Article Text

Original article
A homozygous nonsense CEP250 mutation combined with a heterozygous nonsense C2orf71 mutation is associated with atypical Usher syndrome
  1. Samer Khateb1,
  2. Lina Zelinger1,
  3. Liliana Mizrahi-Meissonnier1,
  4. Carmen Ayuso2,
  5. Robert K Koenekoop3,
  6. Uri Laxer4,
  7. Menachem Gross5,
  8. Eyal Banin1,
  9. Dror Sharon1
  1. 1Department of Ophthalmology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
  2. 2Department of Genetics, Instituto de Investigacion Sanitaria—Fundacion Jimenez Diaz (IIS-FJD), CIBERER, ISCIII, Madrid, Spain
  3. 3Departments of Human Genetics, Paediatric Surgery and Ophthalmology, McGill University Health Centre, Montreal, Quebec, Canada
  4. 4Department of Pulmonology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
  5. 5Department of Otolaryngology, Head and Neck Surgery, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
  1. Correspondence to Professor Dror Sharon, Department of Ophthalmology, Hadassah-Hebrew University Medical Center, POB 12000, Jerusalem 91120, Israel; dror.sharon1{at}gmail.com

Abstract

Background Usher syndrome (USH) is a heterogeneous group of inherited retinitis pigmentosa (RP) and sensorineural hearing loss (SNHL) caused by mutations in at least 12 genes. Our aim is to identify additional USH-related genes.

Methods Clinical examination included visual acuity test, funduscopy and electroretinography. Genetic analysis included homozygosity mapping and whole exome sequencing (WES).

Results A combination of homozygosity mapping and WES in a large consanguineous family of Iranian Jewish origin revealed nonsense mutations in two ciliary genes: c.3289C>T (p.Q1097*) in C2orf71 and c.3463C>T (p.R1155*) in centrosome-associated protein CEP250 (C-Nap1). The latter has not been associated with any inherited disease and the c.3463C>T mutation was absent in control chromosomes. Patients who were double homozygotes had SNHL accompanied by early-onset and severe RP, while patients who were homozygous for the CEP250 mutation and carried a single mutant C2orf71 allele had SNHL with mild retinal degeneration. No ciliary structural abnormalities in the respiratory system were evident by electron microscopy analysis. CEP250 expression analysis of the mutant allele revealed the generation of a truncated protein lacking the NEK2-phosphorylation region.

Conclusions A homozygous nonsense CEP250 mutation, in combination with a heterozygous C2orf71 nonsense mutation, causes an atypical form of USH, characterised by early-onset SNHL and a relatively mild RP. The severe retinal involvement in the double homozygotes indicates an additive effect caused by nonsense mutations in genes encoding ciliary proteins.

Keywords
  • Cilia
  • Double homozygote
  • Usher syndrome

Statistics from Altmetric.com

Keywords

Introduction

Retinitis pigmentosa (RP) mendelian inheritance in man (MIM #268000) is the most common inherited retinal degeneration with an estimated worldwide prevalence of 1:4500.13 The disease is highly heterogeneous, both clinically and genetically, and has several patterns of inheritance. At present, 35 genetic loci/genes have been implicated in non-syndromic autosomal recessive RP (ARRP), most of which account for a few per cent of RP cases each. About 20%–30% of patients with RP suffer from associated additional phenotypes (eg, syndromic RP) with Usher syndrome (USH) being the most common one.4 Patients with USH suffer from typical RP accompanied by sensorineural hearing loss (SNHL) of variable severity: congenital SNHL with vestibular areflexia in Usher type 1 (USH1), mild to moderate non-progressive SNHL in USH2, and progressive hearing loss, sometimes accompanied by vestibular involvement, with a variable time of onset in USH3. In addition, atypical USH has been reported in a few patients.57 USH is a heterogeneous AR syndrome and can be caused by mutations in each of 12 different genes (RETNET at https://sph.uth.edu/Retnet/sum-dis.htm). Some of the genes in which mutations cause USH encode ciliary proteins, a cellular structure that is important for the function of both retinal photoreceptors and cochlear hair cells of the inner ear.8 ,9

The wide use of next generation sequencing in the past few years has allowed geneticists an efficient route towards the identification of the majority of disease-causing mutations and genes.10 In particular, the combination of homozygosity mapping and whole exome sequencing (WES) has been highly efficient in the identification of novel retinal disease-causing genes.1114 In addition, WES allows for an insight into the genotype of a large number of genes and identification of potential modifier genes.

In our current study, we used the combination of homozygosity mapping and WES to identify members of the same family who are homozygous for a C2orf71 (MIM*613425) nonsense mutation (leading to typical RP), homozygous for a nonsense mutation in CEP250 (MIM*609689) leading to an atypical form of USH with a relatively mild retinal degeneration or double homozygotes for both mutations (leading to USH with early-onset and severe retinal degeneration).

Methods

Subjects and clinical evaluation

The tenets of the Declaration of Helsinki were followed and prior to donation of a blood sample, written informed consent was obtained from all individuals who participated in this study, after explanation of the nature and possible consequences of the study. The research was approved by the institutional review board at the Hadassah medical centre. DNA was extracted from the index patients as well as from other affected and unaffected family members using the FlexiGene DNA kit (QIAGEN).

Ocular evaluation included a full ophthalmological examination, Goldmann perimetry, electroretinography (ERG), colour vision testing using the Ishihara 38-panel and Farnsworth–Munsell D-15 tests, colour and infrared fundus photos, optical coherence tomography (OCT), and fundus autofluorescence imaging performed as previously described.15

Audiological screening included a comprehensive questionnaire (collecting information regarding any history of exposure to noise, ototoxic agents and genetic factors related to hearing impairment) as well as physical and audiometric examinations. An age-appropriate audiological examination was performed including pure-tone audiometry (250 to 12 000 Hz, air and bone conduction, including masking) and tympanometry for each ear. We used the following scale to grade the degree of SNHL: slight—16–25 dB hearing loss, mild—26–40 dB, moderate—41–55 dB, moderately severe—56–70 dB, severe—71–90 dB and profound—>90 dB hearing loss.16 Specific types of SNHL were determined by audiometric curve patterns: ascending (hearing loss greater at the lower frequencies), flat, descending (hearing loss greater at the higher frequencies), and ‘U-shaped’ (hearing loss at mid-frequencies) curves.

Genetic analyses

Whole genome SNP analysis was performed on DNA samples from seven patients using the Affymetrix 10K microarray system and data analysis was performed using HomozygosityMapper (http://www.homozygositymapper.org/). The 10k markers are usually distributed along the chromosomes, with a lower density in the centromeres and telomeres. Out of the 10 204 SNP markers, 5256 (51.5%) were informative (heterozygous in at least one of the parents). The informative markers were distributed along the genome and the highest number of consecutive non-informative SNP markers was only 23. WES analysis of two affected individuals (MOL0028 II:1 and II:4) was performed at Otogenetics Corporation using Roche NimbleGen V2 (44.1 Mbp) paired-end sample preparation kit and Illumina HiSeq2000 at a 31× coverage (ranging from 0 to 524 reads). Sequence reads were aligned to the human genome reference sequence (build hg19) and variants were called and annotated using the DNAnexus software package. Dataset files including the annotated information were analysed using ANNOVAR according to the dbSNP database (build 135) with the following filtering steps: (1) All variants in known retinal degeneration genes were analysed separately prior to filtering; (2) Variant type: including the following: missense, nonsense, insertions and deletions within the coding region, and splice-site; (3) Variants found within segmental duplications were excluded; (4) Variants with a minor allele frequency greater than 0.5% in the NHLBI Exome Sequencing Project were excluded (http://evs.gs.washington.edu/EVS/); (5) Variants with a SIFT (http://sift.jcvi.org/) score less than 0.05 were excluded; (6) Variants with a PolyPhen2 (http://genetics.bwh.harvard.edu/pph2/) score less than 0.85 were excluded; (7) Single heterozygous variants that do not fit an autosomal recessive inheritance were excluded; and (8) Variants that do not reside within homozygous regions (detected using the homozygosity mapping approach) were excluded.

Primers for the CEP250 (accession number NM_007186.3) and C2orf71 (accession number NM_001029883.2) genes (see online supplementary table S1) were designed using the Primer3 software17 and Sanger sequencing of PCR products was used to verify the mutations and to screen additional patients and ethnicity-matched controls. cDNA was synthesised using the Verso cDNA kit (Thermo) in accordance with the manufacturer's protocol.

Establishment of CEP250-expressing cell line

Fresh blood samples were drawn from patient II:7 and a healthy control. The peripheral blood lymphocytes were isolated by Ficoll-Hypaque density gradient centrifugation. Epstein–Barr virus (EBV) immortalised lymphoblastoid B cell lines (LCLs) were established with the B95-8 virus substrain. Lymphocytes were incubated for 1 h with the virus, washed and incubated in RPMI-1640 medium, containing 10% heat-inactivated fetal calf serum (FCS) and cyclosporin A (1 μg/mL). The typical clumps of the LCL were seen within 5–7 days.18 The cell lines were grown in RPMI-1640 medium, supplemented with 10% FCS, 2 mM l-glutamine, 100 U/mL penicillin G and 100 μg/mL streptomycin at 37°C in a 5% CO2 humidified atmosphere. The tissue culture medium components were purchased from Biological Industries, Beit Haemek, Israel.

Cell synchronisation and mitotic arrest

EBV transformed lymphoblastoid cells from a control individual and a patient homozygous for the CEP250 mutation were used 2 days after passage (total number of cells 5×107/sample) as described previously.19 Briefly, the cells were synchronised into G0 by suspension in FCS-free RPMI-1640 medium for 24 h. The cells were grown in 16% FCS for additional 4 h and then 10 mM hydroxyurea (H8627, Sigma-aldrich) for 12 h was added in order to arrest the cells in S-phase.20 The cells were collected and washed, then grown up in 10% FCS RPMI-1640 medium for 24 h.

Protein extraction and western blotting

EBV transformed lymphoblastoid B cells were homogenised in lysis buffer (20 mM HEPES ph-7.4, 1 mM MgCl2, 10.8% sucrose, 50 mM β-mercaptoethanol, 1 mL/100 mL protease inhibitor cocktail—Sigma P2714); all steps were done at 4°C. The lysate was centrifuged for 10 min at 170 g, and the pellet was separated and used as the nuclear fraction. The pellet was washed once with the lysis buffer (without Nonidet P40) and proteins were extracted by adding equal volume of lysis buffer with 1% Nonidet P40 for 30 min and then centrifuged for 5 min at 8000 g. The supernatant was isolated and the protein concentration was determined by the Bradford method. Samples of 100 μg protein were electrophoresed by SDS-PAGE using a 10% polyacrylamide gel and analysed by western blot analysis using the mouse polyclonal antibody against c-Nap1 (sc-390540, Santa Cruz) at a dilution of 1:200.

Nasal mucosal swab and electron microscopy processing

Nasal mucosal samples were obtained from patients II:1, II:4 and II:5 using cotton swab and preserved in glutaraldehyde vials. The processing and the electron microscopy investigation were done in the pathology department of the Rambam Medical Center, Haifa according to Toskala et al.21

Phylogenetic sequence analysis

Amino acid sequences of CEP250 and rootletin-related sequences were extracted from the Homologene database at National Center for Biotechnology Information (NCBI). Amino acid sequences were aligned with the ClustalW2 multiple alignment tool at European Bioinformatics Institute (EBI). Phylogenetic analysis was performed using the neighbour joining procedure, with Caenorhabditis elegans sequences serving as an outgroup branch. The following sequences were used: NP_055490.3 (Homo sapiens, rootletin), XP_002685844.1 (Bos taurus, rootletin), NP_742120.2 (Mus musculus, rootletin), NP_651216.2 (Drosophila melanogaster, rootletin), XP_311738.4 (Anopheles gambiae, AGAP003449-PA), NP_508848.2 (Caenorhabditis elegans, protein LFI-1), NP_494820.3 (Caenorhabditis elegans, protein DYF-14), NP_009117.2 (Homo sapiens, CEP250), XP_002692496.1 (Bos Taurus, CEP250), NP_001123472.1 (Mus musculus, CEP250), XP_417323.3 (Gallus gallus, CEP250), XP_692550.3 (Danio rerio, LOC564101) and XP_003118566.1 (Homo sapiens, rootletin-like).

Results

For this study, we recruited a consanguineous Iranian Jewish family (MOL0028—figure 1A) with seven affected individuals who suffer from various degrees of hearing and visual loss. The latter was due to retinal degeneration that was either mild or had an early-onset and severe form (figure 1A and clinical details below). One should take into account both shared and unrelated causes of the two clinical manifestations. In addition, the inheritance pattern cannot be determined based on the family tree (figure 1A) and both autosomal dominant and AR inheritance patterns are compatible with disease segregation in the family, although reported consanguinity (I:1 and I:2 are first cousins) makes AR inheritance more likely.

Figure 1

Genetic data of family MOL0028. (A) Family tree of MOL0028: the different clinical phenotypes are colour-coded and the genotype for the C2orf71 and CEP250 mutations is listed below each individual symbol. Consanguinity in generation #1 is marked with a double line and indication of the degree of relationship (2:2, ie, first cousins). M1-c.3463C>T (p.R1155*), M2-c.3289C>T (p.Q1097*). A two-point LOD score calculation of the CEP250 genotype and the USH phenotype yielded a value of 4.6 at θ=0. Homozygosity mapping (HM), patients with available homozygosity data. WES, patients with available WES data. (B) HM results showing two shared homozygous regions on chromosome 2 (including the C2orf71 gene) and chromosome 20 (including the CEP250 gene). Each bar shows the range of the homozygous region and the patient number. (C) Sequence chromatograms of the two identified mutations in C2orf71 and CEP250. For each mutation, the wildtype, heterozygous and homozygous mutants are shown. LOD, logarithm of odds; USH, Usher syndrome; WES, whole exome sequencing.

Our aim was to identify the genetic cause or causes of hearing and vision loss in MOL0028; thus, we performed homozygosity mapping using whole genome SNP arrays on seven family members (figure 1A,B). No genomic regions cosegregated with both phenotypes in an autosomal dominant (AD) pattern. All five affected siblings, however, had large homozygous regions on chromosome 20 (figure 1B), with a shared homozygous region of 17 Mb (between 24 and 41 Mb). However, the affected father, as well as the unaffected mother, was heterozygous for SNP markers in this genomic region on chromosome 20. Analysing the data separately for the two different retinal phenotypes did not reveal any potential candidate regions. It should be noted though that the father (I:1) and three of his children (figure 1B) had significant large homozygous regions on chromosome 2 with a shared homozygous region of 4.1 Mb, including the C2orf71 gene. The remaining affected individuals, however, did not show homozygosity in this region. Since the homozygosity mapping approach did not reveal any locus that fully cosegregted with the phenotype, we chose two individuals for WES analysis: II:1 with the relatively mild retinal phenotype and II:4 with a retinal phenotype compatible with severe, early-onset RP. Both patients had early-onset hearing loss.

We produced an average of 3.3 gigabases of paired-end 100-nucleotide sequence reads, 95% of which were mapped to gene regions. A total of 190 million bases were covered with an average coverage of 31 reads per nucleotide. Alignment of these reads to hg19 revealed about 150 000 variants in each exome (see online supplementary table S2) that underwent a series of filters to exclude sequence changes that are unlikely to be pathogenic. We initially examined all possible sequence variants within genes that were previously associated with inherited retinal diseases. The analysis revealed a C>T transition (c.3289C>T) leading to a premature stop codon (p.Q1097*) in C2orf71 (figure 1C). This mutation was found homozygously in patient II:4 and heterozygously in patient II:1. A full segregation analysis in the family (figure 1A) showed that it cosegregates with early-onset retinal degeneration but the remaining patients with a relatively mild RP are heterozygous for this recessive mutation, which therefore cannot explain the full spectrum of retinal phenotypes and hearing loss in this family. A subsequent analysis of the WES data revealed a C>T transition (c.3463C>T—figure 1C) leading to a premature stop codon (p.R1155*) in the centrosome-associated protein CEP250 (encoding the C-Nap1 protein). Mutations in C2orf71 were reported previously to cause ARRP22 ,23 while the CEP250 gene has not been associated thus far with any inherited human disease. To exclude the possibility that the CEP250—c.3463C>T variant is a common polymorphism in the Eastern Jewish population, we screened 160 ethnicity-matched control chromosomes, and all were negative for c.3463C>T. In addition, this variant was absent in the exome variant server (EVS), containing about 13 000 chromosomes. RT-PCR analysis of blood samples of patients homozygous for the CEP250 mutation revealed expression of a mutant transcript at levels similar to the expression of the wildtype transcript in normal control individuals (data not shown). The mutant transcript results in the production of an abnormal truncated protein, lacking a region that includes the NEK2-phosphorylation site (figure 2A), which is located at the COOH-terminus of the protein.24 Aiming to determine the effect of the CEP250 mutation on protein expression, we established lymphoblastoid cell lines of a homozygous patient (II:7) and a control individual. Cell growth was arrested at S-phase to enhance CEP250 expression.25 Western blotting analysis of these cells revealed the expression of both wildtype and mutant proteins (figure 2B). A similar truncated CEP250 protein was previously shown to disturb the localisation of rootletin at the centrosomes resulting in centrosome separation.26 Aiming to identify additional families with CEP250 mutations, we screened a set of 56 ethnicity-matched patients with retinal degeneration for the CEP250-c.3463C>T mutation. In addition, we screened the whole CEP250 open reading frame for mutations in eight patients with CEP250 homozygous regions as detected by whole genome SNP arrays. Both analyses were negative. To gain more information on the origin of the CEP250 gene, we performed evolutionary analysis of C-Nap1-related protein sequences (figure 2C). Mammalian and avian C-Nap1 proteins form a distinct evolutionary group with rootletin being its most related ortholog, indicating that the two proteins are paralogues in the human genome and arose by genomic duplication about 500 million years ago.

Figure 2

Properties of the CEP250 (C-Nap1) protein. (A) A protein scheme showing the mutation location and the NEK2-phosphorylation region as reported elsewhere.24 In addition, the C-Nap1 protein contains regions that are predicted to form coiled-coil structures (dashed regions).24 (B) Western blotting analysis of a wt sample (lane 1) and patient II:7 who is homozygous for both mutations in C2orf71 and CEP250 (lane 2). Proteins were extracted from lymphoblastoid B cell lines, and a CEP250-specific antibody was used. Wildtype CEP250 protein is expected to appear at 250 Kd and the truncated protein at 150 Kd. (C) A phylogenetic tree of CEP250-related proteins showing that CEP250 and rootletin are orthologues. Human rootletin L refers to a human sequence (XP_003118566.1) that is related to rootletin.

The two mutations are not genetically linked and therefore show an independent segregation. The C2orf71—c.3289C>T mutation segregates with retinal degeneration in a recessive pseudo-dominant fashion, but shows no segregation with SNHL (figure 1A). The father (I:1) and three of his children are homozygous for the mutation and suffer from retinal degeneration, as expected. The remaining five family members are heterozygous for this mutation. The CEP250—c.3463C>T mutation segregates with SNHL and a relatively mild retinal degeneration phenotype. Both parents are heterozygous and six of the seven children are homozygous for the mutation. Overall, four different C2orf71CEP250 genotypes and distinctive associated phenotypes can be identified in this family: three patients who are double homozygotes for both mutations suffer from SNHL accompanied by early-onset and severe RP, three patients who are homozygous for the CEP250 mutation and heterozygous for the C2orf71 mutation suffer from SNHL with a relatively mild form of retinal degeneration, one patient (I:1) who is homozygous for the C2orf71 mutation and heterozygous for the CEP250 mutation suffers from RP and age-depended hearing loss (see below), and two individuals who are double heterozygotes are clinically unaffected.

Aiming to better characterise the phenotype of patients in MOL0028, we performed a comprehensive hearing and vision assessments in eight family members (table 1 and see online supplementary table S3). Audiology testing revealed an age-related and mild SNHL in the father (I:1), a relatively mild SNHL in one of his sons (II:2) and early-onset with severe SNHL in four siblings with flat audiogram (figure 3 and see online supplementary table S3). All family members reported normal vestibular function. Ocular evaluation of seven family members (table 1 and figure 4) revealed two groups of patients, depending on the severity and onset of retinal degeneration. The first group (Group A) includes three patients (II:1, II:5, and II:6) who were homozygous for the CEP250 mutation (and heterozygous for the C2orf71 mutation) with a relatively mild form of retinal degeneration with measurable scotopic and photopic ERG amplitudes (40%, 32% and 43% of mixed, flicker and rod ERG normal lower limits, respectively) and visual acuity (VA) of 0.11 (at age 47) and 0.16 (at age 54—see table 1). Funduscopy (figure 4A–H) revealed peripheral retinal atrophy with white scars in the far periphery that contain a few bone spicule-like pigmentation (BSP) (figure 4A,D,E). OCT showed preservation of retinal layers in the posterior pole (figure 4K,L) and Goldman visual field testing (figure 4O,P) showed a constricted visual field with a preserved central vision of 45 and 35° on average between eyes. The second group (Group B) includes three patients (II:3, II:4 and II:7) who were double homozygotes and suffered from early-onset RP with non-detectable ERG responses on first visit (at ages 49, 16 and 6 years respectively). The VA was light perception in both eyes (table 1). Funduscopy (figure 4I,J) revealed diffuse retinal atrophy with highly abundant BSP pigmentation in the periphery, optic disc pallor and thinning of blood vessels. OCT showed diffused retinal atrophy encroaching the posterior pole (figure 4M,N). In addition, one patient (I:1) who was homozygous for the C2orf71 mutation (and heterozygous for the CEP250 mutation) had measurable ERG responses at the age of 66 years (table 1), which is comparable with other patients with C2orf71 mutations. Three individuals were double heterozygotes and had normal VA and visual function (data not shown). The double homozygosity genotype that is seen in some patients of MOL0028 is unique and allows us to study whether there might be an additive effect on visual function due to homozygosity for two mutations in unrelated ciliary genes. A plot of Snellen VA versus age for each of the three studied genotypes (figure 5) shows that double homozygous patients (Group B) have low VA compared with the two single-gene homozygous groups (t test: Group A (CEP250)—average VA=0.414, Group B—average VA=0.0085, Group C (C2orf71)—average VA=0.407; p value for A–B comparison is 0.014; B–C comparison 0.0013; A–C comparison p=0.96; there was no significant age difference).

Table 1

Ocular data of patients from MOL0028

Figure 3

Audiometric results of patients with different C2orf71 and CEP250 genotypes. Individuals III:1 and II:2 are unaffected and are heterozygous for both mutations. The audiometric test of III:1 shows normal hearing function at a relatively young age of 23 years. The audiometric test of II:2 shows mild to moderate SNHL with a flat pattern, likely to be the result of long exposure to noise. Patient II:4 is a double homozygous and suffers from a severe to profound flat SNHL. Patient II:1 is homozygous for the CEP250 mutation (and heterozygous for the C2orf71 mutation) and suffers from a severe flat SNHL. Patient I:1 is homozygous for the C2orf71 mutation (and heterozygous for the CEP250 mutation) and suffers from an age-related SNHL. SNHL, sensorineural hearing loss.

Figure 4

Ocular phenotype of patients with CEP250 and/or C2orf71 mutations. (A–H) Colour fundus photographs of patients II:5 (47 YO; panels A–D) and II:1 (54 YO; panels E–H) who are homozygous for the CEP250 nonsense mutation and heterozygous for the C2orf71 nonsense mutation. Panels B, F, C, G represent the posterior pole, OD and OS, respectively, with the classic findings of RP including retinal atrophy, pale optic discs and narrow blood vessels. Panels A, D, E, H show unique clusters of white lesions encroaching the atrophic periphery including bone spicule-like pigmentations (BSPs) within some of which, OD and OS, respectively. (I, J) Colour fundus photographs of patient II:7 (36 YO) who is homozygous for both nonsense mutations. The retinal atrophy together with the optic discs pallor, blood vessels narrowing and BSPs encroaching the whole posterior pole including the macula. (K) OD horizontal and (L) OD vertical OCT sections of patient II:5 showing thinning of the retinal layers in the periphery with preservation of the layers within the posterior pole. (M, N) OCT horizontal sections, OD and OS, respectively, of patient II:7 showing diffuse retinal atrophy. (O, P) The averaged Goldmann visual field of patients II:5 (47 YO) and II:1 (54 YO), respectively, is depicted showing a moderate visual field damage. OCT, optical coherence tomography; OD, oculus dexter; OS, oculus sinister; RP, retinitis pigmentosa; YO, years old.

Figure 5

Snellen visual acuity versus age in patients with different genotypes. Each point represents an average between the two eyes. Values obtained from patients who are double homozygotes for the C2orf71 and CEP250 mutations are shown as a rhombus shape (II:3 in red, II:4 in blue and II:7 in yellow). Values obtained from patients who are homozygous for the CEP250 mutation and heterozygous for the C2orf71 mutation are shown as a square shape (II:1 in brown, II:5 in blue and II:6 in green). Values obtained from patients reported previously to have C2orf71 mutations on both alleles22 ,33 ,34 are shown as a green triangle.

Aiming to examine the ciliary structure of patients with CEP250 mutations, we performed electron microscopy analysis on a nasal biopsy of three patients (II:1, II:4 and II:5) representing the two different combinations of genotypes. We did not detect any abnormalities in the structure of the cilia in the respiratory system (see online supplementary figure S1).

Discussion

We report here retinal degeneration and SNHL in patients from a single family who carry mutations in two ciliary genes, C2orf71 known to cause ARRP when mutated and CEP250, a novel cause of atypical USH due to a nonsense mutation. Since all USH patients studied here were homozygous for the nonsense CEP250 mutation and had at least one null C2orf71 allele, we cannot exclude the possibility that the C2orf71 mutation is required, in addition to the bialleleic CEP250 mutation, to develop USH. There is growing evidence of the essential role of ciliary proteins in photoreceptor function.27 ,28 The cilium is present in many different cell types and the importance of its function can be appreciated by the growing number of genetic diseases caused by mutations in genes encoding cilia-associated proteins.29 ,30 In particular, mutations in a large number of ciliary genes are known to cause either retinal degeneration or SNHL, as well as USH, affecting both vision and hearing.31 In addition, a large number of genes that can cause Bardet–Biedl syndrome (BBS) when mutated encode proteins that interact to produce the BBSome ciliary complex.32 One of the ciliary genes that cause ARRP when mutated is C2orf71 that was identified simultaneously by two groups 22 ,23 and was reported to cause non-syndromic ARRP with a variable disease severity, but neither deafness nor hearing loss was reported.22 ,23 ,33 ,34 Although the exact localisation of C2orf71 in photoreceptors is unknown, colocalisation assays performed in cell cultures indicate ciliary expression.23

CEP250 encodes the C-Nap1 protein which belongs to the CEP family of proteins,35 including over 30 proteins that form the active component of centrosome and play a vital role in centriole biogenesis, centrosome cohesion and cell cycle progression control.36 C-Nap1 dissociates from spindle poles during mitosis, but reaccumulates at centrosomes at the end of cell division and was shown to be required for cell cycle-regulated centrosome cohesion.37 C-Nap1 was previously reported to be expressed in photoreceptor cilia and is also known to interact with other ciliary proteins, including rootletin and NEK2.24 Both proteins are highly important for retinal function: rootletin is a major structural component of the photoreceptor basal body38 and NEK2 (MIM*604043) was recently shown to cause ARRP when mutated.39 At the time of centriole duplication, Nek2 phosphorylates C-Nap1, which in turn mediates the release of centriole connecting fibre and helps in centriole disjunction.35 No data are available regarding the expression of CEP250 in the inner ear, but the reported expression in the basal body,26 similar to other USH proteins, suggests that CEP250 is expressed and functions in the inner ear as well.

Three of the patients we identified in the current study are ‘double homozygotes’ for C2orf71 and CEP250 mutations, and both encode ciliary proteins. It is therefore intriguing to assess whether mutations in these genes show an additive effect. To the best of our knowledge, two individuals were reported so far as ‘double homozygotes’ for mutations in two retinal-disease genes. The first patient was found to be homozygous for missense mutations in GPR98 (MIM*602851), mutations in which are known to cause Usher type 2, and PDE6B (MIM*180072), mutations in which are known to cause ARRP.40 Retinal examination and funduscopy of this patient showed a more advanced retinal degeneration comparing with her relatives who harbour only one of the mutations; however, ERG data were not presented. The second patient was found to be homozygous for frameshift mutations in MYO7A (MIM*276903), mutations in which are known to cause Usher type 1, and PDE6B.41 Although ERG was performed on this individual and compared with his relative with single homozygous mutations, a clear-cut conclusion regarding the severity of his disease could not be made due to the patient’s age (32 months old). In the current study, we compared VA and ERG parameters between double homozygous patients, patients who were homozygous for the CEP250 mutation and patients who were homozygous for C2orf71 mutation. A statistically significant difference in VA was obtained between the double homozygotes and the two other groups, indicating that mutations in CEP250 and C2orf71 have an additive effect on retinal burden and can cause a more severe retinal phenotype when they occur at the same cell. It is likely to be assumed that the function of these proteins at the same photoreceptor region (eg, the connecting cilium) contributes to this effect.

In summary, our data indicate that a homozygous nonsense mutation in CEP250 causes a novel form of USH that is characterised by early-onset hearing loss and a relatively mild retinal degeneration phenotype. The more severe retinal involvement in the double CEP250 and C2orf71 homozygotes may indicate an additive effect caused by nonsense mutations in genes encoding two ciliary proteins.

Acknowledgments

The authors thank the patients and their families for their participation in the study. The authors thank Dr Tamar Ben-Yosef and Professor Michael Steinitz for their excellent help.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

Footnotes

  • EB and DS contributed equally to this study.

  • Contributors Performed the experiments: SK, LZ, LM-M. Analysed data: SK, LZ, CA, RKK, UL, MG, EB, DS. Drafted the manuscript: SK, DS, EB. Revised the manuscript: LZ, LM-M, CA, RKK, US, MG.

  • Competing interests This study was financially supported by the Foundation Fighting Blindness USA (BR-GE-0510-0490-HUJ to DS), the Yedidut 1 research grant (to EB) and a grant from the Israeli Ministry of Health (3-00000-9177 to SK). RKK was supported by the Foundation Fighting Blindness Canada, the CIHR and NIH.

  • Ethics approval The Institutional Review Board (IRB) at the Hadassah Medical Center.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.