss. Further temporal bone findings. Arch. Otolaryngol. HeadNeck Surg. 119: 106–108.hetarpal, U., Schuknecht, H. F., Gacek, R. R., and Holmes, L. B.(1991). Autosomal dominant sensorineural hearing loss. Pedi-grees, audiologic findings, and temporal bone findings in two kin-dreds. Arch. Otolaryngol. Head Neck Surg. 117: 1032–1042.udoh, J., Nagamine, K., Asakawa, S., Abe, I., Kawasaki, K., Maeda,H., Tsujimoto, S., Minoshima, S., Ito, F., and Shimizu, N. (1997).Localization of 16 exons to a 450-kb region involved in the auto-immune polyglandular disease type I (APECED) on human chro-mosome 21q22.3. DNA Res. 4: 45–52.

i, X. C., Everett, L. A., Lalwani, A. K., Desmukh, D., Friedman,T. B., Green, E. D., and Wilcox, E. R. (1998). A mutation in PDScauses non-syndromic recessive deafness. Nat. Genet. 18: 215–217.cKusick, V. A. (1992). “Mendelian Inheritance in Man,” John Hop-kins Univ. Press, Baltimore.uenke, M., Bone, L. J., Mitchell, H. F., Hart, I., Walton, K., Hall-Johnson, K., Ippel, E. F., Dietz-Band, J., Kvaloy, K., and Fan,C. M. (1995). Physical mapping of the holoprosencephaly criticalregion in 21q22.3, exclusion of SIM2 as a candidate gene forholoprosencephaly, and mapping of SIM2 to a region of chromo-some 21 important for Down syndrome. Am. J. Hum. Genet. 57:1074–1079.tt, J. (1991). “Analysis of Human Genetic Linkage,” Johns HopkinsUniv. Press, Baltimore.

rud’homme, J. F., Fridlansky, F., Le Cunff, M., Atger, M., Mercier-Bodart, C., Pichon, M. F., and Milgrom, E. (1985). Cloning of agene expressed in human breast cancer and regulated by estrogenin MCF-7 cells. DNA 4: 11–21.

io, M. C., Bellocq, J. P., Daniel, J. Y., Tomasetto, C., Lathe, R.,Chenard, M. P., Batzenschlager, A., and Chambon, P. (1988).Breast cancer-associated pS2 protein: Synthesis and secretion bynormal stomach mucosa. Science 241: 705–708.obertson, N. G., Lu, L., Heller, S., Merchant, S. N., Eavey, R. D.,McKenna, M., Nadol, J. B. J., Miyamoto, R. T., Linthicum, F. H. J.,Lubianca, N. J., Hudspeth, A. J., Seidman, C. E., Morton, C. C.,and Seidman, J. G. (1998). Mutations in a novel cochlear genecause DFNA9, a human nonsyndromic deafness with vestibulardysfunction. Nat. Genet. 20: 299–303.

ambrook, J., Fritsch, E. F., and Maniatis, T. (1989). “MolecularCloning: A Laboratory Manual,” Cold Spring Harbor LaboratoryPress, Cold Spring Harbor, NY.

cott, D. A., Wang, R., Kreman, T. M., Sheffield, V. C., and Kar-nishki, L. P. (1999). The Pendred syndrome gene encodes a chlo-ride–iodide transport protein. Nat. Genet. 21: 440–443.

eib, T., Blin, N., Hilgert, K., Seifert, M., Theisinger, B., Engel, M.,Dooley, S., Zang, K. D., and Welter, C. (1997). The three humantrefoil genes TFF1, TFF2, and TFF3 are located within a region of55 kb on chromosome 21q22.3. Genomics 40: 200–202.

ertie, A. L., Quimby, M., Moreira, E. S., Murray, J., Zatz, M.,Antonarakis, S. E., and Passos-Bueno, M. R. (1996). A gene whichcauses severe ocular alterations and occipital encephalocele (Kno-bloch syndrome) is mapped to 21q22.3. Hum. Mol. Genet. 5: 843–847.

iegel, S. (1956). “The c2 test for two independent samples. In “Non-parametric Statistics for the Behavioral Sciences,” McGraw–Hill,

New York.

29REFINED LOCALIZATION OF DFNB10 TO ,1 Mb ON 21q22.3

Smyth, C., Kalsi, G., Curtis, D., Brynjolfsson, J., O’Neill, J., Rifkin,L., Moloney, E., Murphy, P., Petursson, H., and Gurling, H. (1997).Two-locus admixture linkage analysis of bipolar and unipolar af-fective disorder supports the presence of susceptibility loci onchromosomes 11p15 and 21q22. Genomics 39: 271–278.

Staden, R. (1996). The Staden sequence analysis package. Mol. Bio-technol. 5: 233–241.

Straub, R. E., Lehner, T., Luo, Y., Loth, J. E., Shao, W., Sharpe, L.,Alexander, J. R., Das, K., Simon, R., and Fieve, R. R. (1994). Apossible vulnerability locus for bipolar affective disorder on chro-mosome 21q22.3. Nat. Genet. 8: 291–296.

Thalmann, R., and Thalmann, I. (1999). Source and role of en-

dolymph macromolecules. Acta Otolaryngol. 119: 293–296.

Tomasetto, C., Rio, M. C., Gautier, C., Wolf, C., Hareuveni, M.,Chambon, P., and Lathe, R. (1990). hSP, the domain-duplicatedhomolog of pS2 protein, is co-expressed with pS2 in stomach butnot in breast carcinoma. EMBO J. 9: 407–414.

Veske, A., Oehlmann, R., Younus, F., Mohyuddin, A., Muller-Myh-sok, B., Mehdi, S. Q., and Gal, A. (1996). Autosomal recessivenon-syndromic deafness locus (DFNB8) maps on chromosome21q22 in a large consanguineous kindred from Pakistan. Hum.Mol. Genet. 5: 165–168.

Weil, D., Blanchard, S., Kaplan, J., Guilford, P., Gibson, F., Walsh,J., Mburu, P., Varela, A., Levilliers, J., and Weston, M. D. (1995).Defective myosin VIIA gene responsible for Usher syndrome type

1B. Nature 374: 60–61.