Genomic copy number alterations in non-syndromic hearing loss

Transcription

1 Clin Genet 2016: 89: Printed in Singapore. All rights reserved Short Report 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd CLINICAL GENETICS doi: /cge Genomic copy number alterations in non-syndromic hearing loss Rosenberg C., Freitas É.L., Uehara D.T., Auricchio M.T.B.M., Costa S.S., Oiticica J., Silva A.G., Krepischi A.C., Mingroni-Netto R.C. Genomic copy number alterations in non-syndromic hearing loss. Clin Genet 2016: 89: John Wiley & Sons A/S. Published by John Wiley & Sons Ltd, 2015 Genetic heteroity has made the identification of s related to hearing impairment a challenge. In the absence of a clear phenotypic aetiology, recurrence risk estimates are often based on family segregation and may be imprecise. We profiled by oligonucleotide array-cgh patients presenting non-syndromic hearing loss with presumptive autosomal recessive (n = 50) or autosomal dominant (n = 50) patterns of inheritance. Rare copy number variants (CNVs) were detected in 12 probands; four of the detected CNVs comprised s previously associated with hearing loss (POU4F3, EYA4, USH2A, andbcap31) and were considered causative, stressing the contribution of genomic imbalance to non-syndromic deafness. In six cases, segregation of the CNVs in pedigrees excluded them as causative. In one case, segregation could not be investigated, while in another case, a point mutation likely explains the phenotype. These findings show that the presumptive patterns of inheritance were incorrect in at least two cases, thereby impacting tic counselling. In addition, we report the first duplication reciprocal to the rare ABCD1, BCAP31, andslc6a8 contiguous deletion syndrome; as with most microduplication syndromes, the associated phenotype is much milder than the respective microdeletion and, in this case, was restricted to hearing impairment. Conflictofinterest The authors declare no conflict of interest C. Rosenberg a,,é.l.freitas a,, D. T. Uehara a, M. T. B. M. Auricchio a, S. S. Costa a, J. Oiticica b, A. G. Silva a,a.c.krepischi a and R. C. Mingroni-Netto a a Department of Genetics and Evolutionary Biology, Institute of Bioscience, and b Department of Otorhinolaryngology, Medical School, University of São Paulo, São Paulo, Brazil These two authors contributed equally to the paper Key words: ADNSHL ARNSHL CNV copy number variants deafness non-syndromic hearing loss Corresponding author: Dr Carla Rosenberg, Departamento de Genética e Biologia Evolutiva, Instituto de Biociências, Universidade de São Paulo, São Paulo, SP, Brazil. Tel.: ; fax: ; carlarosenberg@uol.com.br Received 1 August 2015, revised and accepted for publication 6 October 2015 Hearing loss is one of the most common sensory defects in humans, with a frequency of at least 1/1000 births, 50% of which can be attributed to tic causes. Approximately 70% of hereditary hearing loss is non-syndromic (NSHL), wherein hearing impairment is not associated with any additional clinical phenotypes. Monogenic hearing loss can be inherited in different ways. Autosomal recessive hearing loss (ARNSHL) occurs in 80% of cases and is typically pre-lingual, whereas autosomal dominant hearing loss (ADNSHL) accounts for approximately 20% of cases and is more often post-lingual. In approximately 1 3% of patients, transmission occurs through the X-chromosome or the mitochondria (1). To date, 87 s for NSHL have been identified ( A previous study by our group showed that syndromic deafness is frequently associated with chromosome micro-imbalances (2), while chromosomal alterations in non-syndromic patients tend to be both rare and small, pinpointing specific s (3). Materials and methods Subjects Two Brazilian patient cohorts with non-syndromic hearing loss were used in this study: 50 patients with a presumptive dominant pattern of inheritance (ADNSHL) based on the presence of affected individuals in two consecutive rations, and 50 with a presumptive 473

2 Rosenberg et al. recessive pattern of inheritance (ARNSHL) based either on two affected siblings born to non-affected parents or at least one affected child born to consanguineous parents. Patients were diagnosed at the tic counselling unit of the Institute of Biosciences, University of São Paulo. They had been referred by different institutions mainly DERDIC (Catholic University of São Paulo) and the Academic Hospital of the Medical School of the University of São Paulo. Clinical and alogical data were collected for each affected individual. Signed informed consent was obtained from all subjects or their legal guardians, and the project was approved by the Bioscience Institute Ethics Committee of the University of São Paulo. The following mutations commonly associated with hearing loss were excluded prior to the study: c.35delg and c.167delt GJB2 mutations, delgjb6 [D13S1830] and delgjb6 [D13S1854], the mitochondrial m.a1555g mutation, and other mutations in the coding region of the GJB2. Array-CGH DNA from peripheral blood was isolated from family members following standard procedures. Array-CGH was performed by co-hybridizing genomic DNA from the index patient and a sex-matched reference (Promega) to the platform from Agilent Technologies (Design 22060), containing 170,000 distinct oligonucleotide probes distributed along the genome with 13 kb overall median probe spacing and 11 kb in Refseq s. Hybridization and washings were carried out according to manufacturer s protocol. Scanned images of the arrays were processed using Feature Extraction software, and data were analysed with Genomic Workbench software, both from Agilent Technologies. To make DNA copy number variation (CNV) calls, we used the Aberration Detection Method 2 (ADM2) statistical algorithm with a sensitivity threshold of 6.7. A genomic segment was considered duplicated or deleted when the log2 ratio of the Test/Reference fluorescent intensities of a given region encompassing at least three probes was above 0.3 or below 0.3, respectively. Detected CNVs were compared to data from oligoarray studies documented in the Database of Genomic Variants [DGV; and those found in the normal population were disregarded. Confirmatory studies We applied a loop design in our hybridizations as previously reported (4), resulting in two reverse labelled hybridizations per sample. Alterations not detected in both dye-swap experiments of the same sample were excluded from the analysis. Familial segregation of the rare CNVs identified was investigated either by array-cgh or real-time quantitative PCR (qpcr) using the TaqMan Copy Number Assay system (Applied Biosystems) or the SYBR Green system on a 7500 Fast Real-Time PCR System apparatus as previously described (5). The sequence of the RT-PCR primers is described in Supplemental Table 1, Supporting Information. Results Table 1 summarizes the results of the copy number study. Array-CGH screening of 100 patients with hearing impairment revealed a total of 12 non-recurrent rare DNA copy number variants (one in each individual), 7 from the cohort with a presumptive autosomal dominant pattern of inheritance and the remaining 5 from the autosomal recessive cohort. Supplemental Figure 1 shows RT-PCR confirmatory results for copy number alterations detected in Patients 1, 3, 4, 6 and 7. The sequence of the RT-PCR primers is in Supplemental Table 1. Four variants affected s already known to cause deafness: POUF4F3 and EYA4 heterozygous deletions, USH2A homozygous deletion and BCAP31 duplication. The copy number variants were considered causative in these four cases. Six rare CNVs were found not to segregate with phenotypes and were considered unrelated variants. In one case (Patient 5), the duplication segregated in the family with the phenotype, but an additional mutation in the KCNQ4 (DFNA2A) was identified and is now considered the most plausible explanation for hearing loss (6). We were unable to test the segregation of the duplication detected in the remaining family (Patient 9) because the relatives refused further investigation. This variant harbours a single (FGGY) that remains a candidate for deafness. Discussion An earlier study from our group in patients with syndromic hearing loss revealed that chromosomal imbalances are a common cause of the phenotypes (2). In recent years, some publications have reported that rare copy number alterations are also a cause of non-syndromic hearing loss (3, 7 9). As anticipated, CNVs detected in non-syndromic patient cohorts were small and harboured one or a small number of s. While few new candidates emerged from these CNV studies, it is increasingly clear that genomic imbalances affecting s already implicated in hearing loss are a common cause of these phenotypes and may not have been detected by conventional sequencing. Among the 12 copy number alterations detected in this study, four contained s already known to be associated with hearing loss. Shearer et al. (3), using targeted sequencing of s already known to underlie deafness, found a higher frequency of copy number alterations (15%) that detected in the present study. While the approach used here, in contrast to the study of Shearer et al., investigates genome wide alterations and allows discovery of new loci, the resolution is much lower and smaller alterations would be missed. Two of those known s, namely POU4F3 (DFNA15, MIM#602459, Online Mendelian Inheritance in Man: and EYA4 (DFNA10, 474

3 Genomic copy number alterations in non-syndromic hearing loss Table 1. Rare copy number alterations detected in the cohort of individuals with hearing loss. Genes in underline are known to be associated to hearing impairment. ID Cohort Genomic position (GRCh37/hg19) Size (kb) Copy number type Genes Deafness-associated phenotype Segregation with phenotype Conclusion 1 Dominant chr4:40,099,182-40,342, Duplication N4BP2, RHOH, CHRNA9 2 Dominant chr5:145,702, ,758, Deletion POU4F3 Autosomal dominant 15 (MIM #602459) No Unrelated variant Yes (Freitas et al 2012) Dominant, known 3 Dominant chr6:65,480,408-65,760, Deletion EYS No Unrelated variant 4 Dominant chr6:133,517, ,693, Deletion EYA4 Autosomal dominant Yes Dominant, known 10 (MIM#601316) 5 Dominant chr7:110,943, ,381, Duplication IMMP2L, DOCK4 Yes Unrelated variant 6 Dominant chr17:70,879,778-70,967, Duplication SLC39A11 No Unrelated variant 7 Dominant chr1:216,017, ,067, Deletion USH2A Usher syndrome, type 2A (MIM #276901) Yes Recessive, known 8 Recessive chr5:147,386, ,514, Deletion SPINK5 No Unrelated variant 9 Recessive chr1:59,853,969-60,046, Duplication FGGY Not tested Unrelated variant 10 Recessive chr8:2,350,092-3,737, Duplication CSMD1 No (inherited from normal father) 11 Recessive chr2:20,596,950-20,835, Deletion RHOB, HS1BP3-IT1, HS1BP3 Unrelated variant No Unrelated variant 12 Recessive chrx:152,953, ,969, Duplication SLC6A8, BCAP31 De novo X-linked, known s 475

4 Rosenberg et al. A B Fig. 1. Pseudo-dominant inheritance of hearing loss in the family of Patient 7. (a) Pedigree of the family. (b) Array-comparative genomic hybridization profile of the deleted chromosomal region at chr1q41 encompassing the USH2A. The affected members of the family show homozygous deletions, while the unaffected father and brother show a heterozygous deletion. MIM#601316), result in ADNSHL; accordingly, the alterations were detected in patients from the dominant cohort. Although deletion of EYA4 has been described in association with dilated cardiomyopathy in addition to deafness (MIM#605362), no other clinical signs have been identified in Patient 4. One homozygous deletion harbouring the recessive USH2A was detected in Patient 7, who was part of the dominant cohort based on transmission of the phenotype from mother to sons (pedigree in Fig. 1a). While array-cgh revealed homozygous USH2A deletion in the affected individuals, the father and unaffected son were heterozygous (Fig. 1b). Mutations in USH2A causes Usher syndrome type 2A (MIM#276901), an autosomal recessive condition characterized by congenital bilateral sensorineural hearing loss and late onset retinitis pigmentosa. Based on the array-cgh results, family members were re-evaluated and loss of vision due to retinitis pigmentosa was detected in the mother and affected sons. The presence of identical deletions, homozygous in the mother and heterozygous in the father, was surprising as they self-reported as non-consanguineous; however, both parents were born in the same small village of approximately 8000 inhabitants. A founder effect, followed by undetected inbreeding, might explain this unusual finding. The presence of a recessive phenotype in two rations due to marriage of the affected individual to a heterozygous carrier has been referred in the literature as pseudodominant inheritance (10, 11). The fourth variant was detected in Patient 12, whose hearing impairment was classified as recessive based 476

5 Genomic copy number alterations in non-syndromic hearing loss on parental consanguinity. However, array-cgh showed that the patient carries a de novo 16 kb microduplication at Xq28, encompassing the BCAP31 and SLC6A8 s. Mutations and deletions of BCAP31 have been associated with a severe phenotype of deafness, dystonia, and cerebral hypomyelination (MIM#300475). The is also included in the rare ABCD1, BCAP31,andSLC6A8 contiguous deletion syndrome associated with a severe clinical phenotype that includes marked developmental delays, significant growth failure, liver dysfunction, and early death (12). This is the first description of a reciprocal duplication of BCAP31 and SLC6A8; if this variant is causative, the finding suggests that duplications BCAP31 and SLC6A8 are associated with a much milder phenotype, apparently consisting only of hearing impairment. However, further studies are required to confirm a causative association between this Xq28 duplication and deafness. We did not detect a relatively common copy number present in 1% of the population, the heterozygous deletion encompassing the STRC on chromosome 15q15.3 (13). This deletion, when in homozygosity, was reported as cause of deafness in few families (13, 14). Given the small size of our cohort, the absence of this CNV is not unexpected. The eight remaining rare alterations could encompass candidate s for deafness, but six were found not to segregate with the phenotype in the families. Although the duplication at 7q31.1, which harbours IMMP2L and DOCK4, segregated with the disorder, the family was later found to carry a mutation in KCNQ4 (DFNA2A) that is likely to be pathogenic and that segregates with the phenotype (6). One of the alterations (duplication of 1p32.1 encompassing FGGY) could not be tested for segregation and remains a candidate for deafness. FGGY (carbohydrate kinase domain containing) is a ubiquitously expressed whose function is poorly characterized and has been recently implicated by genome-wide association studies in sporadic amyotrophic lateral sclerosis (ALS; MIM*611370). Currently, given the impossibility of testing the family, there is little evidence to support the role of FGGY in deafness. While this study did not reveal novel candidate causative s for hearing impairment, it makes evident the relevance of a proper molecular diagnosis for tic counselling. As discussed above, the initial classification of inheritance patterns based only on phenotypic segregation proved incorrect for at least two families, and molecular cytotic results led to major changes in the estimated risks of recurrence in affected and unaffected family members. Supporting Information Additional supporting information may be found in the online version of this article at the publisher s web-site. Acknowledgements This work was supported by FAPESP and CNPq. The authors are indebted to all of the professionals at DERDIC (Pontifícia Universidade Católica) for collaborating on hearing evaluations. References 1. Smith RJH, Shearer AE, Hildebrand MS, Van CG. Deafness and Hereditary Hearing Loss Overview. Seattle, WA: University of Washington, Catelani AL, Krepischi AC, Kim CA et al. Chromosome imbalances in syndromic hearing loss. Clin Genet 2009: 76: Shearer AE, Kolbe DL, Azaiez H et al. Copy number variants are a common cause of non-syndromic hearing loss. Genome Med 2014: 6: Allemeersch J, Van VS, Hannes F, De MB, Vermeesch JR, Moreau Y. An experimental loop design for the detection of constitutional chromosomal aberrations by array CGH. BMC Bioinformatics 2009: 10: Silva AG, Krepischi AC, Torrezan GT et al. Does germ-line deletion of the PIP constitute a widespread risk for cancer? Eur J Hum Genet 2014: 22: Mingroni-Netto RC, Uehara D, Freitas E et al. A novel KCNQ4 mutation and a private IMMP2L-DOCK4 duplication segregating with nonsyndromic hearing loss in a Brazilian family. Hum Genome Var In press. 7. Bademci G, Diaz-Horta O, Guo S et al. Identification of copy number variants through whole-exome sequencing in autosomal recessive nonsyndromic hearing loss. Genet Test Mol Biomarkers 2014: 18: Haraksingh RR, Jahanbani F, Rodriguez-Paris J et al. Exome sequencing and genome-wide copy number variant mapping reveal novel associations with sensorineural hereditary hearing loss. BMC Genomics 2014: 15: Ji H, Lu J, Wang J, Li H, Lin X. Combined examination of sequence and copy number variations in human deafness s improves diagnosis for cases of tic deafness. BMC Ear Nose Throat Disord 2014: 14: Pampanos A, Neou P, Iliades T et al. Pseudodominant inheritance of DFNB1 deafness due to the common 35delG mutation. Clin Genet 2000: 57: Tekin M, Duman T, Bogoclu G, Incesulu A, Cin S, Akar N. Moderate hearing loss and pseudodominant inheritance due to L90P/35delG mutations in the GJB2 (connexin 26). Genet Couns 2003: 14: van de Kamp JM, Errami A, Howidi M et al. Genotype-phenotype correlation of contiguous deletions of SLC6A8, BCAP31 and ABCD1. Clin Genet 2015: 87: Hoppman N, Aypar U, Brodersen P, Brown N, Wilson J, Babovic-Vuksanovic D. Genetic testing for hearing loss in the United States should include deletion/duplication analysis for the deafness/infertility locus at 15q15.3. Mol Cytot 2013: 6: Knijnenburg J, Lesnik Oberstein SA, Frei K et al. A homozygous deletion of a normal variation locus in a patient with hearing loss from non-consanguineous parents. J Med Genet 2009: 46: