Wolframin expression induces novel ion channel activity in endoplasmic reticulum membranes and increases intracellular calcium

Wolfram syndrome is an autosomal recessive neuro-degenerative disorder associated with juvenile onset non-autoimmune diabetes mellitus and progressive optic atrophy. The disease has been attributed to mutations in the WFS1 gene, which codes for a protein predicted to possess 9-10 transmembrane segments. Little is known concerning the function of the WFS1 protein (wolframin). Endoglycosidase H digestion, immunocytochemistry, and subcellular fractionation studies all indicated that wolframin is localized to the endoplasmic reticulum in rat brain hippocampus and rat pancreatic islet beta-cells, and after ectopic expression in Xenopus oocytes. Reconstitution of wolframin from oocyte membranes into planar lipid bilayers demonstrated that the protein induced a large cation-selective ion channel that was blocked by Mg2+ or Ca2+. Inositol triphosphate was capable of activating channels in the fused bilayers that were similar to channel components induced by wolframin expression. Expression of wolframin also increased cytosolic calcium levels in oocytes. Wolframin thus appears to be important in the regulation of intracellular Ca2+ homeostasis. Disruption of this function may place cells at risk to suffer inappropriate death decisions, thus accounting for the progressive beta-cell loss and neuronal degeneration associated with the disease.     

Mutations in the WFS1 gene that cause low-frequency sensorineural hearing loss are small non-inactivating mutations

Hereditary hearing impairment is an extremely heterogeneous trait, with more than 70 identified loci. Only two of these loci are associated with an auditory phenotype that predominantly affects the low frequencies (DFNA1 and DFNA6/14). In this study, we have completed mutation screening of the WFS1 gene in eight autosomal dominant families and twelve sporadic cases in which affected persons have low-frequency sensorineural hearing impairment (LFSNHI). Mutations in this gene are known to be responsible for Wolfram syndrome or DIDMOAD (diabetes insipidus, diabetes mellitus, optic atrophy, and deafness), which is an autosomal recessive trait. We have identified seven missense mutations and a single amino acid deletion affecting conserved amino acids in six families and one sporadic case, indicating that mutations in WFS1 are a major cause of inherited but not sporadic low-frequency hearing impairment. Among the ten WFS1 mutations reported in LFSNHI, none is expected to lead to premature protein truncation, and nine cluster in the C-terminal protein domain. In contrast, 64% of the Wolfram syndrome mutations are inactivating. Our results indicate that only non-inactivating mutations in WFS1 are responsible for non-syndromic low-frequency hearing impairment.     

A PCR-RFLP assay for the A716T mutation in the WFS1 gene,a common cause of low-frequency sensorineural hearing loss

Nonsyndromic low-frequency sensorineural hearing loss (LFSNHL) is an unusual type of hearing loss that affects frequencies at 2,000 Hz and below. Recently, we reported five different heterozygous missense mutations in the Wolfram syndrome gene, WFS1, found to be responsible for LFSNHL in six families. One of the five mutations, A716T, may be a common cause of LFSNHL, as it has been reported in three families to date (Bespalova et al., 2001; Young et al., 2001). We have developed a PCR-based restriction fragment-length polymorphism (RFLP) assay to detect the A716T mutation in a simple, specific test. This method was evaluated with DNA samples from a family in which the A716T mutation was segregating with LFSNHL. This simple assay successfully detected the presence of the A716T mutation in all of the individuals predicted to be affected, based on audiologic results. Therefore, this assay can be routinely used for initial screening of the A716T mutation in patients with LFSNHL, before screening the entire coding region of the WFS1 gene.     

Endoplasmic reticulum stress and N-glycosylation modulate expression of WFS1 protein

Mutations of the WFS1 gene are responsible for two hereditary diseases, Wolfram syndrome and low frequency sensorineural hearing loss. The WFS1 protein is a glycoprotein located in the endoplasmic reticulum (ER) membrane but its function is poorly understood. Herein we show WFS1 mRNA and protein levels in pancreatic islets to be increased with ER-stress inducers, thapsigargin and dithiothreitol. Another ER-stress inducer, the N-glycosylation inhibitor tunicamycin, also raised WFS1 mRNA but not protein levels. Site-directed mutagenesis showed both Asn-663 and Asn-748 to be N-glycosylated in mouse WFS1 protein. The glycosylation-defective WFS1 protein, in which Asn-663 and Asn-748 had been substituted with aspartate, exhibited an increased protein turnover rate. Consistent with this, the WFS1 protein was more rapidly degraded in the presence of tunicamycin. These data indicate that ER-stress and N-glycosylation play important roles in WFS1 expression and stability, and also suggest regulatory roles for this protein in ER-stress induced cell death.     

Identification of two novel missense WFS1 mutations, H696Y and R703H,in patients with non-syndromic low-frequency sensorineural hearing loss

Non-syndromic low-frequency sensorineural hearing loss (LFSNHL) is an unusual type of hearing loss in which frequencies ≤2000 Hz predominantly are affected. To date, different mutations in two genes, DIAPH1 and WFSI, have been found to be associated with LFSNHL.Here, we report a five-generation Chinese family with postlingual and progressive LFSNHL. We mapped the disease locus to a 2.5 Mb region on chromosome 4p16 between markers SNP_A-2167174 and D4S431, overlapping with the DFNA6/14/38 locus. Sequencing of candidate gene revealed a heterozygous c.2086C＞T substitution in exon 8 of WFS1, leading to p. H696Y substitution at the C-terminus of Wolframin (WFS1).In addition, we performed mutational screening of WFS1 in 37 sporadic patients, 7-50 years of age, with LFSNHL. We detected a heterozygous c.2108G＞A substitution in exon 8 of WFS1, leading to p. R703H substitution in a patient. The H696 and R703 in WFS1 are highly conserved across species, including human, orangutan, rat, mouse, and frog (Xenopus). Sequence analysis demonstrated the absence of c.2086C＞T or c.2108G＞A substitutions in the WFS1 genes among 200 unrelated control subjects of Chinese background, supporting the hypothesis that they represent causative mutations, and not rare polymorphisms. Our data provide additional molecular and clinical information for establishing a better genotype-phenotype correlation for LFSNHL.     

Wolfram syndrome: new mutations, different phenotype

Background

Wolfram Syndrome (WS) is an autosomal recessive neurodegenerative disorder characterized by Diabetes Insipidus, Diabetes Mellitus, Optic Atrophy, and Deafness identified by the acronym “DIDMOAD”. The WS gene, WFS1, encodes a transmembrane protein called Wolframin, which recent evidence suggests may serve as a novel endoplasmic reticulum calcium channel in pancreatic β-cells and neurons. WS is a rare disease, with an estimated prevalence of 1/550.000 children, with a carrier frequency of 1/354.The aim of our study was to determine the genotype of WS patients in order to establish a genotype/phenotype correlation.

Methodology/Principal Findings

We clinically evaluated 9 young patients from 9 unrelated families (6 males, 3 females). Basic criteria for WS clinical diagnosis were coexistence of insulin-treated diabetes mellitus and optic atrophy occurring before 15 years of age. Genetic analysis for WFS1 was performed by direct sequencing.Molecular sequencing revealed 5 heterozygous compound and 3 homozygous mutations. All of them were located in exon 8, except one in exon 4. In one proband only an heterozygous mutation (A684V) was found. Two new variants c.2663 C>A and c.1381 A>C were detected.

Conclusions/Significance

Our study increases the spectrum of WFS1 mutations with two novel variants. The male patient carrying the compound mutation [c.1060_1062delTTC]+[c.2663 C>A] showed the most severe phenotype: diabetes mellitus, optic atrophy (visual acuity 5/10), deafness with deep auditory bilaterally 8000 Hz, diabetes insipidus associated to reduced volume of posterior pituitary and pons. He died in bed at the age of 13 years. The other patient carrying the compound mutation [c.409_424dup16]+[c.1381 A>C] showed a less severe phenotype (DM, OA).     

The WFS1 gene,responsible for low frequency sensorineural hearing loss and Wolfram syndrome,is expressed in a variety of inner ear cells

Heterozygous mutations in the WFS1 gene are responsible for autosomal dominant low frequency hearing loss at the DFNA6/14 locus, while homozygous or compound heterozygous mutations underlie Wolfram syndrome. In this study we examine expression of wolframin, the WFS1-gene product, in mouse inner ear at different developmental stages using immunohistochemistry and in situ hybridization. Both techniques showed compatible results and indicated a clear expression in different cell types of the inner ear. Although there were observable developmental differences, no differences in staining pattern or gradients of expression were observed between the basal and apical parts of the cochlea. Double immunostaining with an endoplasmic reticulum marker confirmed that wolframin localizes to this organelle. A remarkable similarity was observed between cells expressing wolframin and the presence of canalicular reticulum, a specialized form of endoplasmic reticulum. The canalicular reticulum is believed to be involved in the transcellular movements of ions, an important process in the physiology of the inner ear. Although there is nothing currently known about the function of wolframin, our results suggest that it may play a role in inner ear ion homeostasis as maintained by the canalicular reticulum.     

Mutations in IMPG2, Encoding Interphotoreceptor Matrix Proteoglycan 2, Cause Autosomal-Recessive Retinitis Pigmentosa

Retinitis pigmentosa (RP) is a heterogeneous group of inherited retinal diseases caused by progressive degeneration of the photoreceptor cells. Using autozygosity mapping, we identified two families, each with three affected siblings sharing large overlapping homozygous regions that harbored the IMPG2 gene on chromosome 3. Sequence analysis of IMPG2 in the two index cases revealed homozygous mutations cosegregating with the disease in the respective families: three affected siblings of Iraqi Jewish ancestry displayed a nonsense mutation, and a Dutch family displayed a 1.8 kb genomic deletion that removes exon 9 and results in the absence of seven amino acids in a conserved SEA domain of the IMPG2 protein. Transient transfection of COS-1 cells showed that a construct expressing the wild-type SEA domain is properly targeted to the plasma membrane, whereas the mutant lacking the seven amino acids appears to be retained in the endoplasmic reticulum. Mutation analysis in ten additional index cases that were of Dutch, Israeli, Italian, and Pakistani origin and had homozygous regions encompassing IMPG2 revealed five additional mutations; four nonsense mutations and one missense mutation affecting a highly conserved phenylalanine residue. Most patients with IMPG2 mutations showed an early-onset form of RP with progressive visual-field loss and deterioration of visual acuity. The patient with the missense mutation, however, was diagnosed with maculopathy. The IMPG2 gene encodes the interphotoreceptor matrix proteoglycan IMPG2, which is a constituent of the interphotoreceptor matrix. Our data therefore show that mutations in a structural component of the interphotoreceptor matrix can cause arRP.     

WFS1 and non-syndromic low-frequency sensorineural hearing loss: A novel mutation in a Portuguese case

Low-frequency sensorineural hearing loss (LFSNHL) is an unusual type of HL in which frequencies at 2000 Hz and below are predominantly affected. Most of the families with LFSNHL carry missense mutations in WFS1 gene, coding for wolframin.     

Mutations in the calcium-binding motifs of CDH23 and the 35delG mutation in GJB2 cause hearing loss in one family

We have ascertained a multi-generation family with apparent autosomal recessive non-syndromic childhood hearing loss (DFNB). Failure to demonstrate linkage in a genome-wide scan with 300 polymorphic markers has suggested genetic heterogeneity for the hearing loss in this family. This heterogeneity could be demonstrated by analysis of candidate loci and genes for DFNB. Patients in one branch of the family (branch C) are homozygous for the 35delG mutation in the GJB2 gene (DFNB1). Patients in two other branches (A and B) carry two new mutations in the cadherin 23 ( CDH23) gene (DFNB12). A homozygous CDH23 c.6442G-->A (D2148N) mutation is present in branch A. Patients in branch B are compound heterozygous for this mutation and the c.4021G-->A (D1341N) mutation. The substituted aspartic acid residues are highly conserved and are part of the calcium-binding sites of the extracellular cadherin (EC) domains. Molecular modeling of the mutated EC domains of CDH23 based on the structure of E-cadherin indicates that calcium-binding is impaired. In addition, other aspartic and glutamic acid residue substitutions in the highly conserved calcium-binding sites reported to cause DFNB12 are also likely to result in a decreased affinity for calcium. Since calcium provides rigidity to the elongated structure of cadherin molecules enabling homophilic lateral interaction, these mutations are likely to impair interactions of CDH23 molecules either with CDH23 or with other proteins. DFNB12 is the first human disorder that can be attributed to inherited missense mutations in the highly conserved residues of the extracellular calcium-binding domain of a cadherin.     

Identification of a novel mutation in WFS1 in a family affected by low-frequency hearing impairment

Previously we confirmed linkage of autosomal dominantly inherited low-frequency sensorineural hearing impairment (LFSNHI) in a German family to the genetic locus DFNA6/DFNA14 on chromosome 4p16.3 close to the markers D4S432 and D4S431. Analysis of data from the Human Genome Project, showed that WFS1 is located in this region. Mutations in WFS1 are known to be responsible for Wolfram syndrome (DIDMOAD, MIM #606201), which follows an autosomal recessive trait. Studies in low-frequency hearing loss families showed that mutations in WFS1 were responsible for the phenotype. In all affected family members analysed, we detected a missense mutation in WFS1 (K705N) and therefore confirm the finding that the majority of mutations responsible for LFSNHI are missense mutations which localise to the C-terminal domain of the protein.     

Comorbidity of GJB2 and WFS1 mutations in one family

It is rarely reported that two distinct genetic mutations affecting hearing have been found in one family. We report on a family exhibiting comorbid mutation of GJB2 and WFS1. A four-generation Japanese family with autosomal dominant sensorineural hearing loss was studied. In 7 of the 24 family members, audiometric evaluations and genetic analysis were performed. We detected A-to-C nucleotide transversion (c.2576G>C) in exon 8 of WFS1 that was predicted to result in an arginine-to-proline substitution at codon 859 (R859P), G-to-A transition (c.109G>A) in exon 2 of GJB2 that was predicted to result in a valine-to-isoleucine substitution at codon 37 (V37I), and C-to-T transition (c.427C>T) in exon 2 of GJB2 that was predicted to result in an arginine-to-tryptophan substitution at codon 143 (R143W). Two individuals who had heterozygosity of GJB2 mutations and heterozygosity of WFS1 mutations showed low-frequency hearing loss. One individual who had homozygosity of GJB2 mutation without WFS1 mutation had moderate, gradual high tone hearing loss. On the other hand, a moderate flat loss configuration was seen in one individual who had compound heterozygosity of GJB2 and heterozygosity of WFS1 mutations. Our results indicate that the individual who has both GJB2 and WFS1 mutations can show GJB2 phenotype.     

WFS1 protein modulates the free Ca(2+) concentration in the endoplasmic reticulum

The WFS1 gene, encoding an endoplasmic reticulum (ER) membrane glycoprotein, is mutated in Wolfram syndrome characterized by diabetes mellitus and optic atrophy. Herein, Ca(2+) dynamics were examined in WFS1-knockdown and -overexpressing HEK293 cells. Studies using ER-targeted Ca(2+)-sensitive photoprotein aequorin demonstrated WFS1 protein to positively modulate ER Ca(2+) levels by increasing the rate of Ca(2+) uptake. Furthermore, Ca(2+) imaging with Fura-2 showed the magnitude of the store-operated Ca(2+) entry to parallel WFS1 expression levels. These data indicate that WFS1 protein participates in the regulation of cellular Ca(2+) homeostasis, at least partly, by modulating the filling state of the ER Ca(2+) store.     

Identification and functional characterization of novel compound heterozygotic mutations in the TECTA gene

Mutations of the TECTA gene, which encodes alpha-tectorin, are associated with both dominant (DFNA8/A12) and recessive (DFNB 21) modes of inherited nonsyndromic sensorineural hearing loss, respectively. Although clinical data and genetic analysis for TECTA gene have been reported from different groups, there is no report that compound heterozygous mutations in the TECTA gene result in nonsyndromic sensorineural hearing loss. Here, we identified a missense mutation (p.C1691F) and a splicing mutation (c.6162 + 3insT), one in each TECTA allele, in the patient with hearing loss. Also, we demonstrated that the splicing mutation results in the abnormal skipping of an exon, which leads to a truncated protein as determined by exon-trapping analysis. To the best of our knowledge, this is the first report of an in vitro functional study of splice site mutations in the TECTA gene.     

Novel mutations in the connexin 26 gene (GJB2) that cause autosomal recessive (DFNB1) hearing loss.

Mutations in the connexin 26 (Cx26) gene (GJB2) are associated with the type of autosomal recessive nonsyndromic neurosensory deafness known as "DFNB1." Studies indicate that DFNB1 (13q11-12) causes 20% of all childhood deafness and may have a carrier rate as high as 2. 8%. This study describes the analysis of 58 multiplex families each having at least two affected children diagnosed with autosomal recessive nonsyndromic deafness. Twenty of the 58 families were observed to have mutations in both alleles of Cx26. Thirty-three of 116 chromosomes contained a 30delG allele, for a frequency of .284. This mutation was observed in 2 of 192 control chromosomes, for an estimated gene frequency of .01+/-.007. The homozygous frequency of the 30delG allele is then estimated at .0001, or 1/10,000. Given that the frequency of all childhood hearing impairment is 1/1,000 and that half of that is genetic, the specific mutation 30delG is responsible for 10% of all childhood hearing loss and for 20% of all childhood hereditary hearing loss. Six novel mutations were also observed in the affected population. The deletions detected cause frameshifts that would severely disrupt the protein structure. Three novel missense mutations, Val84Met, Val95Met, and Ser113Pro, were observed. The missense mutation 101T-->C has been reported to be a dominant allele of DFNA3, a dominant nonsyndromic hearing loss. Data further supporting the finding that this mutation does not cause dominant hearing loss are presented. This allele was found in a recessive family segregating independently from the hearing-loss phenotype and in 3 of 192 control chromosomes. These results indicate that 101T-->C is not sufficient to cause hearing loss.     

Single-amino-acid deletion in the RYR1 gene, associated with malignant hyperthermia susceptibility and unusual contraction phenotype

Malignant hyperthermia (MH) is an anesthetic-drug-induced, life-threatening hypermetabolic syndrome caused by abnormal calcium regulation in skeletal muscle. Often inherited as an autosomal dominant trait, MH has linkage to 30 different mutations in the RYR1 gene, which encodes a calcium-release-channel protein found in the sarcoplasmic reticulum membrane in skeletal muscle. All published RYR1 mutations exclusively represent single-nucleotide changes. The present report documents, in exon 44 of RYR1 in two unrelated, MH-susceptible families, a 3-bp deletion that results in deletion of a conserved glutamic acid at position 2347. This is the first deletion, in RYR1, found to be associated with MH susceptibility. MH susceptibility was confirmed among some family members by in vitro diagnostic pharmacological contracture testing of biopsied skeletal muscle. Although a single-amino-acid deletion appears to be a subtle change in the protein, the deletion of Glu2347 from RYR1 produces an unusually large electrically evoked contraction tension in MH-positive individuals, suggesting that this deletion produces an alteration in skeletal-muscle calcium regulation, even in the absence of pharmacological agents.     

Identification of a novel nonsynonymous mutation of EYA1 disrupting splice site in a Korean patient with BOR syndrome

The EYA1 gene is known as the causative gene of BOR (Branchio-oto-renal) syndrome which is a genetic disorder associated with branchial cleft cysts of fistulae, hearing loss, ear malformation, and renal anomalies. Although approximately 40 % of patients with BOR syndrome have mutations in the EYA1 gene and over 130 disease-causing mutations in EYA1 have been reported in various populations, only a few mutations have been reported in Korean families. In this study, genetic analysis of the EYA1 gene was performed in a Korean patient diagnosed with BOR syndrome and his parents. A de novo novel missense mutation, c.418G>A, located at the end of exon 6, changed glycine to serine at amino acid position 140 (p.G140S) and was suspected to affect normal splicing. Our in vitro splicing assay demonstrated that this mutation causes exon 6 skipping leading to frameshift and truncation of the protein to result in the loss of eyaHR. To the best of our knowledge, this is the first report revealing that a missense mutation in the exon disturbs normal splicing as a result of a substitution of the last nucleotide of an exon in EYA1.     

The Role of Mitochondrial DNA Mutations in Hearing Loss

Mutations in mitochondrial DNA (mtDNA) are one of the most important causes of hearing loss. Of these, the homoplasmic A1555G and C1494T mutations at the highly conserved decoding site of the 12S rRNA gene are well documented as being associated with either aminoglycoside-induced or nonsyndromic hearing loss in many families worldwide. Moreover, five mutations associated with nonsyndromic hearing loss have been identified in the tRNA^Ser(UCN) gene: A7445G, 7472insC, T7505C, T7510C, and T7511C. Other mtDNA mutations associated with deafness are mainly located in tRNA and protein-coding genes. Failures in mitochondrial tRNA metabolism or protein synthesis were observed from cybrid cells harboring these primary mutations, thereby causing the mitochondrial dysfunctions responsible for deafness. This review article provides a detailed summary of mtDNA mutations that have been reported in deafness and further discusses the molecular mechanisms of these mtDNA mutations in deafness expression.