|Home | About | Journals | Submit | Contact Us | Français|
A missense mutation of Gipc3 was previously reported to cause age-related hearing loss in mice. Point mutations of human GIPC3 were found in two small families, but association with hearing loss was not statistically significant. Here, we describe one frameshift and six missense mutations in GIPC3 cosegregating with DFNB72 hearing loss in six large families that support statistically significant evidence for genetic linkage. However, GIPC3 is not the only nonsyndromic hearing impairment gene in this region; no GIPC3 mutations were found in a family cosegregating hearing loss with markers of chromosome 19p. Haplotype analysis excluded GIPC3 from the obligate linkage interval in this family and defined a novel locus spanning 4.08 Mb and 104 genes. This closely linked but distinct nonsyndromic hearing loss locus was designated DFNB81.
We previously mapped a nonsyndromic recessive hearing loss locus DFNB72 to chromosome 19p13.3 based on analyses of three large consanguineous Pakistani families, PKDF335, PKDF793, and PKDF291, each independently yielding a LOD score of greater than 3.0 (Ain et al. 2007). Assuming locus homogeneity as the least complex explanation of our data, the smallest interval of shared homozygosity within these families spanned 1.16 megabases (Mb) between markers D19S216 and D19S1034, which excluded GIPC3. Recently, a missense mutation of Gipc3 was reported to be associated with age-related sensorineural hearing loss ahl5 and audiogenic seizures in mouse (Charizopoulou et al. 2011). In the same paper, hearing loss (HL) segregating in two small human families was also reported to be due to mutations of GIPC3. However, these two human families are not large enough to provide statistically significant evidence of linkage. Thus, the question remains whether mutations of GIPC3 are associated with HL in humans.
For two of the original DFNB72 families that we reported (PKDF335 and PKDF793), the linkage interval contains 204 genes including GIPC3. Additional linkage data and mutation analysis of GIPC3 now provide evidence of GIPC3 as the cause of DFNB72 HL. Here, we report seven homozygous recessive mutations of GIPC3 associated with mild to profound HL segregating in seven large consanguineous families. Our data also indicate that on chromosome 19p there is another locus for nonsyndromic HL that genetically excludes GIPC3. This locus is adjacent to but genetically distinct from DFNB72 and is designated DFNB81.
Written informed consent was obtained from participants following approval for the study from the Combined Neuroscience Institutional Review Board (IRB OH93-DC-0016) at the National Institutes of Health, Bethesda, MD, USA, the IRBs at the National Centre of Excellence in Molecular Biology, University of the Punjab, Lahore, Pakistan, Quaid-I-Azam University Islamabad, Baylor College of Medicine, and Cincinnati Children’s Hospital Research Foundation. Ascertainment and linkage analysis of families PKDF335, PKDF793, and PKDF291 was described previously (Ain et al. 2007). Families DEM4322, PKDF1048, DEM4197, and PKSR22A were ascertained through special education schools from Pakistan while family PKDF1258 contacted us on the recommendation of an audiologist.
Audiological examinations for families PKDF1258, PKDF1048, PKDF335, PKDF793, and PKDF291 were provided by audiologists in Pakistan. For families DEM4322 and DEM4197, audiology was conducted under ambient conditions. Audiograms for family PKSR22A are neither available nor are additional diagnostic assessments of the ear and temporal bone for any of the families. All audiometric data from this study were evaluated by C.C.B. at the NIDCD/NIH. Funduscopic examinations were carried out by ophthalmologists in Pakistan. Tandem gait and Rhomberg tests were performed to evaluate balance.
Genomic DNA was extracted following standard procedures from blood samples donated by members of families newly ascertained for this study. Families PKDF1258, PKDF1048, and PKSR22A were genotyped using short tandem repeat (STR) markers located in the DFNB72 interval. For family DEM4322, genotyping was performed using the Illumina HumanLinkage-12 panel which contains 6,090 SNP marker loci while for the DEM4197 pedigree, 396 fluorescently labeled STRs were genotyped. For families DEM4322 and DEM4197, genome-wide genotyping was performed at the Center for Inherited Disease Research (CIDR).
To determine if human GIPC3 mutations are associated with DFNB72 HL, we sequenced the coding region of GIPC3 in one affected subject from each of the eight families. Primers used to PCR amplify and sequence the coding exons of GIPC3 were designed using Primer3 software (http://frodo.wi.mit.edu/primer3/). DNA sequence of the coding exons of GIPC3 was determined using a BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems). The nucleotide sequence of exon 1 of GIPC3 has high GC-content (77%) and was amplified using Advantage GC Genomic LA Polymerase Mix (Clontech). Sequencing products were analyzed on an ABI3730 capillary sequencing instrument (Applied Biosystems), and the resulting sequence traces were aligned to the reference sequence using Lasergene 8 (DNASTAR).
Seventeen affected individuals in families DEM4322, PKDF1258, PKDF1048, DEM4197, PKDF335, and PKDF793 had bilateral HL that ranged from mild to profound (Fig. 1a). Bone conduction thresholds, when tested, confirmed the HL as sensorineural. Three affected individuals from family PKDF291 have bilateral, severe to profound mixed HL while the fourth affected individual has a profound sensorineural HL (Fig. 1b). Hearing loss in these families was not accompanied by obvious vestibular dysfunction, retinal dysfunction, or report of other anomalies. These data suggest that the families are segregating nonsyndromic HL although a yet unrecognized syndrome cannot be ruled out.
We previously defined a 1.16-Mb critical linkage interval for DFNB72 that extends between markers D19S216 and D19S1034 (Ain et al. 2007) (Fig. 2a, b, c). In this study, we used SNP and STR markers to establish linkage between the HL phenotype segregating in additional five Pakistani families and chromosomal region 19p13.3–p13.2. The linkage interval of family PKDF1258 refined the critical DFNB72 interval to that between markers D19S209 and D19S894 (Fig. 2c, gray shaded area). This refined interval spans 1.08 Mb and contains at least 36 genes including GIPC3. We sequenced the coding region of GIPC3 and found one homozygous frameshift and six different homozygous missense mutations in the affected individuals from seven families. The frameshift mutation identified in family PKDF335 arose as a result of a duplication of a guanine nucleotide at position 685 of the mRNA (NM_133261.2). The c.685dupG is predicted to alter the open reading frame by introducing nine missense amino acids followed by a premature stop codon (p.Ala229GlyfsX10). This may render the transcript susceptible to nonsense-mediated mRNA decay (Nicholson et al. 2010). In four DFNB72 families, we identified four homozygous transition mutations of guanine nucleotide (G) to adenine (A) at positions 136, 264, 281, and 767 of the mRNA. These four transition mutations are predicted to substitute Arg for wild-type Gly at position 46, Ile for Met at residue 88, Asp for Gly at position 94, and Asp for Gly at amino acid position 256 of the full length protein (NP_573568.1), respectively. In two DFNB72 families, we detected pyrimidine transitions of cytosine (C) to thymine (T) at positions 565 and 662 of the mRNA that are predicted to replace Arg189 and Thr221 of wild-type GIPC3 with Cys and Ile residues, respectively (Fig. 3a).
The seven homozygous mutations of GIPC3 cosegregated with HL in the corresponding families and are not present in dbSNP build 130 or in the 1000 Genomes database. We also did not find these nucleotide variants in 572–590 ethnically matched control chromosomes (Table 1). The amino acids affected by the six missense mutations of GIPC3 reported here are conserved among the three human GIPC paralogs and the orthologous genes in Xenopus and Drosophila (Katoh 2002), and among vertebrate GIPC3 genes (Fig. 3b). The functional consequences of the six missense mutations were evaluated in silico using Mutation Taster (http://www.mutationtaster.org/), SIFT (http://sift.jcvi.org/), and PolyPhen (http://genetics.bwh.harvard.edu/pph/) and were predicted to be pathogenic, disrupting GIPC3 function.
GIPC3 encodes a 312 amino acid protein that contains three predicted low complexity regions and a central conserved PDZ domain named for a motif found in the proteins PSD-95, Dlg, and ZO-1 (Katoh 2002; Saitoh et al. 2002). Two of the three low complexity regions are similar in sequence among GIPC family members and are referred to as GIPC homology domains (GH1 and GH2) (Katoh 2002). The GH2 domain of GIPC1 interacts directly with the actin-based molecular motor myosin 6, in which mutations cause HL in humans and mice (Ahmed et al. 2003; Avraham et al. 1995). The frameshift mutation identified in family PKDF335 is located in exon 4 encoding the GH2 domain (Fig. 3c). If the c.685dupG (p.Ala229Gly fsX10) transcript survives nonsense-mediated mRNA decay, it would produce a truncated GIPC3 protein that lacks 85% of the GH2 domain. With one exception (p.Arg189Cys), all of the mutations of GIPC3 identified to date are located in one of the two GH domains (Fig. 3c).
Interestingly, haplotype analyses in families PKDF1258 and PKDF1048 segregating GIPC3 mutations did not reveal shared regions of homozygosity with the HL locus identified in family PKDF291, indicating HL locus heterogeneity on chromosome 19p. Thus, there are at least two closely linked but non-overlapping DFNB loci between markers D19S886 and D19S916 (Fig. 2c). GIPC3 is located outside of the linkage interval for HL segregating in family PKDF291 (Fig. 2b, c). Nevertheless, we used genomic DNA of an affected PKDF291 family member and sequenced the six exons of GIPC3. As expected, we did not find a pathogenic variant of GIPC3. However, we did detect a heterozygous SNP (rs8113232; G>A) in exon 2 of GIPC3. The parents were homozygous for different alleles of rs8113232. Their children are obligate rs8113232 heterozygotes (Fig. 2b). Together, these data indicate that the cause of HL in family PKDF291 is unrelated to GIPC3, and a maximum multipoint LOD score of 3.35 at the marker locus D19S391 demonstrates a third genetically distinct recessive HL locus on chromosome 19p designated DFNB81.
A human genome contains deleterious mutations in excess of even H. J. Muller’s prediction (Muller 1950). Massively parallel sequencing of exomes or genomes of apparently healthy humans has revealed hundreds of heterozygous and even homozygous nonsense polymorphisms along with thousands of synonymous and missense variants (Li et al. 2010; MacArthur and Tyler-Smith 2010; Yngvadottir et al. 2009). A well-established strategy to distinguish between a benign variant and a pathogenic mutation, especially for heterogeneous disorders such as hearing loss (HL), is to genetically link the phenotype segregating in a large family to a chromosomal interval before attempting to identify a causal mutation (Cavalli-Sforza and King 1986; Dror and Avraham 2009; Friedman and Griffith 2003; Hilgert et al. 2009; Morton and Nance 2006).
The possible involvement of two mutations of GIPC3 in human nonsyndromic HL was reported, but neither of the two families segregating mutations of GIPC3 provides a sufficient number of informative meioses to produce a significant LOD score (Charizopoulou et al. 2011). One consanguineous family has two affected individuals (Charizopoulou et al. 2011) and could only yield a maximum LOD score of 1.92. In the second family of Indian origin (Charizopoulou et al. 2011), HL was previously genetically mapped to chromosome 3 and chromosome 19 with the maximum LOD score of 2.78, and both loci were designated DFNB15 (Chen et al. 1997).
Our data do provide statistically significant evidence of linkage of human HL to mutations of GIPC3 (Table 1). In addition to describing seven GIPC3 mutations as the cause of DFNB72 HL, none of which has been previously reported, we also provide evidence of an autosomal recessive HL locus closely linked to but distinct from DFNB72 (GIPC3). The linkage interval for HL segregating in family PKDF291 is bounded by markers D19S216 and D19S916 on chromosome 19p and defines an unreported locus designated as DFNB81 (Human Nomenclature Committee). This interval spans 4.08 Mb and contains 104 genes. The DFNB81 linkage interval does not include GIPC3, and it does not overlap with DFNB68, the only other reported recessive HL locus mapped to chromosome 19 with statistically significant evidence of linkage (Fig. 2c) (Santos et al. 2006). Genetically mapping three closely linked nonsyndromic HL loci on chromosome 19 is not unexpected since chromosome 19 has more than double the gene density compared to the genome-wide average (Grimwood et al. 2004). Future studies will take advantage of massively parallel sequencing technologies (Rehman et al. 2010; Walsh et al. 2010) to identify the causative recessive mutation responsible for DFNB81 HL and explore its function in the auditory system.
We thank the families who participated in this study and Andrew J. Griffith, Dennis Drayna, and Julie M. Schultz for valuable suggestions. This work was supported by grants from the National Institute on Deafness and Other Communication Disorders (NIDCD/NIH) R00-DC009287-03 to Z.M.A, from the Higher Education Commission, Islamabad to W.A., and from NIDCD/NIH DC03594 to S.M.L. Genotyping services were provided to S.M.L. by the Center for Inherited Disease Research through a fully funded federal contract from the NIH to The Johns Hopkins University, Contract Number N01-HG-65403. Work in Pakistan was also supported by the Higher Education Commission, EMRO/WHO23 COMSTECH and Ministry of Science and Technology (MoST, Lahore), and the International Center for Genetic Engineering and Biotechnology, Trieste, Italy under project CRP/PAK08-01 Contract no. 08/009 to Sh.R. Work at NIDCD/NIH was supported by intramural funds DC00039-14 to T.B.F.
Ethical standards Experiments for this study were performed in Pakistan and in the United States, and comply with the current laws of the country in which they were performed.
Conflict of interest The authors declare that they have no conflict of interest.
Atteeq U. Rehman, Laboratory of Molecular Genetics, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Rockville, MD 20850, USA.
Khitab Gul, National Centre of Excellence in Molecular Biology, Punjab University, Lahore, Pakistan.
Robert J. Morell, Laboratory of Molecular Genetics, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Rockville, MD 20850, USA.
Kwanghyuk Lee, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA.
Zubair M. Ahmed, Division of Pediatric Ophthalmology, Cincinnati Children’s Hospital Research Foundation, Cincinnati, OH 45229, USA. Division of Otolaryngology, Head and Neck Surgery, Cincinnati Children’s Hospital Research Foundation, Cincinnati, OH 45229, USA.
Saima Riazuddin, Division of Pediatric Ophthalmology, Cincinnati Children’s Hospital Research Foundation, Cincinnati, OH 45229, USA. Division of Otolaryngology, Head and Neck Surgery, Cincinnati Children’s Hospital Research Foundation, Cincinnati, OH 45229, USA.
Rana A. Ali, National Centre of Excellence in Molecular Biology, Punjab University, Lahore, Pakistan.
Mohsin Shahzad, National Centre of Excellence in Molecular Biology, Punjab University, Lahore, Pakistan.
Ateeq-ul Jaleel, National Centre of Excellence in Molecular Biology, Punjab University, Lahore, Pakistan.
Paula B. Andrade, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA.
Shaheen N. Khan, National Centre of Excellence in Molecular Biology, Punjab University, Lahore, Pakistan.
Saadullah Khan, Department of Biochemistry, Faculty of Biological Sciences, Quaid-I-Azam University, Islamabad 45320, Pakistan.
Carmen C. Brewer, Otolaryngology Branch, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Bethesda, MD 20850, USA.
Wasim Ahmad, Department of Biochemistry, Faculty of Biological Sciences, Quaid-I-Azam University, Islamabad 45320, Pakistan.
Suzanne M. Leal, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA.
Sheikh Riazuddin, Allama Iqbal Medical College/Jinnah Hospital Complex, University of Health Sciences, Lahore 54550, Pakistan.
Thomas B. Friedman, Laboratory of Molecular Genetics, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, Rockville, MD 20850, USA.