Introduction

Nonsyndromic forms of deafness transmitted as a recessive trait are the most frequent cause of hereditary hearing loss and often exhibit the most severe hearing phenotype (Cohen et al. 1995). Nonsyndromic autosomal recessive deafness is usually clinically homogeneous and is nonprogressive in nature (Van Camp et al. 1997). As expected from the structural and functional complexity of the inner ear, sensorineural deafness exhibits a high degree of genetic heterogeneity. It has been estimated that at least 1% of human protein-coding genes are involved in the hearing process (Friedman and Griffith 2003). Presently, there are 36 published recessive deafness loci (DFNB), and 20 of the corresponding genes have been identified (Hereditary Hearing Loss Homepage, http://www.uia.ac.be/dnalab/hhh/).

Based on our ongoing genetic studies, a large repertoire of human genes associated with deafness remains to be mapped and identified, and these genes can be used to elucidate the molecular bases of hearing impairment. Because of the high degree of consanguinity, the Pakistani population represents a genetic resource for the identification of these novel deafness loci/genes (Hussain and Bittles 1998). Consanguineous families have contributed significantly to the identification of mutated genes associated with hearing loss (Friedman and Griffith 2003). This communication, describes two such large families segregating recessively inherited, profound congenital deafness, which is linked to markers that lie on chromosome 5q12.3-q14.1 and that define a novel locus, DFNB49 (Fig. 1).

Fig. 1
figure 1

Chromosome 5 markers co-segregating with deafness in families PKDF041 and PKDF141 define DFNB49 (black symbols individuals are deaf). The ‘at risk’ haplotypes are boxed in all generations. The genetic linkage markers and their relative positions in centimorgans (cM) according to the Marshfield map are shown left of each pedigree. a PKDF041; haplotypes showing a region of homozygosity flanked by markers D5S427 (69.23 cM) and D5S2110 (135.25 cM). b PKDF141; haplotypes showing a refined region of 11 cM delimited by markers D5S647 (74.07 cM) and D5S1501 (85.25 cM). Normal hearing individual IV:5 provided the proximal breakpoint at marker D5S647 (74.07 cM). The distal recombination is provided by individuals IV:6, IV:7, V:3, V:4, and V:5 at marker D5S1501 (85.25 cM).

Full size image

Materials and methods

Family enrollment

Approval for this study was obtained from the IRB at the National Centre of Excellence in Molecular Biology, Lahore, Pakistan (FWA00001758) and the NINDS/NIDCD IRB at the National Institutes of Health, USA (OH-93-N-016). Signed informed consent was obtained from all participants. Genomic DNA was extracted from 10 ml blood samples by using a standard protocol (Grimberg et al. 1989).

Clinical evaluation

Clinical histories were obtained for all of the individuals of families PKDF041 and PKDF141. Physical evaluations were performed to rule out any obvious extra auditory phenotype. Pure tone audiometry tests for air and bone conduction were performed at frequencies from 250 to 8,000 Hz (Fig. 2). Ocular funduscopy and electroretinography (ERG) was performed to detect the presence of retinopathy. Vestibular function was evaluated by testing tandem gait ability and by using the Romberg test.

Fig. 2
figure 2

Pure tone air (right and left ear) and bone conduction (right ear) audiograms (O, X right and left ear, respectively). a Normal hearing individual VI:2 (9 years) and affected individual V:5 (42 years) of PKDF041. b Normal hearing individual IV:5 (35 years) and affected individual V:5 (22 years) of PKDF141. Unaffected individuals have normal hearing, whereas affected individuals are profoundly deaf with hearing thresholds at or above 90 dB at all tested frequencies.

Full size image

Linkage analysis

Linkage to known DFNB loci was excluded for families PKDF041 and PKDF141 by using at least three microsatellite markers for each of the reported deafness loci (Hereditary Hearing Loss Homepage, http://www.uia.ac.be/dnalab/hhh/). Consequently, a genome-wide scan was performed with 388 fluorescent dye-labeled microsatellite markers from ABI Prism Linkage Mapping Set, Version 2.5 (Applied Biosystems) spaced at an average of 10 cM across the human genome. The microsatellite markers were amplified by the polymerase chain reaction (PCR) on a Gene Amp PCR system 9700 (Perkin Elmer) and were analyzed on an ABI Prism 3100 Genetic Analyzer. The alleles were assigned by using Genescan and Genotyper Software (Applied Biosystems).

Lod score calculations

Lod scores were calculated by using MLINK of the FASTLINK computer package (Schaffer 1996). The disease was coded as fully penetrant, and the disease-allele frequency was set at 0.001. Meiotic recombination frequencies were considered to be equal for males and females. Allele frequencies for microsatellite markers were calculated by genotyping 90 randomly selected unaffected individuals from the same population.

Mutational analysis

Primers for PCR amplification and subsequent sequencing of SLC30A5 and SLC12A2 were designed by using software at the Primer3 Web site (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) to flank all the exon–intron boundaries. PCR amplification, sequencing reactions, and mutation analysis were performed as described elsewhere (Ahmed et al. 2001). The sequence of primers for SLC30A5 and SLC12A2 will be provided upon request.

Nomenclature

DFNB49 was assigned to the new locus reported herein by the Human Genome Organization (HUGO) committee, and all gene symbols follow the recommendations of the HUGO Gene Nomenclature Committee (Povey et al. 2001).

Results

Regardless of age, all affected individuals in families PKDF041 and PKDF141 displayed congenital bilateral profound sensorineural hearing loss, whereas characteristic audiograms of unaffected individuals of both families showed normal hearing (Fig. 2). The unaffected individual of family PKDF141 (IV:5) had a mild conductive hearing loss at lower frequencies. Audiometric evaluations of two other normal hearing individuals IV:3 and IV:4 (data not shown) also showed the same hearing profile, which was different from the profound sensorineural hearing loss observed in the affected individuals. Vestibular function was normal in affected individuals from both families, and clinical evaluations suggested no features of ophthalmologic, skin, or renal anomalies. Funduscopy examinations of affected individuals of PKDF041, viz., V:5 (42 years), VI:3 (16 years), VI:4 (10 years), VI:5 (20 years), and of PKDF141, viz., V:5 (22 years), showed no signs of retinitis pigmentosa (RP). Moreover, ERGs were normal for affected members of PKDF041, viz., V:5 (42 years) and V:7 (30 years), thus excluding a novel Usher syndrome gene segregating in these two families.

The deafness phenotype segregating in families PKDF041 and PKDF141 was not found to be linked to the published deafness loci, thus indicating the involvement of at least one new locus. For deafness segregating in family PKDF041, evidence of linkage was detected on chromosome 5q12.1-q23.3 following a genome-wide scan. Additional markers were genotyped in the DFNB49 interval for all participating members. Haplotype analysis revealed a region of homozygosity of approximately 66 cM (Fig. 1a) flanked by markers D5S427 (69.23 cM) and D5S2110 (135.25 cM). A maximum two-point lod score (Zmax) of 4.44 at recombination fraction θ=0 was obtained for marker D5S2055 (Table 1).

Table 1 Two-point lod scores
Full size table

Genome-wide linkage analysis of family PKDF141 segregating nonsyndromic deafness also showed linkage to markers at chromosome 5q12.3-q14.1, and haplotype analysis refined the linked region to 11 cM (Fig. 1b) delimited by markers D5S647 (74.07 cM) and D5S1501 (85.25 cM). Individual IV:5 of PKDF141 had normal hearing and was homozygous for markers in the proximal portion of the linked haplotype, with a recombination at D5S647 defining the proximal breakpoint of the linked region at 74.07 cM. Individuals IV:6, IV:7, V:3, V:4, and V:5 provided the distal breakpoint of the critical linked region at D5S1501. The parents of affected individuals V:6 and V:7 were unrelated. The deafness phenotype in these individuals may have been the result of compound heterozygosity for allelic mutations. A Zmax of 5.94 at θ=0 was found for D5S424 (Table 1).

Discussion

The region of homozygosity on 5q12.3-q14.1 for affected individuals of family PKDF041 overlaps with the reported interval for Usher syndrome (USH2C; Pieke-Dahl et al. 2000). Usher syndrome is an autosomal recessive disorder characterized by sensorineural hearing loss, RP and variable vestibular dysfunction (Ahmed et al. 2003b). Mutations of the gene encoding VLGR1 (MIM 602851), the very large G-protein-coupled receptor-1, have recently been reported in affected individuals of several USH2C-linked families (Weston et al. 2004). Although affected members of families PKDF041 do not have RP, VLGR1 cannot be excluded as a candidate gene. Mutant alleles of several Usher syndrome genes are reported to be associated with nonsyndromic hearing loss (Weil et al. 1997). We have previously shown that mutations of CDH23, USH1C, and PCDH15 can cause nonsyndromic hearing loss (Bork et al. 2001; Ahmed et al. 2002, 2003a).

Assuming the deafness segregating in these two families is caused by allelic mutations, the linkage region for DFNB49 is 11 cM in length and is delimited by D5S647 (74.07 cM) and D5S1501 (85.25 cM), which excludes VLGR1 as a candidate gene. Family PKDF141 delineates this smallest linkage region, which contains a number of candidate genes (Fig. 3). As the haplotypes of both families differ from each other, most likely the two mutant alleles are not the same.

Fig. 3
figure 3

Deafness segregating in families PKDF041 and PKDF141 show linkage to a region of human chromosome 5q12.3-q14.1. Candidate genes for DFNB49 deafness are derived from data available on the UCSC Genome Browser (http://genome.ucsc.edu/cgi-bin/hgGateway, July 2003 assembly).

Full size image

Three deafness loci (DFNA42, DFNA1, and DFNA15) have been previously mapped to chromosome 5q, but their genetic intervals do not overlap with the larger interval for DFNB49 defined by family PKDF041 (Xia et al. 2002; Leon et al. 1992; Vahava et al. 1998).

One of the plausible candidate genes in the region of DFNB49 locus is solute carrier family member SLC30A5. Mutations in a solute carrier family member, SLC26A4, are responsible for DFNB4 (Li et al. 1998), Pendred syndrome (Everett et al. 1997), and Enlarged Vestibular Aqueduct syndrome (Usami et al. 1999). SLC26A4 mutations account for approximately 10% of hereditary deafness in diverse populations that include eastern and southern Asians (Park et al. 2003). However, all 16 protein-coding exons of SLC30A5 have been sequenced by using DNA samples from two affected individuals of families PKDF041and PKDF141, and no mutation has been found.

Other interesting candidate genes within the region of the DFNB49 locus include OCLN, GTF2H2, and BTF3, encoding the tight-junction protein occludin, RNA polymerase II transcription initiation factor, and basic transcription factor 3, respectively. Tight-junction proteins contribute to ion-selective paracellular barriers, and mutant alleles of the claudin 14 gene CLDN14 are responsible for nonsyndromic deafness in DFNB29 linked families (Wilcox et al. 2001; Ben-Yosef et al. 2003). OCLN is, therefore, a good candidate for the deafness gene segregating in our two families. The occludin knock-out mouse has not been reported to have hearing loss, but its hearing status does not seem to have been evaluated (Saitou et al. 2000). Genes encoding transcription factors EYA4, POU3F4, POU4F3, and TFCP2L3 are reported to be mutated in DFNA10 (Wayne et al. 2001), DFN3 (De Kok et al. 1995), DFNA15 (Vahava et al. 1998), and DFNA28 (Peters et al. 2002), respectively. TFCP2L3 is expressed in the epithelial cells of many tissues, and yet the only phenotype of a dominant mutant allele of TFCP2L3 is progressive loss of hearing (Peters et al. 2002).

An alternative to a one-locus hypothesis explaining the segregation of deafness in our two families is that there are two different, closely linked, deafness loci. In the linkage interval defined by family PKDF041, there are more than 100 genes including KCNN2, a member of the potassium channel genes. KCNN2 is an attractive candidate, because KCNQ4, another potassium channel gene, is mutated in dominant deafness DFNA2 (Kubisch et al. 1999). However, SLC12A2 is perhaps the most interesting candidate gene in this interval, because there is a deaf mouse model for SLC12A2 (Delpire et al. 1999). The 27 protein-coding exons of SLC12A2 have been sequenced by using DNA from affected individuals of family PKDF041, but no mutation has been found.

Rather than continuing to screen the two families for mutations that define DFNB49, we propose that it would be more efficient to identify additional families segregating recessive hearing loss mapping to this interval and to use these families to refine the critical region(s). Localization of locus DFNB49 is a first step toward the identification of a novel gene that is necessary for the development and/or maintenance of normal hearing.

Electronic database information

URLs for data in this article are as follows:

Center for Medical Genetics, Marshfield Medical Research Foundation: http://research.marshfieldclinic.org/genetics/

Hereditary Hearing Loss Homepage: http://www.uia.ac.be/dnalab/hhh/

Online Mendelian Inheritance in Man (OMIM): http://www.ncbi.nlm.nih.gov/entrez/OMIM

Primer3 Web-Based Server (primer3_www.cgi v 0.2): http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi

The Human Genome Organization (HUGO): http://www.gene.ucl.ac.uk/hugo/

UCSC Genome Bioinformatics: http://genome.ucsc.edu/