A Novel GJB2 compound heterozygous mutation c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) causes sensorineural hearing loss in a Chinese family

Abstract

Objective

To investigate whether a novel compound heterozygous mutations c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) in GJB2 result in hearing loss.

Methods

Allele‐specific PCR‐based universal array (ASPUA) screening and sequence analysis were applied to identify these mutations. 3D model was built to perform molecular dynamics (MD) simulation to verify the susceptibility of the mutations. Furthermore, WT‐ and Mut‐GJB2 DNA fragments, containing the mutation of c.257C>G and c.176del16 were respectively cloned and transfected into HEK293 and spiral ganglion neuron cell (SGNs) by lenti‐virus delivery system to indicate the subcellular localization of the WT‐ and Mut‐CX26 protein.

Results

A novel compound heterozygous mutation c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) in GJB2 was identified in a Chinese family, in which 4 siblings with profound hearing loss, but the fifth child is normal. By ASPUA screening and sequencing, a compound heterozygote mutations in GJB2 c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) were identified in these four deaf children, each of the mutated GJB2 gene were inherited from their parents. There is no mutation of GJB2 gene identified in the normal child. Besides, the compound heterozygous mutation GJB2 c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) could lead to the alterations of the subcellular localization of each corresponding mutated CX26 protein and could cause the hearing loss, which has been predicted by MD simulation and verified in both 293T and SGNs cell line.

Conclusion

The c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) compound mutations in GJB2 detected in this study are novel, and which may be associated with hearing loss in this Chinese family.

1. INTRODUCTION

To our knowledge, there are at least more than 134 genetic loci and 80 different genes identified to associate with sensorineural hearing loss (SNHL), one of the most common human sensory disorders1 [http://hereditaryhearingloss.org]. Gap junctions (GJs), one type of proteins working as an intercellular channel that mediates cell‐cell communication and tissue homeostasis, have been reported to involve pathological process of hearing loss.2

CX26, is one type of the Gap junctions most frequently involved in autosomal recessive nonsyndromic hearing loss in some population.3, 4 As a key factor, it has been proposed to play an important role in the recycling of the K⁺ ion between the endolymph and perilymph.5 CX26 is located in the fibrocytes of spiral ligaments and limbi, basal cells of striavascularis, and all types of supporting cells in the organ of Corti. The encoded gene of CX26 is known as GJB2, which could be responsible for most autosomal recessive nonsyndromic phenotype hearing loss and tend to be prelingual,6, 7, 8 but some autosomal dominant transmission mutations in GJB2 gene tend to be postlingual, progressive, and associated with some other syndromic phenotype9, 10, 11 (http://davinci.crg.es/ deafness). Currently, the most common GJB2 mutations were identified, including c.35delG, c.167del16, and c.235delC. These mutations have been demonstrated to responsible for the hearing loss identified among different populations.12, 13, 14, 15, 16, 17

In this study, we report that a novel compound heterozygote mutation in the GJB2 gene has been identified successfully. Furthermore, its pathogenic impacts on hearing function were also evaluated via bioinformatic structural analysis and subcellular localization in HEK293 and primary SGN cells.

2. MATERIALS AND METHODS

2.1. Clinical analysis

All individuals in this family were examined by auditory brainstem responses (ABRs) testing. Their medical histories were obtained by using of a questionnaire according to the aspects of this condition followed: medication, noise exposure, pathological changes in the ear, and other relevant clinical manifestations. The study was approved by Research Ethics Committee of Xuzhou medical University.

2.2. Auditory brainstem responses (ABRs) evaluations

ABRs were tested by using of the IHS System (Intelligent Hearing System, Smart EP, FL, USA) and ER‐3A insert earphone. The button‐shaped electrode was placed on the skin: the recording electrode was placed on the middle of the top of the forehead, the reference electrodes were placed on both earlobes (or mastoid), and the ground electrode was placed at the nasion. The 100‐μs clicks were presented at a repetition rate of 19.3/s at a bandpass of 100‐3000 Hz. The sweeps were typically 1.024 seconds. The maximum output level for the clicks was 100 dB nHL. The threshold of ABR was defined as the minimum sound intensity to evoke repeatable wave V. When there was no response or absent waveforms at 90 dB nHL, the subjects were tested with 95 and 100 dB nHL.

2.3. Genotyping analysis

We isolated Genomic DNA from EDTA‐anticoagulated blood samples of the five siblings and their parents by using blood DNA Isolation kits (Tiangen, china). Nine hotspot mutations of deafness genes present in Chinese populations were screened by using a universal array approach, termed a multiplex allele‐specific PCR‐based universal array (ASPUA), as previously described.18, 19 Each blood DNA samples were further examined by sequencing.

2.4. Molecular dynamics simulation

Molecular dynamics (MD) simulation of the two systems was carried out using AMBER 12 molecular simulation package with the standard AMBER03 force field (ff03).20 Each system was immersed in a rectangular box of TIP3P water molecules, which extended 12 Å away from any solute atoms. To neutralize the systems, appropriate numbers of Na⁺ were added. Each system was firstly energy minimizing using the sander program in AMBER 12 via the following three steps: the water molecules and ions were firstly optimized by restraining the protein (2500 cycles of steepest descent minimization and 2500 cycles of conjugated gradient minimization); Secondly, the side chains of the protein were relaxed by restraining all backbone atoms (2500 cycles of steepest descent minimization and 2500 cycles of conjugated gradient minimization); finally, the whole system was relaxed without any restrain (5000 cycles of steepest descent minimization and 5000 cycles of conjugated gradient minimization). After energy minimization, each system was gradually heated in the NVT ensemble from 0 to 300 K within 60 ps. For the production run, a 10‐ns MD simulation with a time step of 2.0 fs was performed under a constant temperature of 300 K. During MD simulation, the SHAKE procedure and particle mesh Ewald (PME) were employed to constrain all bonds involving at least one hydrogen atom and deal with the long‐range electrostatic interactions, respectively.

2.5. Cells culture

HEK293 cells were obtained from ATCC (Manassas, VA, USA). 293T cells were maintained in Dulbecco's minimum essential medium (DMEM, Gibco), supplemented with 10% FBS, 100 mg/mL penicillin and 100 mg/mL streptomycin in a humidified atmosphere containing 5% CO₂ at 37°C. Primary SGN cells preparations have been done as previous study.21

2.6. Localization analysis of mutant CX26 protein in vitro

Normal human GJB2 sequence fragments, GJB2 c.257C>G and GJB2 c.176del16 mutant sequences were obtained from the blood genomic DNAs of all members in the family carrying this compound heterozygous mutations. They were then subcloned into the pWPXLD vectors fused with the EGFP sequence using specific primers TTTATGGATTGGGGCACGCT‐3′ and reverse, 5′‐TCCCTAACTGGCTTTT TTGA C‐3′. The correct clones were identified with DNA sequencing. The lenti‐virus encoding WT or mutated CX26 tagged with fluorescent protein markers were transduced into HEK293 cell line (ATCC) and primary SGN cells to test whether functional gap junctions can be formed. The intracellular localization of WT and mutant CX26 proteins were observed through a fluorescent microscope 48‐72 hours after transduction.

3. RESULTS

3.1. Identification of a novel compound heterozygous mutations in GJB2 c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18)

From ABR testing, four of the five children in the studied family showed a profound hearing loss. No significant hearing loss was identified in their parents (Figure 1A,B). Through the following molecular screening by ASPUA, none of the potential hotspot mutations in SLC26A4, GJB3, and mtDNA RNR1 genes were identified as a determinant of their hearing loss, but the GJB2 c.176 del 16, a prevalent heterozygous mutation in the GJB2 gene, was identified in the four children and their mother (Figure 1C). However, these results were not consistent with the family phenotype, suggesting that there are additional mutations in GJB2 gene awaiting for further identification.

Figure 1.

Compound heterozygous mutations c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) and hearing impairment in a Chinese family. A: The hereditary maps of the family. Four of five filial generation individuals were found to have a profound hearing loss, and one individual of five filial generations showed a normal hearing. Both of mother and father present a normal hearing. B: A profound hearing loss detected in the four individuals in filial generation in this family. Blue and red lines indicated absent ABR waveforms evoked even using high intensity sound stimuli (90, 95 and 100 dB nHL). C: The 9 common‐site screening of a universal array (Allele‐specific PCR‐based universal array, ASPUA), the white frame showing the WT or 176 del 16 site mutation in GJB2 gene in the DNA samples obtained from this family membrane. D: 176 del 16 site mutation and 257C>G, a rare site mutations in GJB2 gene, was further identified in the same samples of this family using GJB2 coding exons sequencing

To explore a possible causive mutations, we sequenced the coding exon of GJB2 genes of each member in this family and found a compound heterozygote mutations in GJB2 c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) in the four children who showed a profound hearing loss (Figure 1D). In comparison, the father and mother showed GJB2‐WT/GJB2 c.257C>G (p.T86R) and GJB2‐WT/c.176del16 (p.G59A fs*18), respectively.

3.2. Compound heterozygous mutations in GJB2 genec.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) disrupt formation of CX26 hexamer

Here, a 3D model for the structural analysis of WT/Mut CX26 proteins was built to illustrate the possible mechanism underlying the mutant CX26 induce hearing impairment. We found that the functional gap junction structure, also named as hexamer, is composed of six WT‐CX26 mono molecules (Figure 2A). As we know, normal formation of hexamer depends on hydrophobic interaction between adjacent CX26 molecules. However, the MD simulation revealed that the Mut‐CX26 protein (p.T86R) alters spatial structure of the dimerization, especially in the residues of three amino acids, namely Leu27, Ile185, and Leu187 due to that the side chain residues of arginine in the 86^th mutated CX26 (p.T86R) are longer than threonine in the same site of WT‐CX26 (Figure 2B). Moreover, such alterations of spatial structure was measured by the distance between the core carbon atom of Thr or Arg and the core carbon atom of corresponding amino acids (Leu27, Ile185 and Leu187) in adjacent CX26 molecules (Figure 2C). Additionally, MD simulation showed that the mutation of CX26 (p.G59A fs*18) which is translated from GJB2 c.176 del 16 cause the formation of truncated mutated proteins, and also suggested a disruption of normal dimerization of adjacent CX26 molecules (Figure 3).

Figure 2.

3D structures analysis of potential pathogenic mechanism underlying mutant CX26 proteins using MD simulation. A: Hemichannels structure of normal gap junctions composed of six CX26 mono molecules. B: Molecular structure of WT (green) and c.257C>G (p.T86R) (yellow) mutated CX26 proteins. Enlarged area (black circle indicated) highlights the major changes in Leu27, Ile185, and Leu187 in interface of adjacent CX26 mono molecules. Red arrow shows that once the 86th amino acids are changed to be Arg, their side chains will consequently alter and tend to be longer than that of structure which 86th amino acids are Thr. C: Evaluation of the distance between core carbon atom of Thr or Arg and crew carbon atom of corresponding amino acids (Leu27,Ile185 and Leu187) in adjacent CX26 molecules, suggesting a structural alterations between WT CX26 (left) and the c.257C>G (p.T86R) mutated CX26 protein (right)

Figure 3.

A truncated CX26 protein caused by the c.176del16 GJB2 mutation inhibits the formation of the hemichannels structure of gap junctions. Enlarged area of the red circle (left picture) highlights the truncated CX26 protein (right picture, red long arrow indicated), suggesting that the format dimer structure is unable to form due to the binding structures have been deleted (gray areas indicated)

3.3. GJB2 c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) alter membrane locatization of CX26 protein

In this study, HEK 293 cells were infected by lenti‐virus carrying GFP tagged WT or Mut (257C>G or 176del16) GJB2 respectively, which allowed us to track the intracellular location of CX26. WT‐CX26 has been found to form a massive focis located in cell membrane (Figure 4). Differently, the CX26 p.T86R (GJB2 c.257C>G) and CX26 p.G59Afs*18 (GJB2 c.176del16) mutants have not been identified in the cell membrane which suggested that these mutations may not involve in the formation of gap junctions.

Figure 4.

Mutated CX26‐GFP proteins cause a disruption of cellular membrane localization. Subcellular localization analysis of WT and mutated CX26‐GFP proteins in 293T and primary mouse cochlear spiral ganglion cells (SGCs). White arrows in the WT pictures (left, 293T and SGCs) show WT‐CX26‐GFP florescent foci normally accumulated in the surface of cell members, suggesting that a normal gap junction structures were formed in these cells. In contrast, no any normal structures can be found in c.176 del 16 and c.257 G>C mutated ‐CX26‐GFP (white arrow head indicated), suggesting that normal structural formation has been disrupted by mutated CX26‐GFP proteins. Bar = 10 μm

Then similar phenomena were also found in primary SGN cells. When the GFP‐tagged CX26‐WT or mutant proteins were introduced into primary SGN cells via a lenti‐virus system, the same gap junction plaques was assembled by CX26‐WT proteins. However, neither CX26 p.G59Afs*18 (GJB2 c.176del16) nor CX26 p.T86R (GJB2 c.257C>G) mutant proteins can assemble the gap junction plaques in SGN cells (Figure 4). Therefore, we conclude that the CX26 p.T86R (GJB2 c.257C>G)/CX26 p.G59Afs*18 (GJB2 c.176del16) mutation might be highly responsible for the profound hearing loss in the four individuals of this family due to the cells membrane localization of each mutated CX26 protein has been inhibited.

4. DISCUSSION

Our study demonstrates that compound heterozygous mutation c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) in the GJB2 gene is novel, further, the mutation is identified in a Chinese family in which 4 siblings are confirmed to have profound hearing loss carrying this compound heterozygous mutation. It is not consistent with their parents who are diagnosed with normal hearing which indicating an autosomal recessive inheritance pattern. Moreover, our study demonstrates that the compound mutations may lead to the profound hearing loss occurred in this family. In addition, cell study and analysis of molecular dynamics simulation reveal significant influence that compound heterozygous mutation CX26 p.T86R (GJB2 c.257C>G)/CX26 p.G59Afs*18 (GJB2 c.176del16) could lead to an alterations of subcellular localization which has been confirmed corresponding to the susceptibility of hearing disorder.

Currently, mutations in GJB2 are considered as one of the major causes of nonsyndromic hearing loss following by a dominant or recessive inheritance pattern.17, 18, 19 The effects of these mutations on the function of CX26 protein could be various and depend on the different positions and types of amino acid substitution. Previous studies proposed that hemichannels structure (hexmer) is crucial for the smooth function of ion channels in the WT‐CX26 protein. Considering that GJB2 c.257C>G (p.T86R) may serve a rare mutation in the second transmembrane domain of the CX26 protein and could be able to convert an uncharged amino acid (threonine) at codon 86 into a positively charged structure (arginine), and logically, it is possible that the rare mutation may cause a functional null protein.19 Our study showed that the steric hindrance became greater as a result that threonine has been replaced by arginine (Figure 2), the evidence indicates that the c.257C>G (p.T86R) mutation in GJB2 gene could impede the dimerization of the CX26 protein. The disruption of the dimer of WT‐CX26 is made up of hydrophobic amino acids, the Arg, which is normally in the 86^th amino acid, tend to form a longer side chains and greater polarity compared with the Thr in the same site. The alterations are most likely to disrupt normal formation of hexmers.

GJB2 c.176del16 (p.G59A fs*18), a frameshift mutation, located in the second topological domain and can truncate the second transmembrane domain of CX26 protein. Once the mutations occur, a stop codon (TGA) will be introduced into the GJB2 gene, which leading to the formation of a truncated CX26 protein comprising 75 amino acids as a result. The abnormal protein could significantly inhibit the formation of hexmers, this altered process is successfully mimicked by using of MD simulation analysis in this study (Figure 3).

It has been fully proposed that the mutation of GJB2 c.257C>G is associated with hearing loss, either homozygously2, 22, 23 or as part of such compound heterozygous mutations as c.235delC and c.299delAT17 As a common hearing loss inducing mutation, c.176 del.16 is fully explored due to that it is one of the most common hearing loss inducing mutation in GJB2. Although GJB2 c.257C>G/c.176 del.16 mutations could be theoretically to associate with hearing loss, the compound heterozygous mutations c.257C>G and c.176 del.16 in GJB2 gene are never identified and proposed before this study. Moreover, it is the first report to show how the compound heterozygous mutations disrupt the function of CX26 protein by using of both investigations of cellular biology and MD simulations; therefore, this study still presents a clinical value for further genetic screening of hearing disorder.

5. CONCLUSION

Here, we demonstrate that the c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) compound mutations in the GJB2 gene are novel, and the mutations are identified in a Chinese family. Furthermore, we demonstrate that the mutations are pathogenicity and may highly lead to a profound hearing loss. Thus, our study could provide a clinical value for genetic screening in further application.

Shi X, Zhang Y, Qiu S, et al. A Novel GJB2 compound heterozygous mutation c.257C>G (p.T86R)/c.176del16 (p.G59A fs*18) causes sensorineural hearing loss in a Chinese family. J Clin Lab Anal. 2018;32:e22444 10.1002/jcla.22444