Molecular Genetics of Hearing Impairment Due toMutations in Gap Junction Genes Encoding BetaConnexinsRaquel Rabionet,1 Paolo Gasparini,2 and Xavier Estivill1*1Medical and Molecular Genetics Center, Cancer Research Institute, Barcelona, Catalonia, Spain2Medical Genetics Service, Ospedale Casa Sollievo della Sofferenza, San Giovanni Rotondo, Foggia, Italy
Received 21 March 2000; accepted revised manuscript 8June 2000.
*Correspondence to: Xavier Estivill, M.D., Ph.D., Medical andMolecular Genetics Center, Autovia de Castelldefels, Km 2.7,L’Hospitalet de Llobregat, 08907 Barcelona, Catalonia, Spain.E-mail: [email protected]
Contract grant sponsor: Fundació La Marató de TV3; Con-tract grant sponsor: Servei Català de la Salut; Contract grant spon-sor: Fondo de Investigaciones Sanitarias de la Seguridad Social(FIS); Contract grant number: 98/9207.
GAP JUNCTION MUTATIONS AND HEARING LOSS 191
INTRODUCTION
Hearing impairment is a common disorder thataffects about 10% of the general population andits prevalence increases with age. One in 1000 chil-dren is born deaf [Nadol, 1993], with about 60%of cases being due to genetic factors in developedcountries [Morton, 1991; Cohen and Gorlin,1995]. About 20% of subjects over 65 years of agehave severe hearing impairment, mainly due tootosclerosis, presbycusis, and unknown factors,probably as the result of a complex interactionbetween environmental and genetic factors[Gorlin et al., 1995]. Hearing impairment can beclassified in different degrees according to the in-tensity of the hearing loss: mild (25–40 dB), mod-erate (45–60 dB), severe (65–85 dB) and profound(> 85 dB) hearing loss.
The pattern of inheritance of hearing loss can beautosomal dominant, autosomal recessive, X-linked,and mitochondrial. A polygenic/multifactorialethiology can be postulated for the most commonlate-onset cases of hearing impairment. Overall, itis believed that several hundred genes could be in-volved in hearing impairment [Morton, 1991].
About 30% of early onset cases of hearing im-pairment are associated to other clinical featuresand are designated as syndromic forms of deafness[Gorlin et al., 1995]. Most of the non-syndromichearing impairment (NSHI) cases (80%) have anautosomal recessive pattern of inheritance (auto-somal recessive non-syndromic hearing impairment,ARNSHI), about 18% are inherited in an autoso-mal dominant fashion (autosomal dominant non-syndromic hearing impairment, ADNSHI), and1–2% are due to X-linked genes [Cohen and Gorlin,1995]. The role of alterations in the mitochondrialDNA in early onset cases seems to be small, al-though a single mutation (A1555G) accounts for alarge proportion (up to 50%) of familial cases of pro-gressive deafness in some populations [Estivill et al.,1998a; Fischel-Ghodsian, 1999].
As the result of positional cloning, candidategene analysis, and murine model comparison ap-proaches, progress in the identification of genescausing syndromic deafness has been achieved forseveral disorders that include Waardenburg syn-drome types I and II [Tassabehji et al., 1992, 1994]and Pendred syndrome [Li et al., 1998], amongmany others (Van Camp G, Smith RJH. Heredi-tary Hearing Loss Homepage: World Wide WebURL: http://dnalab-www.uia.ac.be/dnalab/hhh/).
Over 50 loci responsible for NSHI have beenidentified. During the last three years 15 of the
genes involved in NSHI have been identified bypositional cloning or positional candidate geneapproaches. Seven genes are responsible forARNSHI: GJB2 (GDB: 125247; MIM# 121011,220290, 601544; GenBank: M86849, U43932;URL: http://www.iro.es/ cx26deaf.html) [Kelsell etal., 1997; Zelante et al., 1997], PDS [Li et al.,1998], MYO7A [Liu et al., 1997], MYO15 [Wanget al., 1998], TECTA [Mustapha et al., 1999],OTOF [Yasunaga et al., 1999], and GJB3 (GDB:127820; MIM# 600101, 603324, 133200; Gen-Bank: AF099730, AF052692, AJ004856) [Xia etal., 1998]; 10 for ADNSHI: GJB2 [Kelsell et al.,1997; Denoyelle et al., 1998], HDIA1 [Lynch etal., 1997], MYO7A [Liu et al., 1997], GJB3 [Xiaet al., 1998], DFNA5 [Van Laer et al., 1998],TECTA [Verhoeven et al., 1998], COCH [Robert-son et al., 1998], POU4F3 [Vahava et al., 1998],KCNQ4 [Kubisch et al., 1999], and GJB6 (GDB:9958357; GenBank: AJ005585) [Grifa et al.,1999]; one for X-linked non-syndromic hearingimpairment: POU3F4 [de Kok et al., 1995]; andone for maternal (mitochondrial) inheritance ofnon-syndromic hearing loss: 12S rRNA [Prezantet al., 1993; Estivill et al., 1998a] (for details seeHereditary Hearing Loss Homepage).
Three of these genes (GJB2, GJB3, and GJB6)encode for connexin proteins (Connexin26,Connexin30, and Connexin31, respectively), andhave been found to be involved in dominant andrecessive NSHI, and in syndromic forms of deaf-ness (Table 1; Connexins and Deafness Homepage:World Wide Web URL: http://www.iro.es/cx26-deaf.html). In addition, mutations in anotherconnexin gene, GJB1 (GDB: 125246; MIM#302800, 304040; GenBank: x04325; URL: http://www.uwcm. ac.uk/uwcm/mg/search/125246.html),cause X-linked Charcot-Marie-Tooth (CMTX), adisorder also associated with late onset hearingimpairment [Bergoffen et al., 1993; Stojkovic etal., 1999].
Connexins are transmembrane proteins thatform channels that allow a rapid transport of ionsand small molecules between cells, with selectiveproperties for size and charge [Bruzzone et al.,1996]. To form these channels connexins aregrouped in hexamers (connexons) that are joinedextracellularly. The assembly of several connexonsforms a gap junction (Fig. 1). Connexins form amultigenic family of more than 13 polytopic mem-brane proteins that are grouped into alpha and betasubtypes, named GJA or GJB followed by a num-ber [reviewed in Bruzzone et al., 1996].
192 RABIONET ET AL.
CONNEXIN26 (GJB2) IS A MAJOR GENE
FOR DEAFNESS
Despite the fact that more than 25 loci havebeen described for non-syndromic autosomal re-cessive deafness (DFNB), a single locus, DFNB1,accounts for a large proportion of families and pa-tients, with variability depending on the popula-tion. This locus on chromosome 13q12 was firstidentified by Guilford et al. [1994] in two consan-guineous Tunisian families. Subsequent studies infamilies from New Zealand [Maw et al., 1995] andPakistan [Brown et al., 1996] suggested that the
DFNB1 locus could be an important one inARNSHI. A more extensive work by Gasparini etal. [1997] showed that DFNB1 was responsible forabout 80% of all recessive deafness families in theMediterranean region, and further reduced theDFNB1 locus region to 14 cM between markersD13S175 and D13S115.
The identification of GJB2 as the gene causingDFNB1 was reached independently by two groups.It was first reported by Kelsell et al. [1997], whosearched for mutations in the GJB2 gene (Con-nexin26) in a family in which palmoplantar kera-
TABLE 1. Phenotypes and Mutations Causing Hearing Loss Due to Mutation in β-Connexin (GJB) Genes
Gene Protein Phenotype Mutation Reference
GJB1 Connexin32 X-linked Charcot-Marie-Tooth and deafness 230 mutations Bergoffen et al. [1993]GJB2 Connexin26 ADNSHI and palmoplantar keratoderma M34T Kelsell et al. [1997]
ADNSHI W44C Denoyelle et al. [1998]ADNSHI and palmoplantar keratoderma R75W Richard et al. [1998b]
ADNSHI and palmoplantar hyperkeratosis G59A Heathcote et al. [2000]ADNSHI and mutilating keratoderma D66E Maestrini et al. [1999]
ARNSHI W77X, W24X Kelsell et al. [1997]ARNSHI 35delG, 167delT Zelante et al. [1997]
GJB3 Connexin31 ADNSHI R180X, E183K Xia et al. [1998]ARNSHI I141V, I141del Liu et al. [2000]
FIGURE 1. Schematic representation of a β-connexin protein (GJB), a connexon, and a gap junction. IC1 to IC3,intracytoplasmic domains; TM1 to TM4, transmembrane domains; EC1 and EC2, extracytoplasmic domains.
GAP JUNCTION MUTATIONS AND HEARING LOSS 193
toderma (PPK) co-occurred with autosomal domi-nant sensorineural deafness. The study was basedon the assumption that GJB2 was a good candi-date for PPK, due to the potential overexpressionof Connexin26 in the epidermis. Although theseinvestigators did not find GJB2 mutations segre-gating with PPK, they detected a methionine tothreonine change at codon 34 (M34T) in some ofthe patients with deafness of this family. SinceGJB2 maps to the chromosomal region whereDFNA3 and also DFNB1 were localized, theseinvestigators analyzed a Pakistani family linked toDFNB1 [Brown et al., 1996] and found homozy-gosity for a nonsense mutation (W77X) in the deafsubjects of this family. They also identified anothernonsense mutation (W24X) in two small Pakistanifamilies [Kelsell et al., 1997]. These results indi-cated that GJB2 was the gene for DFNB1 and prob-ably for DFNA3.
In an independent approach of positional clon-ing and candidate gene analysis of the DFNB1 lo-cus, Zelante et al. [1997] observed severalrecombinants in Caucasoid families from the Medi-terranean region that narrowed the candidate re-gion to approximately 5 cM, between markersD13S141 and D13S232. Since the GJB2 gene waslocalized on chromosome 13 [Willecke et al., 1990;Mignon et al., 1996] within this interval definedby linkage studies, and previous work showed thatcochlea supporting cells are interconnected via gapjunctions, which are formed by connexons con-taining Connexin26 [Kikuchi, 1995], the GJB2gene was considered a good candidate for DFNB1.A deletion of one G within a stretch of six Gs atpositions 30 to 35 of GJB2 (mutation 35delG),causing a frameshift and a premature terminationat codon 13, was detected in 63% of the chromo-somes linked to chromosome 13. Another patienthad a deletion of one T at position 167 (167delT),also causing premature chain termination [Zelanteet al., 1997]. These results confirmed that GJB2was the DFNB1 locus and defined a major muta-tion in deaf patients of Caucasoid origin.
Forty-seven different mutations have been iden-tified in the GJB2 gene in patients with deafness(Table 2). Most GJB2 mutations are located in thecoding region of the gene (exon 2), with the ex-ception of a splice site mutation located at the endof the non-coding exon 1 [Denoyelle et al., 1999;Green et al., 1999]. Eight nonsense mutations (atcodons 24, 44, 47, 57, 65, 77, 124, and 136), 14frameshift mutations, one in-frame deletion (358-360delGAG), and 23 missense mutations havebeen identified (Fig. 2). In addition, five amino acid
changes (V27I, F83L, E114G, G160S, and I203T)have been observed in control subjects, and havebeen considered polymorphisms. Finally, severalnucleotide changes, not leading to amino acid sub-stitutions in Connexin26, have been identified(Table 3).
COMMON MUTATIONS IN GJB2
CAUSING DEAFNESS
Mutations in GJB2 have been detected in manypopulations. While some mutations have beenfound more than once, only few have a high fre-quency in deafness patients. Mutation 35delG[Zelante et al., 1997] (also named 30delG) is com-mon in most Caucasoid populations studied[Carrasquillo et al., 1997; Denoyelle et al., 1997;Zelante et al., 1997; Estivill et al., 1998b; Kelley etal., 1998; Lench et al., 1998a; Lench et al., 1998b;Scott et al., 1998a], but is especially frequent in theMediterranean region. Mutation 167delT, first de-tected by Zelante et al. [1997], is common inAshkenazi Jewish deaf patients [Morell et al., 1998;Sobe et al., 1999]. Finally, mutation 235delC is com-mon among the Japanese population [Fuse et al.,1999; Abe et al., 2000; Kudo et al., 2000] (Table 4).
Mutations 35delG and 235delC occur withinshort repeated DNA sequences. It has been pro-posed that the high frequency of these mutationscould be due to mutational hot spots [Carrasquilloet al., 1997; Estivill et al., 1998b; Fuse et al., 1999].However, since their frequencies vary betweenpopulations (from absent at all to a large propor-tion of deafness alleles), and a mutational hot spotshould have the same probability of mutation inall populations, a single origin with a founder ef-fect and/or an advantage for heterozygous subjects,or both, should be postulated.
Carrier frequencies for these frequent mutationshave been established in different populations[Estivill et al., 1998b; Lench et al., 1998a;Antoniadi et al., 1999; Storm et al., 1999;Gasparini et al., 2000] (Table 4). Gasparini et al.[2000] has studied mutation 35delG in a total of4700 control samples from different Europeancountries and found carrier frequencies rangingfrom 1 in 22 (in Eslovenia) to 1 in 200 (in theUK). The overall 35delG carrier frequency forEurope is 1 in 50, but it is significantly lower in thenorth and center Europe (1 in 75) than in theMediterranean region (1 in 35). Morell et al.[1998] analyzed the Ashkenazi Jewish populationfor 35delG and found a carrier frequency of 1 in138, while the frequency of mutation 167delT was1 in 25. The importance of 167delT in Jewish deaf
194 RABIONET ET AL.
patients has also been pointed out by other stud-ies [Sobe et al., 1999].
No large carrier frequency studies for mutation235delC have been reported in the Asian popula-tion, where this mutation is specially frequent.While the study of Fuse et al. [1999] in 50 controlsubjects has failed to detect 235delC carriers, otherstudies in the Japanese population have detectedone carrier in 48 [Abe et al., 2000] and one in 101[Kudo et al., 2000] control subjects.
GJB2 AND AUTOSOMAL
DOMINANT DEAFNESS
Several missense mutations in GJB2 have beenfound to segregate with hearing impairment in anautosomal dominant fashion (Fig. 2) Although inthe description of GJB2 as a deafness gene it wasconsidered causative of both DFNB1 and DFNA3[Kelsell et al., 1997], its involvement in DFNA3was soon challenged. The M34T mutation wasfound in normal subjects and in a recessive family,
TABLE 2. Mutations in the Connexin26 Gene (GJB2) in Patients With Deafness
Mutation name Nucleotide change Codon Amino acid change Domain Reference
–3170G –3170G>A — Splice site none Denoyelle et al. [1999]M1V 1A→G 1 Met→Val IC1 Estivill et al. [1998a]31del14 del of 14 nt at 31 11–15 Frameshift IC1 Murgia et al. [1999]31del38 del of 38 nt at 31 11–23 Frameshift IC1 Denoyelle et al. [1997]G12V 35G→T 12 Gly→Val IC1 Rabionet et al. [2000]35delG del of G at 30-35 10–12 Frameshift IC1 Zelante et al. [1997]35insG ins of G at 30-35 10–12 Frameshift IC1 Estivill et al. [1998a]51del12insA del of 12 nt at 52 17–21 Frameshift IC1 Sobe et al. [2000]S19T 56G→C 19 Ser→Thr IC1 Rabionet et al. [2000]W24X 71G→A 24 Trp→Stop TM1 Kelsell et al. [1997]M34Ta 101T→C 34 Met→Thr TM1 Kelsell et al. [1997]V37Ib 109G→A 37 Val→Ile TM1 Kelley et al. [1998]W44C 132G→C 44 Trp→Cys EC1 Denoyelle et al. [1998]W44X 132G→A 44 Trp→Stop EC1 Green et al. [1999]G45E 134G→A 45 Gly→Glu EC1 Fuse et al. [1999]E47X 139G→T 47 Glu→Stop EC1 Denoyelle et al. [1997]167delT del of T at 167 56 Frameshift EC1 Zelante et al. [1997]Q57X 169C→T 57 Gln→Stop EC1 Wilcox et al. [1999]G59A 176C→G 59 Gly→Ala EC1 Heathcote et al. [2000]176-191del16 del of 16 nt at 176 59–64 Frameshift EC1 Kudo et al. [2000]Y65X 195C→G 65 Tyr→Stop EC1 Estivill et al. [1998a]D66H 196G→C 66 Asp→His EC1 Maestrini et al. [1999]R75W 223T→G 75 Arg→Trp EC1 Richard et al. [1998b]W77R 229T→C 77 Trp→Arg TM2 Carrasquillo et al. [1997]W77X 231G→A 77 Trp→Stop TM2 Kelsell et al. [1997]235delC del of C at 233-235 78–79 Frameshift TM2 Fuse et al. [1999]V84L 250G→C 84 Val→Leu TM2 Kelley et al. [1998]L90P 269T→C 90 Leu→Pro TM2 Denoyelle et al. [1999]269insT Ins of T at 269 90 Frameshift TM2 Denoyelle et al. [1999]V95M 283G→A 95 Val→Met IC2 Kelley et al. [1998]R98Q 293G→A 98 Arg→Gln IC2 Green et al. [1999]H100Yc 298C→T 100 His→Tyr IC2 Green et al. [1999]299-300delAT del of AT at 299 100 Frameshift IC2 Abe et al. [2000]314del14d del of 14 nt at 314 104–110 Frameshift IC2 Denoyelle et al. [1997]333-334delAA del of AA at 333-335 111–112 Frameshift IC2 Kelley et al. [1998]S113R 339T→G 113 Ser→Arg IC2 Kelley et al. [1998]358-360delGAGe del of GAG at 358 120 Del of Glu 120 IC2 Denoyelle et al. [1999]K122I 365A→T 122 Lys→Ile IC2 Green et al. [1999]Q124X 370C→T 124 Gln→Stop IC2 Scott et al. [1998a]R127H 380G→A 127 Arg→His IC2 Estivill et al. [1998a]Y136X 408C→A 136 Tyr→Stop IC2 Fuse et al. [1999]R143W 427C→T 143 Arg→Trp IC2 Brobby et al. [1998]509insA ins of A at 509 170 Frameshift TM3 Denoyelle et al. [1999]P175T 523C→T 175 Pro→Thr EC2 Denoyelle et al. [1999]R184P 551G→C 184 Arg→Pro EC2 Denoyelle et al. [1997]S199F 596C→T 199 Ser→Phe EC2 Green et al. [1999]631-632delGT del of GT at 631-632 210 Frameshift IC3 Kelley et al. [1998]aMutation detected in dominant and recessive cases and in the general population.bFirst described as a polymorphism.cThis change was found in heterozygosis with M34T in a recessive family.dMutations 310del14 and 312del14 are probably the same mutation as 314del14.eWrongly described as DelE118.IC1-3, intracellular domains; TM1-4, transmembrane domains; EC1,2, extracellular domains.
GAP JUNCTION MUTATIONS AND HEARING LOSS 195
in both affected and non-affected individuals[Scott et al., 1998a; Scott et al., 1998b]. Thismutation has also been detected in families withARNSHL with another GJB2 mutation beingpresent in the other allele [Kelley et al., 1998].M34T has been detected in about 2% of sub-jects in some populations [Green et al., 1999].Interestingly, the same amino acid change hasbeen found in the Connexin32 (GJB1) gene inpatients with CMTX [Tan et al., 1996], andthree other amino acid changes affecting thesame methionine have been described in CMTXpatients (Fig. 3). M34T is a missense mutationaffecting the first transmembrane region of GJB2.The methionine at this position is conserved inGJB2 and GJB1, but not in other β-connexins,where there is a leucine at this position. Thus,mutation M34T functional studies, its presence
in other disorders, and different forms of inherit-ance suggest that M34T possibly is a partially pen-etrant mutation with a double role in dominantand recessive hearing impairment and with apotential involvement in PPK.
The detection of mutation W44C in the origi-nal family that allowed the assignment of theDFNA3 locus clearly confirms the relationshipbetween GJB2 and DFNA3 [Denoyelle et al.,1998]. Three other missense mutations in GJB2have been found to cause dominant deafness,R75W, D66H, and G59A. But in all cases thesemutations are associated to other clinical features(Table 1 and Fig. 2). Richard et al. [1998b] identi-fied mutation R75W in two affected members of afamily with deafness and diffuse keratodermasegregating together, very similar to the family pre-sented by Kelsell et al. [1997] with mutation M34T.
FIGURE 2. Schematic representation ofConnexin26 (GJB2) and indication ofamino acid substitutions involved in domi-nant and recessive hearing loss. IC1 toIC3, intracytoplasmic domains; TM1 toTM4, transmembrane domains; EC1 andEC2, extracytoplasmic domains. Lightgrey, recessive mutations; dark grey andasterisk, dominant mutations.
TABLE 3. Polymorphisms in the Connexin26 Gene (GJB2)*
Mutation name Nucleotide change Codon Amino acid change Domain Reference
V27I 79G→A 27 Val→Ile TM1 Kelley et al. [1998]I30I 90T→A/C 30 None TM1 Zelante et al. [1997]F83L 249C→G 83 Phe→Leu TM2 Scott et al. [1998a]L89L 267C→A 89 None TM2 Zelante et al. [1997]E114G 341A→G 114 Glu→Gly IC2 Fuse et al. [1999]I128I 384C→T 128 None IC2 Zelante et al. [1997]G160S 478G→A 160 Gly→Ser EC2 Scott et al. [1998a]V182V 546G→C 182 None EC2 Soctt et al. [1998a]I203T 608T→C 203 Ile→Tre TM4 Abe et al. [2000]
*There are other polymorphisms in GJB2 that have been described but are not published and can be consulted at the connexinsand deafness website (http://www.iro.es/cx26deaf.html)IC1-3, intracellular domains; TM1-4, transmembrane domains; EC1,2, extracellular domains.
196 RABIONET ET AL.
TABLE 4. Carrier Frequency for Mutations 35delG, 235delC and 167delT in Different Populations
Mutation Population % n Reference
35delG Caucasoid (North European) 2.8 1826 Gasparini et al. [2000]Caucasoid (South European) 1.3 1444 Gasparini et al. [2000]Caucasoid (US) 2.5 560 Green et al. [1999]Caucasoid (Belgian) 2.5 360 Storm et al. [1999]Caucasoid (Spanish and Italian) 3.2 280 Estivill et al. [1998a]Caucasoid (Greek) 3.5 395 Antoniadi et al. [1999]Caucasoid (US) 0.5 173 Morell et al. [1998]Caucasoid (US) 1 100 Scott et al. [1998a]Caucasoid (US) 2.1 96 Kelley et al. [1998]Ashkenazi Jewish 0.7 551 Morell et al. [1998]Ashkenazi Jewish (US) 0 89 Gasparini et al. [2000]North African Jews 2 100 Gasparini et al. [2000]Persian Jews 3.4 59 Gasparini et al. [2000]Iraqui Jews 0.9 115 Gasparini et al. [2000]Asian 0 53 Morell et al.[1998]African American 0 173 Morell et al. [1998]African American 0 190 Gasparini et al. [2000]Arab 1 58 Gasparini et al. [2000]
167delT Caucasoid (US) 0 175 Morell et al. [1998]Caucasoid (US) 0 96 Kelley et al. [1998]Caucasoid 0 50 Zelante et al. [1997]Ashkenazi Jewish 4 546 Morell et al. [1998]African American 0 171 Morell et al. [1998]Asian 0 52 Morell et al. [1998]
235delC Japanese 1.0 203 Kudo et al. [2000]Japanese 2.1 96 Abe et al. [2000]Japanese 0 50 Fuse et al. [1999]
FIGURE 3. Amino acid residues involved in the β-connexin disorders hearing impairment and X-linked Charcot-Marie-Tooth. IC1 and IC2, intracytoplasmic domains; TM1 to TM4, transmembrane domains; EC1 and EC2, extracytoplasmicdomains. Residues in black correspond to the most conserved residue at this position among connexins. Residues in greyare non-conserved residues. Underlined aminoacids correspond to those for which a mutation involved in deafness or X-linked Charcot-Marie-Tooth has been described.
GAP JUNCTION MUTATIONS AND HEARING LOSS 197
This mutation was nonetheless also identified inone control subject, again raising the issue of par-tial penetrance for some of the GJB2 missensemutations. Arginine at position 75 is conservedamong all connexins, suggesting that this residueis critical for the function of the protein. Anotherdominant mutation (D66H) was identified in threefamilies presenting sensorineural deafness andmutilating keratoderma [Maestrini et al., 1999].This mutation seems to be fully penetrant, as allsubjects in the three pedigrees presenting D66Hwere affected. Heathcote et al. [2000] identifiedmutation G59A in a family with an autosomaldominant syndrome with hearing loss andpalmoplantar hyperkeratosis.
FUNCTIONAL STUDIES OF
GJB2 MUTATIONS
Functional studies should help us to understandthe behavior of GJB2 missense mutations, espe-cially for those proposed to be dominant. Func-tional studies in Xenopus laevis oocytes have shownthat M34T has a dominant negative effect on gapjunction transport of current and ions [White etal., 1998], suggesting that this is a dominant mu-tation. Functional studies in paired Xenopus laevisoocytes have also indicated that R75W has a domi-nant negative effect. Dye transfer studies in HeLacells performed by Martin et al. [1999] for muta-tions M34T, W77R, and W44C [described byDenoyelle et al., 1998] in a large autosomal domi-nant family confirmed an impaired function for allthree mutations, with slight residual transfer in thecase of W77R and M34T. While W77R was ineffi-ciently targeted to the plasma membrane, andM34T failed to form hexameric hemichannels,W44C reached the membrane and efficientlyformed the connexons. The non-functionality ofW44C has been attributed to a modification of thethree-dimensional structure of the protein [Mar-tin et al., 1999]. We have to take into account atthis stage that the data obtained on the function-ality of the different mutations is, nonetheless, onlyvalid for in vitro situations, since the function ofconnexins is only achieved after a complex oligo-merization process. Further studies are still neces-sary to completely understand the functionality ofthese “dominant” connexin alleles.
HEARING IMPAIRMENT CONSEQUENCES
OF GJB2 MUTATIONS
Clinical features of patients affected by muta-tions in GJB2 are quite variable, but most of themare responsible for childhood onset NSHI.
[Denoyelle et al., 1999] analyzed over 100 patientsand concluded that DFNB1 patients (consideringDFNB1 patients those that have two mutationsin GJB2) are prelingually deaf, and present similardegrees of hearing loss for both ears, generatingsloping or flat audiometric curves. The degree ofdeafness in these patients involves all ranges, butmost patients have a severe to profound hearingimpairment. Nevertheless, patients that were com-pound heterozygous for 35delG and another mu-tation were more common severely affected, while35delG homozygous were predominant among pro-found or severely affected patients. Intrafamilialvariability of the severity of hearing loss is not un-common, and it can vary from moderate to severe,from moderate to profound, or even from mild toprofound.
Cohn et al. [1999] did not find any consistentaudiologic phenotype for DFNB1. Severity of deaf-ness also varied from mild-moderate to profound,even in patients homozygous for 35delG, and cer-tain cases presented progression too, with no rela-tion to the causative mutation. Data of this studysuggest that patients homozygous for 167delT havea more severe and less variable phenotype thanpatients that are homozygous for 35delG. How-ever, these differences could be attributed to ge-netic background, since all 167delT homozygouspatients were of Ashkenazi Jewish origin. Cohn etal. [1999] also detected two individuals with ves-tibular dysfunction, one of them with vertigo.
The main conclusion that can be drawn fromthese studies is that the phenotype due to GJB2mutations is highly variable, and mostly indepen-dent of the mutation. This seems to indicate thatthere are either environmental factors or modify-ing genes that can affect the phenotype of a par-ticular GJB2 genotype. One possibility would bethat a second connexin gene shares partial func-tional redundancy with GJB2, but other genes orenvironmental factors can not be excluded. Nowthat we know that other connexins (GJB3 andGJB6) are involved in hearing impairment, it wouldbe interesting to study wether variants in thesegenes [Lopez-Bigas et al., 2000] could play a rolein the severity of the hearing loss of patients thatbear GJB2 mutations.
CONNEXIN31 DEAFNESS AND
SKIN DISORDERS
Richard et al. [1998a] identified three muta-tions in the Connexin31 gene (GJB3) in four fami-lies with erythrokeratodermia variabilis (EKV).Independently, Xia et al. [1998] presented the
198 RABIONET ET AL.
cloning of the human GJB3 gene on chromosome1p33-p35 and found mutations in two small deaf-ness families. In both cases the disorders were in-herited in an autosomal dominant pattern withpartial penetrance (female carriers have eithersubclinical deafness or normal hearing). The aminoacids that are mutated in each disease lie in differ-ent Connexin31 protein domains. In EKV themutations are of the missense type and affect resi-dues 12 and 86. The GJB3 deafness mutationsoccur at amino acid positions 180 and 183; muta-tion at R180 being a nonsense change, which gen-erates a putative truncated protein lacking thefourth transmembrane domain and the third in-tracellular domain (Fig. 4). The fact that somecarriers of GJB3 mutations presented a normalphenotype challenges the involvement of thesemutations in dominant deafness.
GJB3 has recently been related to early onsetautosomal recessive deafness. Two mutations af-fecting the same amino acid (I141) were detectedin compound heterozygosity in two non-consan-guineous Chinese families [Liu et al., 2000]. Iso-leucine 141 lies in the third transmembranedomain, which seems to be critical for the forma-tion of the wall of the gap junction pore. The fact
that mutations in the same gene cause both domi-nant and recessive deafness (as occurs with GJB2)strongly implies that some missense mutationsshould have a dominant negative effect, sincehaploinsufficiency is not sufficient to cause a phe-notype. Functional studies for Connexin31 muta-tions have not yet been performed. Clinical datain patients with GJB3 mutations indicate that thereis a wide variability in the age of onset of hearingloss [Xia et al., 1998; Liu et al., 2000], with somechanges being also present in subjects with appar-ent normal hearing [Richard et al., 1998a]. Thisvariability depends on the type and location of theGJB3 change (Table 5).
CONNEXIN30 AND
SENSORINEURAL DEAFNESS
The gene for human Connexin30 (GJB6) hasbeen cloned recently [Grifa et al., 1999] and athreonine to methionine change at position 5(T5M) has been found in three members of a fam-ily affected by bilateral middle/high-frequencyhearing impairment. Functional studies have in-dicated the non-functionality of the T5M GJB6,demonstrating that it has a dominant negative ef-fect (Table 5).
TABLE 5. Mutations in the Connexin30 (GJB6) and Connexin31 (GJB3) Genes in Patients With Hearing Impairment
GJB3R180X 538C→T 180 Arg→Stop EC2 Adult onset Xia et al. [1998]E183K 547G→A 183 Glu→Lys EC2 Adult onset Xia et al. [1998]I141del 423-425delATT 141 Deletion of Ile TM3 Childhood onset Liu et al. [2000]I141V 423A→G 141 Ile→Val TM3 Childhood onset Liu et al. [2000]
IC1, first intracellular domain; EC2, second extracellular domain; TM3, third transmembrane domain.
FIGURE 4. Schematic representation of β-connexins 26, 30, and 31 (GJB2, GJB6, and GJB3, respectively). The locationof dominant mutations involved in deafness is indicated.
GAP JUNCTION MUTATIONS AND HEARING LOSS 199
GJB6 has been localized 800 kb centromeric toGJB2 on chromosome 13q12 [Grifa et al., 1999].GJB6 and GJB2 are both expressed in the cochlea[Grifa et al., 1999; Lautermann et al., 1999]. Thesetwo connexins are very similar (76% identity), butGJB6 has a carboxy terminus which is 37 aminoacids longer than GJB2 (Fig. 4). The involvementof GJB6 in dominant deafness indicates that it isthe second gene that causes DFNA3. Although ithas been proposed that mutations in GJB6 couldexplain those cases of GJB2 with only one changeidentified [Rabionet et al., 2000], suggesting adigenic inheritance, patients with mutations inboth genes have not yet been described.
FUTURE PROSPECTS
The identification of mutations in connexingenes in patients with hearing impairment opensnew perspectives for molecular diagnosis, carrierdetection, and even prenatal diagnosis of deafness.Molecular diagnosis of these genes is facilitated bythe fact that most mutations are localized in thesingle coding exon of each of these genes. In addi-tion, the presence of a single common mutation inGJB2 in specific populations provides the tools forcarrier detection and even for screening in popu-
lation groups that are likely to have hearing im-pairment.
It has been hypothesized that the abnormali-ties in gap junctions of the supportive cells of thecochlea are responsible for hearing impairment inpatients with connexin mutations. This anomalyin gap junctions is expected to impair the recy-cling of potassium at the endolymph level (Fig. 5).However, this has not been proven for GJB2 orthe other connexins involved in deafness. In addi-tion, the patterns of expression of differentconnexins seems to be different in the cochlea[Lautermann et al., 1999]. While Connexins 26and 30 are expressed during development of therat cochlea, this is not the case of Connexin31,which has not been found to be expressed in thecochlea [Lautermann et al., 1998]. In addition,since Connexin32 is involved in sensorineuralneuropathy, also causing hearing impairment, it isalso possible that some of these connexins are alsoexpressed in Schwann cells and that hearing losscould occur through different mechanisms for dif-ferent connexins.
If hearing impairment due to GJB2 mutationsis not due to developmental abnormalities of thecochlea, then the correction of the molecular al-
FIGURE 5. Schematic representation of a hair cell and supportive cells of the cochlea. Recycling of potassium ions ishighly dependent on gap junctions in the supportive cells. GJB2 and GJB6 are expressed in these cells and probablyparticipate in the recycling of potassium. The role of GJB3 in audition and hearing impairment is still unknown.
200 RABIONET ET AL.
terations by gene therapy could be possible earlyafter birth. However, it is necessary to know thestatus of the cochlea and the potential damage ofhair cells and other structures due to GJB2 muta-tions. Gene therapy approaches for congenitaldeafness due to GJB2 mutations could involve theuse of viruses, liposomes, or other vectors thatcould drive a transgene to the supportive cells ofthe cochlea [Lalwani et al., 1996]. Gene therapyapproaches will have to be tested first in mousemodels for GJB2 mutations. Unfortunately, thehomozygous null mouse for Gjb2 is not viable[Gabriel et al., 1998] and a conditional knockoutwith specific expression in the cochlea has to bedeveloped. Although there is still a long way to goto obtain a correction of deafness, it seems thatresearch in congenital hearing impairment is onthe right track and that we will see progress inunderstanding, diagnosis, and therapy.
ACKNOWLEDGMENTS
This work was supported by grants from theServei Català de la Salut (to X.E.) and the FIS(98/9207 to R.R.).
REFERENCES
Abe S, Usami S, Shinkawa H, Kelley PM, Kimberling WJ, AskewJW, Smith SD, Usami S. 2000. Prevalent connexin 26 gene(GJB2) mutations in Japanese. J Med Genet 37:41–43.
Antoniadi T, Rabionet R, Kroupis C, Aperis GA, Economides J,Petmezakis J, Economou-Petersen E, Estivill X, Petersen MB.1999. High prevalence in the Greek population of the 35delGmutation in the connexin 26 gene causing prelingual deaf-ness. Clin Genet 55:381–382.
Bergoffen J, Scherer SS, Wang S, Scott MO, Bone LJ, Paul DL,Chen K, Lensch MW, Chance PF, Fischbeck KH. 1993.Connexin mutations in X-linked Charcot-Marie-Tooth dis-ease. Science 262:2039–2042.
Brobby GW, Muller-Myhsok B, Horstmann RD. 1998. Connexin26 R143W mutation associated with recessive nonsyndromicsensorineural deafness in Africa. N Engl J Med 338: 548–550.
Brown KA, Janjua AH, Karbani G, Parry G, Noble A, CrockfordG, Bishop DT, Newton VE, Markham AF, Mueller RF. 1996.Linkage studies of non-syndromic recessive deafness (NSRD)in a family originating from the Mirpur region of Pakistanmaps DFNB1 centromeric to D13S175. Hum Mol Genet5:169–173.
Bruzzone R, White TW, Paul DL. 1996. Connections withconnexins: the molecular basis of direct intercellular signal-ing. Eur J Biochem 238:1–27.
Carrasquillo MM, Zlotogora J, Barges S, Chakravarti A. 1997.Two different connexin 26 mutations in an inbred kindredsegregating non-syndromic recessive deafness: implicationsfor genetic studies in isolated populations. Hum Mol Genet6:2163–2172.
Cohen MM Jr, Gorlin RJ. 1995. Epidemiology, etiology and ge-netic patterns. In: Gorlin RJ, Toriello HV, Cohen MM Jr. edi-tors. Hereditary hearing loss and its syndromes. Oxford:Oxford University Press. p 9–21.
WJ. 1999. Clinical studies of families with hearing loss attrib-utable to mutations in the connexin 26 gene (GJB2/DFNB1).Pediatrics 103: 546–550.
de Kok YJ, van der Maarel SM, Bitner-Glindzicz M, Huber I,Monaco AP, Malcolm S, Pembrey ME, Ropers HH, CremersFP. 1995. Association between X-linked mixed deafness andmutations in the POU domain gene POU3F4. Science267:685–688.
Denoyelle F, Weil D, Maw MA, Wilcox SA, Lench NJ, Allen-Powell DR, Osborn AH, Dahl HH, Middleton A, HousemanMJ, Dode C, Marlin S, Boulila-ElGaied A, Grati M, Ayadi H,BenArab S, Bitoun P, Lina-Granade G, Godet J, Mustapha M,Loiselet J, El-Zir E, Aubois A, Joannard A, Levilliers J,Garabédian EN, Mueller RF, McKinlay Gardner RJ, Petit C.1997. Prelingual deafness: high prevalence of a 30delG muta-tion in the connexin 26 gene. Hum Mol Genet 6: 2173-2177
Denoyelle F, Lina-Granade G, Plauchu H, Bruzzone R, Chaib H,Levi-Acobas F, Weil D, Petit C. 1998. Connexin 26 gene linkedto a dominant deafness. Nature 393:319–320.
Denoyelle F, Marlin S, Weil D, Moatti L, Chauvin P, GarabedianEN, Petit C. 1999. Clinical features of the prevalent form ofchildhood deafness, DFNB1, due to a connexin-26 gene de-fect: implications for genetic counselling. Lancet 353:1298–1303.
Estivill X, Fortina P, Surrey S, Rabionet R, Melchionda S,D’Agruma L, Mansfield E, Rappaport E, Govea N, Mila M,Zelante L, Gasparini P. 1998a. Connexin-26 mutations insporadic and inherited sensorineural deafness. Lancet 351:394–398.
Estivill X, Govea N, Barcelo E, Badenas C, Romero E, Moral L,Scozzri R, D’Urbano L, Zeviani M, Torroni A. 1998b. Famil-ial progressive sensorineural deafness is mainly due to themtDNA A1555G mutation and is enhanced by treatment ofaminoglycosides. Am J Hum Genet 62:27–35.
Fischel-Ghodsian N. 1999. Mitochondrial deafness mutationsreviewed. Hum Mutat 13:261–270.
Fuse Y, Doi K, Hasegawa T, Sugii A, Hibino H, Kubo T. 1999.Three novel connexin26 gene mutations in autosomal reces-sive non-syndromic deafness. Neuroreport 10:1853–1857.
Gabriel HD, Jung D, Butzler C, Temme A, Traub O, WinterhagerE, Willecke K. 1998. Transplacental uptake of glucose is de-creased in embryonic lethal connexin26-deficient mice. J CellBiol 140:1453–1461.
Gasparini P, Estivill X, Volpini V, Totaro A, Castellvi-Bel S, GoveaN, Mila M, Della Monica M, Ventruto V, De Benedetto M,Stanziale P, Zelante L, Mansfield ES, Sandkuijl L, Surrey S,Fortina P. 1997. Linkage of DFNB1 to non-syndromic neuro-sensory autosomal-recessive deafness in Mediterranean fami-lies. Eur J Hum Genet 5:83–88.
Gasparini S, Rabionet R, Barbujani G, Melchionda S, PetersenP, Brondum-Nielsen K, Metspalu A, Oitmaa E, Pisano M,Fortina P, Estivill X, Genetic Analysis Consortium of GJB235delG. 2000. High carrier frequency of the 35delG deaf-ness mutation in European populations. Eur J Hum Genet8:19–23.
Gorlin RJ, Toriello HV, Cohen MMJ. 1995. Hereditary hearingloss and its syndromes. Oxford: Oxford University Press.
Green GE, Scott DA, McDonald JM, Woodworth GG, SheffieldVC, Smith RJ. 1999. Carrier rates in the midwestern UnitedStates for GJB2 mutations causing inherited deafness. Jama281:2211–2216.
Grifa A, Wagner CA, D’Ambrosio L, Melchionda S, Bernardi F,Lopez-Bigas N, Rabionet R, Arbones M, Monica MD, EstivillX, Zelante L, Lang F, Gasparini P. 1999. Mutations in GJB6cause nonsyndromic autosomal dominant deafness at DFNA3locus. Nat Genet 23:16–18.
GAP JUNCTION MUTATIONS AND HEARING LOSS 201
Guilford P, Ben Arab S, Blanchard S, Levilliers J, Weissenbach J,Belkahia A, Petit C. 1994. A non-syndrome form of neuro-sensory, recessive deafness maps to the pericentromeric re-gion of chromosome 13q. Nat Genet 6:24–28.
Heathcote K, Syrris P, Carter ND, Patton MA. 2000. A connexin26 mutation causes a syndrome of sensorineural hearing lossand palmoplantar hyperkeratosis (MIM 148350). J Med Genet37:50–51.
Kelley PM, Harris DJ, Comer BC, Askew JW, Fowler T, SmithSD, Kimberling WJ. 1998. Novel mutations in the connexin26 gene (GJB2) that cause autosomal recessive (DFNB1)hearing loss. Am J Hum Genet 62:792–799.
Kikuchi T, Kimura RS, Paul DL, Adams JC. 1995. Gap junctionsin the rat cochlea: immunohistochemical and ultrastructuralanalysis. Anat Embryol (Berl) 191:101–118.
Kubisch C, Schroeder BC, Friedrich T, Lutjohann B, El-AmraouiA, Marlin S, Petit C, Jentsch TJ. 1999. KCNQ4, a novel po-tassium channel expressed in sensory outer hair cells, is mu-tated in dominant deafness. Cell 96:437–446.
Kudo T, Ikeda K, Kure S, Matsubara Y, Oshima T, Watanabe K,Kawase T, Narisawa K, Takasaka T. 2000. Novel mutationsin the Connexin 26 gene (GJB2) responsible for childhooddeafness in the Japanese population. Am J Med Genet 90:141–145.
Lalwani AK, Walsh BJ, Reilly PG, Muzycza N, Mhatre AN. 1996.Development of in vivo gene therapy for hearing disorders:introduction of adeno-associated virus into the cochlea of theguinea pig. Gene Ther 3:588–592.
Lautermann J, ten Cate WJ, Altenhoff P, Grummer R, Traub O,Frank H, Jahnke K, Winterhager E. 1998. Expression of thegap-junction connexins 26 and 30 in the rat cochlea. CellTissue Res 294:415–420.
Lautermann J, Frank HG, Jahnke K, Traub O, Winterhager E,ten Cate WJ, Altenhoff P, Grummer R, Frank H. 1999. De-velopmental expression patterns of connexin26 and -30 inthe rat cochlea. Dev Genet 25:306–311.
Lench N, Houseman M, Newton V, Van Camp G, Mueller R.1998a. Connexin-26 mutations in sporadic non-syndromalsensorineural deafness. Lancet 351:415.
Lench NJ, Markham AF, Mueller RF, Kelsell DP, Smith RJ, WillemsPJ, Schatteman I, Capon H, Van De Heyning PJ, Van CampG. 1998b. A Moroccan family with autosomal recessive sen-sorineural hearing loss caused by a mutation in the gap junc-tion protein gene connexin 26 (GJB2). J Med Genet35:151–152.
Li XC, Everett LA, Lalwani AK, Desmukh D, Friedman TB, GreenED, Wilcox ER. 1998. A mutation in PDS causes non-syndromic recessive deafness. Nat Genet 18:215–217.
Liu XZ, Walsh J, Mburu P, Kendrick-Jones J, Cope MJ, Steel KP,Brown SD. 1997. Mutations in the myosin VIIA gene causenon-syndromic recessive deafness. Nat Genet 16:188–190.
Liu XZ, Xia XJ, Xu LR, Pandya A, Liang CY, Blanton SH, BrownSD, Steel KP, Nance WE. 2000. Mutations in connexin31underlie recessive as well as dominant non- syndromic hear-ing loss. Hum Mol Genet 9:63–67.
Lopez-Bigas N, Rabionet R, Martínez E, Banchs I, Volpini V,Vance JM, Arbones ML, Estivill X. 2000. Identification ofseven novel SNPs (five nucleotide and two amino acid sub-stitutions) in the connexin31 (GJB3) gene. Hum Mutat (online) 15:481–482.
Lynch ED, Lee MK, Morrow JE, Welcsh PL, Leon PE, King MC.1997. Nonsyndromic deafness DFNA1 associated with mu-
tation of a human homolog of the Drosophila gene diapha-nous. Science 278:1315–1318.
Maestrini E, Korge BP, Ocana-Sierra J, Calzolari E, Cambiaghi S,Scudder PM, Hovnanian A, Monaco AP, Munro CS. 1999. Amissense mutation in connexin26, D66H, causes mutilatingkeratoderma with sensorineural deafness (Vohwinkel’s syn-drome) in three unrelated families. Hum Mol Genet 8:1237–1243.
Martin PE, Coleman SL, Casalotti SO, Forge A, Evans WH. 1999.Properties of connexin26 gap junctional proteins derived frommutations associated with non-syndromal hereditary deafness.Hum Mol Genet 8:2369–2376.
Maw MA, Allen-Powell DR, Goodey RJ, Stewart IA, NancarrowDJ, Hayward NK, Gardner RJ. 1995. The contribution of theDFNB1 locus to neurosensory deafness in a Caucasian popu-lation. Am J Hum Genet 57:629–635.
Mignon C, Fromaget C, Mattei MG, Gros D, Yamasaki H, MesnilM. 1996. Assignment of connexin 26 (GJB2) and 46 (GJA3)genes to human chromosome 13q11—q12 and mouse chro-mosome 14D1-E1 by in situ hybridization. Cytogenet CellGenet 72:185–186.
Morell RJ, Kim HJ, Hood LJ, Goforth L, Friderici K, Fisher R,Van Camp G, Berlin CI, Oddoux C, Ostrer H, Keats B, Fried-man TB. 1998. Mutations in the connexin 26 gene (GJB2)among Ashkenazi Jews with nonsyndromic recessive deafness.N Engl J Med 339:1500–1505.
Morton NE. 1991. Genetic epidemiology of hearing impairment.Ann N Y Acad Sci 630:16–31.
Murgia A, Orzan E, Polli R, Martella M, Vinanzi C, Leonardi E,Arslan E, Zacchello F. 1999. Cx26 deafness: mutation analy-sis and clinical variability. J Med Genet 36:829–832.
Mustapha M, Weil D, Chardenoux S, Elias S, El-Zir E, BeckmannJS, Loiselet J, Petit C. 1999. An alpha-tectorin gene defectcauses a newly identified autosomal recessive form of senso-rineural pre-lingual non-syndromic deafness, DFNB21. HumMol Genet 8:409–412.
Nadol JB Jr. 1993. Hearing loss. N Engl J Med 329:1092–1102.Prezant TR, Agapian JV, Bohlman MC, Bu X, Oztas S, Qiu WQ,
Arnos KS, Cortopassi GA, Jaber L, Rotter JI, Shohat M,Fischel-Ghodsian N. 1993. Mitochondrial ribosomal RNA mu-tation associated with both antibiotic-induced and non-syndromic deafness. Nat Genet 4:289–294.
Richard G, Smith LE, Bailey RA, Itin P, Hohl D, Epstein EH Jr,DiGiovanna JJ, Compton JG, Bale SJ. 1998a. Mutations inthe human connexin gene GJB3 cause erythrokeratodermiavariabilis. Nat Genet 20:366–369.
Richard G, White TW, Smith LE, Bailey RA, Compton JG, PaulDL, Bale SJ. 1998b. Functional defects of Cx26 resulting froma heterozygous missense mutation in a family with dominantdeaf-mutism and palmoplantar keratoderma. Hum Genet103:393–399.
Robertson NG, Lu L, Heller S, Merchant SN, Eavey RD, McKennaM, Nadol JB Jr, Miyamoto RT, Linthicum FH Jr, LubiancaNeto JF, Hudspeth AJ, Seidman CE, Morton CC, SeidmanJG. 1998. Mutations in a novel cochlear gene cause DFNA9,a human nonsyndromic deafness with vestibular dysfunction.Nat Genet 20:299–303.
Scott DA, Kraft ML, Carmi R, Ramesh A, Elbedour K, Yairi Y,Srisailapathy CR, Rosengren SS, Markham AF, Mueller RF,Lench NJ, Van Camp G, Smith RJ, Sheffield VC. 1998a. Iden-tification of mutations in the connexin 26 gene that causeautosomal recessive nonsyndromic hearing loss. Hum Mutat11:387–394.
202 RABIONET ET AL.
Scott DA, Kraft ML, Stone EM, Sheffield VC, Smith RJ. 1998b.Connexin mutations and hearing loss. Nature 391:332.
Sobe T, Erlich P, Berry A, Korostichevsky M, Vreugde S,Avraham KB, Bonne-Tamir B, Shohat M. 1999. High fre-quency of the deafness-associated 167delT mutation in theconnexin 26 (GJB2) gene in Israeli Ashkenazim. Am J MedGenet 86:499–500.
Sobe T, Vreugde S, Shahin H, Davis N, Berlin M, Kanaan M,Yaron Y, Orr-Urtreger A, Frydman M, Shohat M, AvrahamKB. 2000. The prevalence and expression of inheritedconnexin26 mutations associated with non-syndromic hear-ing loss in the Israeli population. Hum Genet 106:50–57.
Stojkovic T, Latour P, Vandenberghe A, Hurtevent JF, VermerschP. 1999. Sensorineural deafness in X-linked Charcot-Marie-Tooth disease with connexin 32 mutation (R142Q). Neu-rology 52:1010–1014.
Storm K, Willocx S, Flothmann K, Van Camp G. 1999. Deter-mination of the carrier frequency of the common GJB2(connexin-26) 35delG mutation in the Belgian populationusing an easy and reliable screening method. Hum Mutat14:263–266.
Tan CC, Ainsworth PJ, Hahn AF, Macleod PM. 1996. Novelmutations in the Connexin-32 gene associated with x-linkedCharcot-Marie-Tooth disease. Hum Mutat 7:167–171.
Tassabehji M, Read AP, Newton VE, Harris R, Balling R, GrussP, Strachan T. 1992. Waardenburg’s syndrome patients havemutations in the human homologue of the Pax-3 paired boxgene. Nature 355:635–636.
Tassabehji M, Newton VE, Read AP. 1994. Waardenburg syn-drome type 2 caused by mutations in the human mi-crophthalmia (MITF) gene. Nat Genet 8:251–255.
Vahava O, Morell R, Lynch ED, Weiss S, Kagan ME, Ahituv N,Morrow JE, Lee MK, Skvorak AB, Morton CC, BlumenfeldA, Frydman M, Friedman TB, King MC, Avraham KB. 1998.Mutation in transcription factor POU4F3 associated withinherited progressive hearing loss in humans. Science279:1950–1954.
Van Laer L, Huizing EH, Verstreken M, van Zuijlen D, WautersJG, Bossuyt PJ, Van de Heyning P, McGuirt WT, Smith RJ,Willems PJ, Legan PK, Richardson GP, Van Camp G. 1998.
Nonsyndromic hearing impairment is associated with a mu-tation in DFNA5. Nat Genet 20:194–197.
Verhoeven K, Van Laer L, Kirschhofer K, Legan PK, HughesDC, Schatteman I, Verstreken M, Van Hauwe P, Coucke P,Chen A, Smith RJ, Somers T, Offeciers FE, Van de HeyningP, Richardson GP, Wachtler F, Kimberling WJ, Willems PJ,Govaerts PJ, Van Camp G. 1998. Mutations in the humanalpha-tectorin gene cause autosomal dominant non-syndromic hearing impairment. Nat Genet 19:60–62.
Wang A, Liang Y, Fridell RA, Probst FJ, Wilcox ER, TouchmanJW, Morton CC, Morell RJ, Noben-Trauth K, Camper SA,Friedman TB. 1998. Association of unconventional myosinMYO15 mutations with human nonsyndromic deafnessDFNB3. Science 280:1447–1451.
White TW, Deans MR, Kelsell DP, Paul DL. 1998. Connexinmutations in deafness. Nature 394:630–631.
Wilcox SA, Osborn AH, Allen-Powell DR, Maw MA, Dahl HH,Gardner RJ. 1999. Connexin26 deafness in several intercon-nected families. J Med Genet 36:383–385.
Willecke K, Jungbluth S, Dahl E, Hennemann H, Heynkes R,Grzeschik KH. 1990. Six genes of the human connexin genefamily coding for gap junctional proteins are assigned to fourdifferent human chromosomes. Eur J Cell Biol 53:275–280.
Xia JH, Liu CY, Tang BS, Pan Q, Huang L, Dai HP, Zhang BR,Xie W, Hu DX, Zheng D, Shi XL, Wang DA, Xia K, Yu KP,Liao XD, Feng Y, Yang YF, Xiao JY, Xie DH, Huang JZ. 1998.Mutations in the gene encoding gap junction protein beta-3associated with autosomal dominant hearing impairment.Nat Genet 20:370–373.
Yasunaga S, Grati M, Cohen-Salmon M, El-Amraoui A,Mustapha M, Salem N, El-Zir E, Loiselet J, Petit C. 1999. Amutation in OTOF, encoding otoferlin, a FER-1-like pro-tein, causes DFNB9, a nonsyndromic form of deafness. NatGenet 21:363–369.
Zelante L, Gasparini P, Estivill X, Melchionda S, D’Agruma L,Govea N, Mila M, Monica MD, Lutfi J, Shohat M, MansfieldE, Delgrosso K, Rappaport E, Surrey S, Fortina P. 1997.Connexin26 mutations associated with the most commonform of non-syndromic neurosensory autosomal recessivedeafness (DFNB1) in Mediterraneans. Hum Mol Genet6:1605–1609.
