Saber Masmoudi

Mutations in a new gene encoding a protein of the hair bundle cause non-syndromic deafness at the DFNB16 locus.

Hearing impairment affects about 1 in 1,000 children at birth. Approximately 70 loci implicated in non-syndromic forms of deafness have been reported in humans and 24 causative genes have been identified (see also http://www.uia.ac.be/dnalab/hhh). We report a mouse transcript, isolated by a candidate deafness gene approach, that is expressed almost exclusively in the inner ear. Genomic analysis shows that the human ortholog STRC (so called owing to the name we have given its protein—stereocilin), which is located on chromosome 15q15, contains 29 exons encompassing approximately 19 kb. STRC is tandemly duplicated, with the coding sequence of the second copy interrupted by a stop codon in exon 20. We have identified two frameshift mutations and a large deletion in the copy containing 29 coding exons in two families affected by autosomal recessive non-syndromal sensorineural deafness linked to the DFNB16 locus. Stereocilin is made up of 1,809 amino acids, and contains a putative signal petide and several hydrophobic segments. Using immunohistolabeling, we demonstrate that, in the mouse inner ear, stereocilin is expressed only in the sensory hair cells and is associated with the stereocilia, the stiff microvilli forming the structure for mechanoreception of sound stimulation.

Pendred syndrome: Phenotypic variability in two families carrying the same PDS missense mutation

Pendred syndrome comprises congenital sensorineural hearing loss, thyroid goiter, and positive perchlorate discharge test. Recently, this autosomal recessive disorder was shown to be caused by mutations in the PDS gene, which encodes an anion transporter called pendrin. Molecular analysis of the PDS gene was performed in two consanguineous large families from Southern Tunisia comprising a total of 23 individuals affected with profound congenital deafness; the same missense mutation, L445W, was identified in all affected individuals. A widened vestibular aqueduct was found in all patients who underwent computed tomography (CT) scan exploration of the inner ear. In contrast, goiter was present in only 11 affected individuals, who interestingly had a normal result of the perchlorate discharge test whenever performed. The present results question the sensitivity of the perchlorate test for the diagnosis of Pendred syndrome and support the use of a molecular analysis of the PDS gene in the assessment of individuals with severe to profound congenital hearing loss associated with inner ear morphological anomaly even in the absence of a thyroid goiter.

TMC1 but not TMC2 is responsible for autosomal recessive nonsyndromic hearing impairment in Tunisian families.

Hereditary nonsyndromic hearing impairment (HI) is extremely heterogeneous. Mutations of the transmembrane channel-like gene 1 (TMC1) have been shown to cause autosomal dominant and recessive forms of nonsyndromic HI linked to the loci DFNA36 and DFNB7/B11, respectively. TMC1 is 1 member of a family of 8 genes encoding transmembrane proteins. In the mouse, MmTmc1 and MmTmc2 are both members of Tmc subfamily A and are highly and almost exclusively expressed in the cochlea. The restricted expression of Tmc2 in the cochlea and its close phylogenetic relationship to Tmc1 makes it a candidate gene for nonsyndromic HI. We analyzed 3 microsatellite markers linked to the TMC1 and TMC2 genes in 85 Tunisian families with autosomal recessive nonsyndromic HI and without mutations in the protein-coding region of the GJB2 gene. Autozygosity by descent analysis of 2 markers bordering the TMC2 gene allowed us to rule out its association with deafness within these families. However, 5 families were found to segregate deafness with 3 different alleles of marker D9S1837, located within the first intron of the TMC1 gene. By DNA sequencing of coding exons of TMC1 in affected individuals, we identified 3 homozygous mutations, c.100C→T (p.R34X), c.1165C→T (p.R389X) and the novel mutation c.1764G→A (p.W588X). We additionally tested 60 unrelated deaf Tunisian individuals for the c.100C→T mutation. We detected this mutation in a homozygous state in 2 cases. This study confirms that mutations in the TMC1 gene may be a common cause for autosomal recessive nonsyndromic HI.

A missense mutation in DCDC2 causes human recessive deafness DFNB66, likely by interfering with sensory hair cell and supporting cell cilia length regulation

Hearing loss is the most common sensory deficit in humans. We show that a point mutation in DCDC2 (DCDC2a), a member of doublecortin domain-containing protein superfamily, causes non-syndromic recessive deafness DFNB66 in a Tunisian family. Using immunofluorescence on rat inner ear neuroepithelia, DCDC2a was found to localize to the kinocilia of sensory hair cells and the primary cilia of nonsensory supporting cells. DCDC2a fluorescence is distributed along the length of the kinocilium with increased density toward the tip. DCDC2a-GFP overexpression in non-polarized COS7 cells induces the formation of long microtubule-based cytosolic cables suggesting a role in microtubule formation and stabilization. Deafness mutant DCDC2a expression in hair cells and supporting cells causes cilium structural defects, such as cilium branching, and up to a 3-fold increase in length ratios. In zebrafish, the ortholog dcdc2b was found to be essential for hair cell development, survival and function. Our results reveal DCDC2a to be a deafness gene and a player in hair cell kinocilia and supporting cell primary cilia length regulation likely via its role in microtubule formation and stabilization.

Localization of a Novel Autosomal Recessive Non-syndromic Hearing Impairment Locus DFNB63 to Chromosome 11q13.3-q13.4

Hereditary hearing impairment is the most genetically heterogeneous trait known in humans. So far, 50 published autosomal recessive non‐syndromic hearing impairment (ARNSHI) loci have been mapped, and 23 ARNSHI genes have been identified. Here, we report the mapping of a novel ARNSHI locus, DFNB63, to chromosome 11q13.3‐q13.4 in a large consanguineous Tunisian family. A maximum LOD score of 5.33 was obtained with microsatellite markers D11S916 and D11S4207. Haplotype analysis defined a 5.55 Mb critical region between microsatellite markers D11S4136 and D11S4081. DFNB63 represents the sixth ARNSHI locus mapped to chromosome 11. We positionally excluded MYO7A from being the DFNB63‐causative gene. In addition, the screening of two candidate genes, SHANK2 and KCNE3, failed to reveal any disease‐causing mutations.

Screening of the DFNB3 locus: identification of three novel mutations of MYO15A associated with hearing loss and further suggestion for two distinctive genes on this locus.

Recessive mutations of MYO15A are associated with nonsyndromic hearing loss (HL) in humans (DFNB3) and in the shaker-2 mouse. Human MYO15A has 66 exons and encodes unconventional myosin XVA. Analysis of 77 Tunisian consanguineous families segregating recessive deafness revealed evidence of linkage to microsatellite markers for DFNB3 in four families. In two families, sequencing of MYO15A led to the identification of two novel homozygous mutations: a nonsense (c.4998C>A (p.C1666X) in exon 17 and a splice site mutation in intron 54 (c.9229 + 1G>A). A novel mutation of unknown significance, c.7395 + 3G>C, was identified in the third family, and no mutation was found in the fourth family. In conclusion, we discovered three novel mutations of MYO15A, and our data suggest the possibility that there are two distinct genes at the DFNB3 locus.

High Frequency of the p.R34X Mutation in the TMC1 Gene Associated with Nonsyndromic Hearing Loss Is Due to Founder Effects

Founder mutations, particularly 35delG in the GJB2 gene, have to a large extent contributed to the high frequency of autosomal recessive nonsyndromic hearing loss (ARNSHL). Mutations in transmembrane channel-like gene 1 (TMC1) cause ARNSHL. The p.R34X mutation is the most frequent known mutation in the TMC1 gene. To study the origin of this mutation and determine whether it arose in a common ancestor, we analyzed 21 polymorphic markers spanning the TMC1 gene in 11 unrelated individuals from Algeria, Iran, Iraq, Lebanon, Pakistan, Tunisia, and Turkey who carry this mutation. In nine individuals, we observed significant linkage disequilibrium between p.R34X and five polymorphic markers within a 220 kb interval, suggesting that p.R34X arose from a common founder. We estimated the age of this mutation to be between 1075 and 1900 years, perhaps spreading along the third Hadramaout population movements during the seventh century. A second founder effect was observed in Turkish and Lebanese individuals with markers in a 920 kb interval. Screening for the TMC1 p.R34X mutation is indicated in the genetic evaluation of persons with ARNSHL from North African and Southwest Asia.

A Novel Autosomal Recessive Non-Syndromic Deafness Locus, DFNB66, Maps to Chromosome 6p21.2-22.3 in a Large Tunisian Consanguineous Family

Hereditary non-syndromic deafness is extremely heterogeneous. Autosomal recessive forms account for approximately 80% of genetic cases. Autosomal recessive non-syndromic sensorineural deafness segregating in a large consanguineous Tunisian family was mapped to chromosome 6p21.2-22.3. A maximum lod score of 5.36 at Θ = 0 was obtained for the polymorphic microsatellite marker IR2/IR4. Haplotype analysis defined a 16.5-Mb critical region between microsatellite markers D6S1602 and D6S1665. The screening of 3 candidate genes, COL11A2, BAK1 and TMHS, did not reveal any disease causing mutation, suggesting that this is a novel deafness locus, which has been named DFNB66. A search in the Human Cochlear EST Library for ESTs located in this critical interval allowed us to identify several candidates. Further investigations on these candidates are needed in order to identify the deafness-causing gene in this Tunisian family.

Mapping of a new autosomal recessive nonsyndromic hearing loss locus (DFNB32) to chromosome 1p13.3-22.1

Approximately 80% of the hereditary hearing loss is nonsyndromic. Isolated deafness is the most genetically heterogeneous trait. We have ascertained 10 individuals from a large consanguineous Tunisian family with congenital profound autosomal recessive deafness. All affected individuals are otherwise healthy. Genotype analysis excluded linkage to known recessive deafness loci in this family. Following a genome wide screening, a linkage was detected only with locus D1S206 on chromosome 1, thereby defining a novel deafness locus, DFNB32. In order to confirm linkage and for fine mapping the genetic interval, 12 individuals belonging to this family were added and 19 microsatellite markers were tested. A maximum two-point lodscore of 4.96 was obtained at a new polymorphic marker D1S21401. Haplotype analysis defined a 16 Mb critical region between D1S2868 and afmb014zb9. The interval of DFNB32 locus overlap with DFNA37 locus and the Marshall and Stickler syndromes locus. The entire coding region of COL11A1, responsible of the later syndromes, was screened and no mutation was observed. Towards the identification of the DFNB32 gene, a search on the Human Cochlear cDNA Library and EST Database was done. The genes corresponding to the ESTs found in the DFNB32 interval are being screened for deafness-causing mutations.

Analysis of GJB2 mutation : evidence for a Mediterranean ancestor for the 35delG mutation

To the Editor: Despite the fact that more than 22 genes have been identified in autosomal recessive nonsyndromal hearing loss, a single gene, GJB2, accounts for the highest proportion of the cases (http://webhost.ua.ac.be/hhh/). One mutation in this gene, 35delG, was shown to have a high allele frequency in theMediterraneanpopulation (1). The frequencies of this mutation among DFNB1 cases are 0.88 in Tunisia and 0.94 in Lebanon (2, 3). Seventy and 12 unrelated deaf individuals with autosomal recessive non-syndromic profound hearing loss, originating from Tunisia and Morocco, respectively, were screened for GJB2 mutations by PCR-mediated site-directed mutagenesis and sequencing (4, 5). We detected the 35delG mutation in the homozygous state in 16 Tunisian and six Moroccan individuals. One deaf Tunisian individual was found to be compound heterozygous for the 35delG and 291insA mutations. The fluctuation in 35delG carrier frequency among normal hearing populations and its low prevalence in some ethnic groups (6, 7) suggest one common origin for this mutation (8). Haplotype sharing in a very small chromosomal interval of individuals homozygous for this mutation indicates a founder effect in the Caucasian and the Northern European populations (9, 10). In order to test the origin of this mutation in the Mediterranean population, we screened 31 unrelated deaf individuals homozygous for the 35delG mutation (16 Tunisians, six Moroccans, and nine Lebaneses) and 116 unrelated hearing individuals who did not carry this mutation (45 Tunisians, 41 Moroccans, and 30 Lebaneses). Two microsatellite markers and one single-nucleotide polymorphism (SNP) were used for the haplotype analyses. The primers designed to amplify the SNP region were CTTTGTGTCCCGGCCCAC and CCCCTAAAGCCTCAAAACAAA. PCR products were digested with BfaI. Using the Yule’s coefficient (11), we detected significant linkage disequilibrium between the 35delG mutation and the SNP (Table 1). For the marker D13S175, allele 6 was the only allele shared by Tunisian and Moroccan patients. This allele was found only in 61.11% of Lebanese patients. Allele distribution for this marker seemed to be most consistent with a single occurrence of 35delG mutation with a recombination event occurring in the 80 kb region betweenGJB2 and D13S175 in the Lebanese population. For the most remote marker, D13S115, the 35delG mutation was found to be in significant disequilibrium with allele 7 in deaf individuals (Table 1). However, this allele was also relatively commonamonghearingcontrols.The35delGmutation was associated with one haplotype T-6-7 for SNP-D13S175-D13S115, respectively. This haplotype was rarely found among normal hearing individuals (Table 1). Our data suggest that the 35delG mutation originates from one common ancestor chromosome carrying the T allele of the SNP in populations for the three Mediterranean countries studied here. This SNP allele was also found to be associated with the 35delG mutation among deaf individuals from Belgium, Britain, and America (9). This might suggest the possibility of a single mutational event. However, we cannot rule out the possibility that 35delG independently arose multiple times on the same haplotype. Using the formula of Diaz et al. (12) and Luria – Delbruck correction (13), we have estimated the age of the 35delG mutation to be about 216 generations in the Lebanese population vs 111 generations in the Tunisian and the Moroccan populations. Because the real recombination rate is unknown, the range of error for this calculation is probably wide. Using the same formula and 112 patients from Belgium, Britain, and America, the age of the 35delG mutation was estimated to be about 500 generations (9). *These authors contributed equally to this work. Clin Genet 2005: 68: 188–189 Copyright # Blackwell Munksgaard 2005 Printed in Singapore. All rights reserved CLINICALGENETICS doi: 10.1111/j.1399-0004.2005.00474.x