Search for new genes, molecular mechanisms and diagnostic tools in postlingual hereditary hearing loss

Abstract

Non-syndromic sensorineural hearing loss (SNHL) is the most common sensory disorder, and it presents a high genetic heterogeneity. New genes and variants associated with hearing loss are described at a very rapid pace, making the use of next-generation sequencing (NGS) approaches essential to undertake a genetic diagnosis. We have used OTO-NGS-panel V2, a custom-designed NGS panel containing 117 genes associated with hearing loss to study the mutational spectrum of autosomal dominant non-syndromic hearing loss in the Spanish clinical population. OTO-NGS-panel V2 was validated for the identification of single nucleotide variants (SNVs), small insertions and deletions (Indels) and copy number variations (CNVs), and 108 Spanish families with autosomal dominant non-syndromic hearing loss (ADNSHL) were analysed. We identified the causative variant in 39 cases, constituting a diagnostic rate of 36.11%. The genes with the highest prevalence are WFS1 (7.41%) followed by MYO7A (5.56%), MYO6 (4.63%), TECTA (2.78%) and ACTG1 (2.78%). To identify the genetic cause of deafness in the cases that remained undiagnosed upon analysis with OTO-NGS-panel V2, we designed a candidate gene panel, including genes known to cause deafness in mice but with no human mutations reported so far. Candidate variants were found in the genes XIRP2, USP31, PLS1 and KCNK5. Furthermore, we have used cellular and animal models to investigate the pathogenic mechanisms of hearing loss due to mutations in CCDC50 and MIR96, respectively. As part of our clinical genetic studies, we ascertained a novel mutation in CCDC50 (c.828_858del, p.(Asp276Glufs*40)) segregating with the hearing impairment in a Spanish family with autosomal dominant SNHL that is predicted to disrupt the protein function. To gain insight into the mechanism behind CCDC50 mutations, we carried out in vitro studies on a set of artificial mutants and on the p.(Asp276Glufs*40) and p.(Phe292Hisfs*37) human mutations and determined that only the mutants containing the six amino acid sequence CLENGL as part of their aberrant protein tail showed an abnormal distribution consisting of perinuclear aggregates of the CCDC50-encoded protein Ymer. Therefore, we conclude that the CLENGL sequence is necessary to form the aggregates. These results, together with the normal hearing thresholds in Ccdc50 loss-of-function mouse mutants (Ccdc50tm1a and Ccdc50tm1b), suggest that the two human mutations in CCDC50 exert their effect through a dominant-negative or gain-offunction mechanism rather than by haploinsufficiency. Finally, we wanted to investigate the underlying pathological mechanisms of mutations in miR-96, which is important for hearing since it acts as a transcriptional regulator in the inner ear and coordinates hair cell maturation. Point mutations in the seed region of miR-96 cause dominant progressive hearing loss in humans and mice. Here, we present two mouse mutants carrying two point mutations identified as underlying progressive hearing loss in humans (Mir96s403 and Mir96s1334). Homozygotes of both mouse lines exhibit profound hearing loss, Mir96s1334 heterozygous mice have mild progressive hearing loss, but Mir96s403 heterozygotes have normal hearing in contrast to humans carrying this mutation. Structural analyses showed hair cell degeneration and misshapen hair bundles in both mutants, with Mir96s1334 mice being more severely affected. Our new results indicate that Mir96s403 homozygotes have a reduction in the number of inner hair cell ribbon synapses at four weeks old. Furthermore, bulk RNA-seq of the organ of Corti of 4-days-old mice showed that Mir96s1334 homozygous mice have a larger number of differentially-expressed genes than Mir96s403 homozygotes, which correlates with the more severe structural phenotype observed in these mutants. Sylamer analysis suggests that the phenotype of Mir96s403 and Mir96s1334 mutants is a combination of the lack of repression of normal targets and the gain of repression of novel targets. Ocm and Slc26a5 are strongly downregulated in Mir96s403 and Mir96s1334 mutant mice, but neither has a predicted wildtype or mutant miR-96 target site, suggesting that their downregulation is a downstream effect. To determine whether these genes regulate each other in the miR-96 regulatory network, we studied the effect of knocking down each of the genes in mice on the expression of the other gene by using Ocmtm1e and Slc26a5tm1Cre mutant mice. We used RT-qPCR to investigate the inner ear of 5-days-old and 10-weeks-old mice. The mRNA levels of Slc26a5 were unaffected in Ocmtm1e homozygotes, and Ocm levels were unaffected in Slc26a5tm1Cre homozygotes. Our findings indicate that it is unlikely that oncomodulin regulates prestin, or vice versa. Overall, this thesis provides insights into the genes with more mutations in a cohort of Spanish families with SNHL and suggests possible mechanisms for hearing loss caused by mutations in CCDC50 and MIR96.