Molecular Vision 2010; 16:26-35
Received 19 May 2009 | Accepted 6 January 2010 | Published 12 January 2010
1Department of Ophthalmology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 2Department of Otorhinolaryngology, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands; 3Department of Human Genetics, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
Correspondence to: Barend F.T. Hogewind, MD, 400 Ophthalmology, Radboud University Nijmegen Medical Centre, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands; Phone: +31-24-3614448; FAX: +31-24-3540522; email: email@example.com
Purpose: To describe the phenotype of a novel Wolframin (WFS1) mutation in a family with autosomal dominant optic neuropathy and deafness. The study is designed as a retrospective observational case series.
Methods: Seven members of a Dutch family underwent ophthalmological, otological, and genetical examinations in one institution. Fasting serum glucose was assessed in the affected family members.
Results: All affected individuals showed loss of neuroretinal rim of the optic nerve at fundoscopy with enlarged blind spots at perimetry. They showed a red-green color vision defect at color vision tests and deviations at visually evoked response tests. The audiograms of the affected individuals showed hearing loss and were relatively flat. The unaffected individuals showed no visual deviations or hearing impairment. The affected family members had no glucose intolerance. Leber hereditary optic neuropathy (LHON) mitochondrial mutations and mutations in the Optic atrophy-1 gene (OPA1) were excluded. In the affected individuals, a novel missense mutation c.2508G>C (p.Lys836Asn) in exon 8 of WFS1 was identified.
Conclusions: This study describes the phenotype of a family with autosomal dominant optic neuropathy and hearing impairment associated with a novel missense mutation in WFS1.
Hereditary optic neuropathy (HON) is a disease entity characterized by symmetric, bilateral, central visual loss with deviations of the papillomacular nerve fiber bundle resulting in cupping of the disk and in central or cecocentral scotomas and generalized constriction of the visual fields. In later stages, visual loss becomes severe, usually worse than 20/200 . HON is seen in isolated autosomal dominant optic neuropathy, in Leber hereditary optic neuropathy (LHON; OMIM 535000), in Wolfram Syndrome (OMIM 222300), and in diseases with primarily neurologic or systemic manifestations such as hereditary ataxias, hereditary polyneuropathies, hereditary spastic paraplegias, hereditary muscular dystrophies, and storage diseases .
The combination of autosomal dominant optic neuropathy and deafness has been reported in families and in isolated cases with a heterozygous missense mutation in Optic atrophy-1 gene (OPA1; OMIM 605290) [3–6]. Eiberg et al.  described a Danish family who had autosomal dominant optic neuropathy and deafness caused by a mutation in the Wolframin (WFS1) gene. WFS1, on chromosome 4p16.3, contains eight exons. Mutations in this gene are reported to be responsible for Wolfram Syndrome, Deafness Autosomal dominant type 6/14 (DFNA6/14; OMIM 600965, a low-frequency sensorineural hearing loss that is inherited in an autosomal dominant manner), psychiatric disorders, and diabetes mellitus (OMIM 606201). After examination of the family with autosomal dominant optic neuropathy and deafness, Eiberg et al.  concluded that the patients also had impaired glucose tolerance. Valéro described a French family with the same missense mutation . There were only two affected individuals: the proband and his mother suffered diabetes mellitus with congenital hearing loss. At the age of 60 the mother was diagnosed with optic atrophy. The present report describes the phenotype of a third family with autosomal dominant optic neuropathy and deafness that is associated with a novel missense mutation in WFS1.
The proband, a 57-year-old man, was referred to our tertiary referral hospital for progressive hearing loss, which coexisted with optic neuropathy. The question was raised whether he was a good candidate for cochlear implantation. Medical history showed that both his mother and brother also had optic neuropathy and hearing loss. Figure 1 shows the pedigree of this family. Informed consent was obtained from both the proband and his family to participate in this study. The research adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board (Commissie Mensgebonden Onderzoek), Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
All family members were examined in our outpatient clinic and medical history was taken. For all family members except patient II:2, the ophthalmological examinations included best corrected visual acuity measurements, slit-lamp microscopy and ophthalmoscopy. Goldmann perimetry was performed to evaluate visual field size. A morphometric analysis of the optic disc was performed using the Heidelberg Retina Tomograph II (HRT; Heidelberg Engineering, Heidelberg, Germany) . Color vision was assessed with the Hardy-Rand-Ritter (HRR) pseudoisochromatic plates, the Lanthony new color test, the Neitz anomaloscope (Neitz, Tokyo, Japan), and the standard pseudoisochromatic plates test. In addition, visually evoked potentials (VEP; Roland Consult, Brandenburg, Germany) were evaluated. Patient II:2 only underwent standard ocular examinations. All individuals underwent pure-tone audiometry and speech audiometry. Otoscopy was performed on all family members to rule out middle ear pathology. As part of a preoperative selection procedure for cochlear implantation, the proband also underwent electronystagmography, computed tomography (CT), and auditory steady-state response (ASSR) testing.
The affected family members underwent fasting serum testing to exclude diabetes mellitus (serum specific insulin, serum C-peptide and HbA1C were analyzed in the proband and plasma glucose was analyzed in all affected individuals).
Blood samples of all living individuals were collected in EDTA tubes and kept at room temperature. DNA was isolated within five days after withdrawal on a Chemagen MSM1 platform using the Chemagic DNA blood 10k kit (Chemagen, Baesweiler, Germany). Mutation analysis of WFS1 was performed by direct sequencing of the entire coding region (exon 2 to 8). The coding exons and the flanking intronic sequences were PCR amplified and subsequently sequenced on a 3730 automated sequencer using Dye terminator chemistry (Applied Biosystems, Foster City, CA). For primer information and PCR conditions see Table 1. In addition, OPA1 and three known LHON mutations (mtDNA positions m.11778, m.3460, and m.14484) were screened by a combination of dHPLC (Transgenomic, Inc., Omaha, NE) and direct sequencing analysis.
From the pedigree in Figure 1, it can be concluded that the disease has an autosomal dominant or mitochondrial inheritance pattern. Neither the maternal grandparents nor the uncle of the proband had hearing impairment. All affected individuals had no symptoms additional to progressive sensorineural hearing impairment and optic atrophy. No other Wolfram syndrome-related symptoms (diabetes mellitus, diabetes insipidus, renal or psychiatric problems) were mentioned during each participant’s history and physical examination.
The results of the ophthalmological examinations are shown in Table 2 and Figure 2. It should be noted that not all the family members were examined ophthalmologically and that the unaffected individuals that were examined ophthalmologically were of a younger generation than the affected individuals. All affected individuals had an enlarged blind spot at Goldmann perimetry, loss of neuroretinal rim on HRT, and a deviating VEP. They also showed an indication of protan and deutan axes (i.e., a red-green defect) at color vision testing. The unaffected individuals did not have an enlarged blind spot at Goldmann perimetry, loss of neuroretinal rim on HRT, nor a deviating VEP.
The results of otological and audiological examinations are shown in Table 3. Otoscopy showed no abnormalities in all patients. Hearing was normal in both children (IV:1 and IV:2) of the proband and also normal (for his age) in the only living uncle (II:3) of the proband. For all affected individuals, the first available and last-visit pure-tone audiograms are shown in Figure 3, upper panel. They demonstrate a relatively flat-type of hearing loss at last visit. Remarkably, hearing in both the proband and his mother is so diminished that they hardly have any speech recognition. Individual III:3, the youngest of these three affected participants, still has relatively normal maximum speech recognition scores. His first available pure-tone audiogram demonstrated a typical low-frequency sensorineural hearing loss. No progression of hearing loss could be deduced from the pure-tone audiograms of study participants II:2 and III:2.
The proband had normal vestibular function. CT scan of the temporal bone showed no anatomic abnormalities. Central causes of sensorineural hearing loss were excluded by ASSR testing, which revealed normal function of the auditory nerve. The proband underwent successful right ear cochlear implantation with a 22-electrode implant (Cochlear; Nucleus Freedom, Sidney, Australia). Seven months after implantation, he was found to have 83% speech recognition at 70 dB sound pressure level (SPL). Prior to implantation, aided thresholds with a conventional aid on the right were tested: no speech recognition was found at 70 dB SPL and the maximum speech recognition of about 20% was found at 80 dB SPL.
Screening of mtDNA positions m.11778, m.3460 and m.14484 demonstrated that the three most frequent LHON mutations were not present. Screening of WFS1 and OPA1 revealed no mutations in all coding regions of OPA1. In exon 8 of WFS1, a heterozygous mutation was identified in all three affected patients. This variant was not detected in the unaffected family members. At position 2508, a substitution of a cytosine for a guanine (c.2508G>C) leads to the amino acid substitution p.Lys836Asn. Based on the medical history and clinical information available, we assume that this is a de novo mutation. Figure 4 shows the normal and mutated sequences. We believe this variant to be pathogenic because the mutated lysine is evolutionarily highly conserved (Figure 5), is located in a conserved region of the protein, and cosegregates with the disease in this family. In addition, this variant has never been identified in our laboratory (so far we analyzed the WFS1 sequence in 200 European patient chromosomes).
Diabetes mellitus was excluded in the affected individuals by assessing fasting serum glucose (all within the reference value 4.0–5.6 mmol/l) and HbA1C (all within the reference value 4.2%–6.3%). In the proband, insulin (9 mE/l, reference value 8–20) and C-peptide (0.62 mmol/l, reference value 0.17–1) were assessed as well, and found to be in the normal range.
In this article we report a novel missense mutation (p.Lys836Asn) in exon 8 of WFS1 that is associated with autosomal dominant optic neuropathy and deafness. The disease shows 100% segregation with the mutation. The three affected individuals did not have diabetes mellitus or any other symptoms. Since there was no autosomal recessive inheritance, they did not present with Wolfram syndrome spectrum. They also did not have DFNA6/14, the nonsyndromic autosomal dominant low frequency sensorineural hearing loss that is caused by WFS1 mutations. The affected participants did show more severe sensorineural hearing loss with involvement of all frequencies, and they did have optic atrophy.
HON is a clinically and genetically heterogeneous condition. Isolated autosomal dominant optic neuropathy is caused by four genes: OPA1, OPA3, OPA4, and OPA5. OPA1 (OMIM 165500) is the most frequently encountered type of isolated autosomal dominant optic neuropathy. Symptoms usually manifest in the first decade of life, and, in most cases, visual impairment develops gradually over many years. The reduction in visual acuity tends to be mild or moderate [10,11]. Visual field defects mainly involve the central portion of the field and include cecocentral scotomas, paracentral defects, and pseudobitemporal defects. Neuroretinal rim pallor in autosomal dominant optic neuropathy is most pronounced temporally, but usually involves the entire optic disc .
Optic neuropathy is also a sequela of LHON. LHON is the result of maternal mitochondrial DNA mutations. Visual loss usually begins painlessly and centrally in one eye, and the second eye is affected weeks to months later with an acute or subacute course. However, the duration of progression of visual loss in each eye varies and may be difficult to document accurately. Eventually, optic atrophy bundle supervenes .
Optic neuropathy is also one of the main symptoms in Wolfram syndrome. Caused by mutations in WFS1 on chromosome 4p16.3, Wolfram syndrome is an autosomal recessive neurodegenerative syndrome . The minimal criteria for diagnosis are diabetes mellitus and optic atrophy . Sensorineural hearing loss is often an additional finding.
The affected individuals of the described family have severe optic nerve damage as well as sensorineural hearing loss that is inherited in an autosomal dominant way. In 1977, Deutman presented a family with autosomal dominant optic neuropathy and deafness similar to our family: the affected individuals had sensorineural hearing impairment and optic atrophy but no other symptoms. All affected individuals had hearing loss with a relatively flat audiogram, an enlarged blind spot at Goldmann perimetry, deutanomaly, and a deviating VER . There was no information available on the genetic background of this family.
Autosomal dominant optic neuropathy and deafness has also been reported in several families and in isolated cases with a heterozygous missense mutation (p.Arg445His or p.Gly439Val) in OPA1 [3–5,14]. In our family, the presentation of the optic atrophy and the bilateral progressive sensorineural hearing loss is similar to the presentation of the families with autosomal dominant optic neuropathy and deafness and a mutation in OPA1. However, OPA1 mutations were excluded by sequencing analysis. In addition, the OPA1 mutations seem to cause a broader phenotype with ptosis, ophthalmoplegia, ataxia, axonal sensory-motor polyneuropathy, and mitochondrial myopathy [5,14,15]. Another OPA1 mutation (p.Tyr582Cys) is responsible for progressive hearing loss that necessitated cochlear implantation, macrocytic anemia, and hypogonadism . Interestingly, the patient of this case had progressive external ophthalmoplegia and central vision loss. Because no optic pallor was seen on fundoscopy, it was believed the patient’s vision loss was not caused by optic atrophy.
The combination of hearing loss and LHON has not been ascribed to a single mutation so far . However, Hofmann and coworkers  proposed that LHON mutations represent a susceptibility factor for Wolfram syndrome which, by interaction with further exogeneous or genetic factors, might increase the risk for disease. In our family, the main LHON mutations were excluded by sequencing analysis. The phenotype of LHON mutations is characterized by an acute or subacute loss of visual acuity with changes of the optic disc. None of the affected family members suffered such a period.
WFS1 encodes for wolframin, a protein known to contain nine predicted transmembrane domains. So far, about 110 mutations in WFS1 are believed to cause the Wolfram syndrome [19–23]. Eiberg  and Valéro  reported two families who had a WFS1 missense mutation (p.Gln864Lys) in the same conserved region as the mutation that we found. This mutation caused an autosomal dominant clinical triad: congenital hearing impairment, diabetes mellitus, and optic atrophy. In the study by Valéro, however, the proband (thus far) had no optic atrophy . Mutations in WFS1 are also responsible for other conditions such as psychiatric disorders and diabetes mellitus (Lesperance laboratory database). The pleiotropy of this disorder can possibly be explained by alternative splicing: missense mutations would occur in regions that are spliced out in specific organs. Challenging WFS1 expression studies of these organs would be needed to prove or reject this hypothesis.
To date, 26 WFS1 mutations have been reported to cause DFNA6/14 [21–24], an unusual type of hearing loss that affects frequencies at 2,000 Hz and below [25–27]. In general, hearing loss in DFNA6/14 is not progressive, however, some families were reported to have progressive hearing loss that could be attributed to presbycusis [28,29]. Because high frequency hearing is generally preserved, DFNA6/14 patients retain excellent understanding of speech, although presbycusis may cause high-frequency hearing loss later in life. Consequently, DFNA6/14 in younger patients is often asymptomatic, and many patients choose not to wear hearing aids. This contrasts with the affected family members in the current family who showed a flat-type of hearing loss with poor speech recognition.
Due to a lack of previous audiograms, it was difficult to evaluate progression of hearing loss in II:2 and III:2. It should however be noted that the first available audiograms of II:2 were made at the age of 60 years and that the first available audiograms of III:2 were made at the age of 60 years. Progression of their hearing loss may have occurred before these ages. Interestingly, the first pure-tone audiogram of III:3, done at the age of 38 years, shows a more pronounced loss at the lower frequencies, resembling the hearing loss characteristic of DFNA6/14 (see Figure 3, upper and lower panel).
Individual III:3 typically had low-frequency sensorineural hearing loss that resembled DFNA6/14 at the age of 38 years. In the following years, he experienced progression of his hearing loss at 8 kHz (40 dB) that appears to be bigger than the mean deterioration (approximately 1 dB/year) that was reported for this frequency based on analyses in several families with DFNA6/14 . The hearing impairment in his brother, III:2, and his mother, II:2, is too profound for DFNA6/14. Thus it can be concluded that the phenotype of this novel WFS1 mutation is more severe when compared to DFNA6/14. Hearing appears to deteriorate more progressively for the WFS1 mutation and these patients also develop optic neuropathy.
The distribution of wolframin in the mammalian visual system, and the pathogenesis of optic atrophy due to mutations in WFS1 remain unclear. Expression studies, however, have assessed the presence of wolframin in retinal ganglion cells and optic nerve glia cells of the cynomolgus monkey . In rodents, the presence of mRNA and wolframin have been examined in the retina (amacrine cells, Müller cells, photoreceptors, horizontal cells, bipolar cells and retinal ganglion cells), in the optic nerve (particularly in astrocytes), in the optic tract, and in the brain (the superior colliculus, the dorsomedial part of the suprachiasmatic nucleus and layer II of the primary and secondary visual cortices) . Kawano and coworkers hypothesized that mutant wolframin may contribute to the dysfunction of wolframin-expressing neurons as well as glial cells, which, in turn, may lead to optic neuropathy .
The function of wolframin in the inner ear and the mechanisms by which missense mutations cause hearing loss have not been extensively explored . The expression of wolframin has been localized to the mouse cochlea at different developmental stages and is widely distributed in different cochlear cell types, including inner and outer hair cells, a variety of supporting cells, and cells of the lateral wall, spiral ganglion, and vestibule . A similarity has been observed between wolframin expression and the presence of the canalicular reticulum, a specialized form of endoplasmic reticulum that is believed to be involved in transcellular ion transport . Thus, wolframin may be involved in regulation of inner ear ion homeostasis as maintained by the canalicular reticulum [33,34]. The majority of causative DFNA6/14 mutations have been identified in exon 8, which contains the conserved C-terminal domain. This domain seems to have a crucial function in the cochlea [22,33], and the p.Lys836Asn mutation is also located in this domain. Because the proband of our study family highly benefits from his cochlear implant, this suggests that indeed there is a deleterious effect of the present WFS1 mutation in the cochlea without any neurodegenerative symptoms that are so common in Wolfram syndrome. It is still not clear why DFNA6/14 patients show a stable low frequency hearing impairment with minor progression and why the affected family members in our study have such severe progression and profound hearing impairment. Interestingly, in our study family with visual impairment by optic neuropathy, there were no signs that the hearing impairment is caused by auditory neuropathy. According to the literature , most patients with Wolfram syndrome do not have auditory neuropathy, but most have sensorineural hearing loss caused by degeneration of the organ of Corti. In addition, patients with missense mutations in WFS1, causing DFNA6/14, also have sensorineural hearing loss and no auditory neuropathy. Apparently, this specific missense mutation in WFS1 has a different effect in the inner ear than in the eye. Further studies are needed to elucidate the underlying pathogenetic mechanism .
The results of our study suggest that p.Lys836Asn is a novel mutation in WFS1 that is associated with autosomal dominant optic neuropathy and finally a severe to profound hearing loss with relatively flat audiograms. The present phenotype is similar to the phenotype caused by the only other reported WFS1 missense mutation (p.Gln864Lys) that causes autosomal dominant optic neuropathy and deafness. However, some of those patients also have impaired glucose intolerance , and one patient has no optic neuropathy . It is also similar to the phenotype of OPA1 mutations causing autosomal dominant optic neuropathy and deafness, however, in general these mutations cause a more extensive phenotype that includes ptosis, ophthalmoplegia, ataxia, axonal sensory-motor polyneuropathy, mitochondrial myopathy, macrocytic anemia, and hypogonadism [6,14–16]. The symptoms in our study family also clearly differ from the clinical presentation of LHON, Wolfram syndrome, and DFNA6/14. On the basis of this study we advise to perform extensive genetic testing of at least WFS1 and OPA1 in cases of autosomal dominant optic neuropathy and deafness.
We thank the family for their participation in the study. The authors do not have any financial interest in this study.