Exome Sequencing In Gene Identification In Recessive Hereditary Hearing Loss And Pheno Type-genotype Correlation

Hereditary hearing loss will impact the communication and social activities,and is the increasingly severe disease unresolved. Prevention of deafness andimprovement of population is very important for national economy and the people'slivelihood. The root of the prophylaxis was to thoroughly explore the causative geneand mutation spectrum and discover the gene distribution and hot spot, in order toinstruct the marriage and procreation among deafness to avoid the incidence ofdeafness. Most of Chinese hereditary deafness were in recessive mode, andcausative genes in60-70%population were unknown. This study applied the nextgeneration sequencing technique-exome sequencing in autosomal recessivehereditary hearing loss families, and identified CDH23gene to be the causative gene.CDH23gene was the common deafness gene, and occupied the NO.2-3innonsyndromic hearing loss genes. We then srcreened the known mutations ofCDH23related to DFNB12in108recessive deafness families. In the background ofmultiple causative genes identified, this study performed the genotype phenotyperelation analysis about GJB2and SLC26A4genes, and clarified the inpact ofgenotype to phenotype to some extent.PART1: CDH23was identified as causative gene in autosomalrecessive hereditary hearing loss and known mutations screening in108probands of recessive familiesTo identify the causative gene in3families with autosomal recessive hereditary hearing loss, we performed the common causative gene screening (GJB2,SLC26A4and m.1555A>G), and no mutation was found out. Then, from family784and845,2patients and their parents were recrued respectively, and from family855,2patients,1sibling, their mother and uncle were recrued for the father was dead. Thewhole exome sequencing was applied in the13subjects.The quality control report of the sequencing showed: sequencing depth was50X,and about95000SNPs, about2500splice site mutations, about7000indels, about58000introns in10bp near the flanking part of exons.Based on the databases, including dbSNP, Hapmap-8,1000Genomes andYanHuang (YH), the raw data were filtered as follows:Firstly, the raw data from each subject was filtered by such databasesrespectively, and rare mutations were selected. Then the data from patients werefiltered by the data from normal controls. In the third step, according to theco-segregating principle in autosomal recessive hereditary mode, the mutationsappeared homozygous state in patients and heterozygous state in controls were takenas the candidate genes.In the candidate genes, preferred mutations were the indels less than5bp in theconserved region in multiple species, splice site mutations, nonsense mutations, anddamaging missense mutations predicted by SIFT software.In family784, no preferred mutations and genes were identified. In family845,preferred mutation was the mutation of G>A(c.6442G>A,p.D2148N)on73553127of CDH23gene. In family855, a novel mutation(c.403C>T, p.Q135*)in CDH23gene was preferred to be the causative mutation. CDH23gene was the knowncausative gene, and the mutation c.6442G>A was the known mutation while thenovel mutation c.403C>T was thought to be damaging by SIFT software. By theophthalmologic examination, the patient with c.403C>T mutation was found withsevere rentitis pigmentosa (RP), and was diagnosed with Usher1D syndrome.By literature reviews, about29mutations in CDH23gene related with DFNB12were found out, which were scattering in Caucasians, Africans and Mongolians over the world including America, German, Holland, Japan, India, Pakistan and Algeria.c.403C>T was not found in100geographically matched controls.108families with autosomal recessive hereditary hearing loss were performedthe common causative gene screening (GJB2,SLC26A4and m.1555A>G), and nomutation was found out. Known mutations screening was applied in the108familiesby MassARRAY system.3monoallelic mutation carriers were found out.Together with2homozygotes found by exome sequencing, and the mutationfrequency was3.24%(7/216).PART2: Genotype-phenotype correlation analysis in295deafnesscarrying GJB2biallelic mutationsConnexin26coding gene (GJB2) is the primary causative gene fornonsyndromic sensorineural hearing impairment (NSSHI). More than100mutationsin this gene have been reported to be linked to hearing impairment, from mild toprofound hearing loss. In order to precisely estimate the impact of GJB2mutationsin Chinese population, a cross-sectional study was performed to analyze the auditorydata of Chinese NSSHI patients.295unrelated NSSHI patients with biallelic mutations in GJB2were recruitedfrom7provinces in Northern China from2004to2008. The levels of hearingimpairment (HI) and average pure tone audiometry (PTA) were compared acrossdifferent genotypes by χ2testing. The subjects with the genotypes of combinedtruncating mutations had more cases of severe HI than the subjects with genotype ofnontruncating mutations combination. It was also revealed that subjects carryingeither c.[79G>A;341A>G]+[79G>A;341A>G] or c.[109G>A]+[79G>A;341A>G]had significantly less cases of severe HI than the reference group of homozygousc.235delC, while the subjects carrying c.[235delC]+[176₁91del16] had more casesof severe HI than the homozygous c.235delC group.This is the first study to clarify the correlations between different GJB2 biallelic genotypes and NSSHI phenotype in Chinese population. The Chinesesubjects with two truncating mutations in GJB2were shown to correlate with moresevere hearing impairmentPART3: Genotype-phenotype correlation analysis in272childrendeafness with large vestibular aqueductThis study was designed to draw a spectrum of SLC26A4gene mutations inchildren with large vestibular aqueduct (LVA) and to study the phenotype genotyperelationship.272subjects under age12years with nonsyndromic sensory neural hearing loss(NSSHL) and LVA, received SLC26A4gene screening.266received completeauditory examinations. High-resolution computed tomography (HR-CT) of temporalbone were measured in152subjects. According to genotypes, the phenotypesincluding PTA (pure tone average), distribution of subjects, and diameter of externalaperture and middle portion of vestibular aqueduct, were compared by t test or chisquare tests by SPSS13.0. Further, divided by dilated level of vestibular aqueduct,subject distribution in different hearing loss levels were compared by chi squaretests by SPSS13.0.69types of mutations were identified, in which2were novel mutations(c.665G>T and c.1639A>G).170subjects were found with biallelic mutations inSLC26A4gene,56with monoallelic mutation,8with no mutation, and one withthree mutations. The hearing loss was more stable in the subjects with monoallelicmutation than in other genotype groups (P<0.05). The air-bone gap was morefrequently seen in subjects with biallelic missense mutations than those withbiallelic truncating, heterozygous truncating, or monoallelic mutation (all Pvalues<0.05). There is no difference for distribution of diameters distribution indifferent genotypes. And there is no dominant correlation between hearing loss leveland dilated levels of vestibule aqueduct. This is the first study to clarify the correlations between different SLC26A4genotypes and phenotype in Chinese children population with LVAS. In childrenwith LVAS,98%carry mutations in SLC26A4gene. Though we found somecorrelation between genotype and phenotype, it is far away from erecting directcorrelation between the mutation genotype and hearing loss level or LVA diameter.