Advances in Understanding the Molecular Dynamics of Autosomal Dominant Auditory Neuropathy: Unveiling a Novel DIAPH3 Gene Variant Associated With Sensorineural Hearing Loss and Bilateral Vestibular Aqueduct Enlargement
Article information
Abstract
Auditory neuropathy is characterized by abnormal neural conduction in the auditory pathway despite normal outer hair cell function, exhibiting substantial genetic heterogeneity and phenotypic variability. We report the case of a 29-year-old male patient with hearing loss, bilateral enlargement of the vestibular aqueduct (EVA), and vestibular system dysfunction. Based on these features, which are tipically indicative of Pendred syndrome, a molecular investigation including the SLC26A4 gene was performed. This analysis identified a novel heterozygous missense variant, c.411A>C, in exon 4 of the DIAPH3 gene, likely associated with autosomal dominant auditory neuropathy. This point mutation results in substituting glutamic acid with aspartic acid at position 137 p.(Glu137Asp), in the functional Rho-GTPase-binding domain of the DIAPH3 protein. Segregation analysis of the parents and two siblings of the proband revealed the variant’s de novo origin. According to the American College of Medical Genetics and Genomics criteria, this finding underscores the need to reclassify the variant as likely pathogenic. This is the first evidence of an association between a DIAPH3 variant and hearing loss coupled with bilateral EVA and vestibular system dysfunction. This finding contributes to a better understanding of the phenotypic complexity of disorders grouped within the auditory neuropathy spectrum.
Introduction
Sensorineural deafness represents a complex and heterogeneous clinical condition that can arise from multiple genetic and environmental causes. In particular, auditory neuropathy spectrum disorder (ANSD) represents a subset of sensorineural hearing loss that affects the transmission of sound signals from the inner ear to the brain by a disruption in the normal synchronization of auditory nerve impulses [1]. This condition can lead to difficulties in understanding speech, particularly in noisy environments, and can affect a person’s ability to process and interpret sound accurately [2]. The distinctive features of ANSD include a normal functioning of the outer hair cells (OHCs) in the cochlea, coupled with impaired function of the auditory nerve responsible for transmitting electrical signals to the brain [2].
In this context, we encounter the case of a 29-year-old male patient, affected by sensorineural deafness, whose clinical picture has proven to be extraordinarily challenging. Indeed, the clinical investigation revealed the presence of bilateral enlargement of the vestibular aqueducts (EVA), a peculiarity that has drawn attention to a possible association with Pendred syndrome. This latter is an autosomal recessive genetic disorder, which is commonly characterized by sensorineural deafness with anomaly in thyroid structure, congenital EVA alterations and variable vestibular system dysfunction [3]. The syndrome is typically associated with pathogenic variants in the SLC26A4 gene, involved in ion transport across the cell membrane and iodine homeostasis [4]. However, the molecular investigation of the proband unveiled an unexpected twist consisting of the c.411A>C, p.(Glu137Asp), missense variant in the DIAPH3 gene which codes for a protein whose altered expression has been associated with autosomal dominant auditory neuropathy (AUNA1) (OMIM#609129). This form of deafness preserves the normal function of OHCs, which partially maintains the ability to amplify sounds and mitigates some negative effects on auditory perception [5].
The present investigation aims to elucidate the likely pathogenic role of this missense variant and paving the way for a deeper understanding of how variants within the DIAPH3 gene may be involved in the development of challenging clinical phenotypes.
Case Report
The present investigation focused on a 29-year-old male patient who underwent a clinical evaluation for a bilateral hearing impairment that first appeared in early childhood. Since the age of six, the patient has been utilizing bilateral hearing aids and concurrently engaging in speech therapy. A comprehensive pedigree analysis did not identify any occurrences of hearing loss among other family members, and no parental consanguinity was reported.
Audiometric analysis, performed at the age of 5, revealed a bilateral mild sensorineural hearing loss at low frequencies, moderate loss at mid frequencies, and severe impairment at high frequencies, more pronounced on the right side, which was associated to mild conductive deafness (Fig. 1A). Molecular investigation of GJB2 gene yielded normal results. At the age of 26, the hearing loss was stable for low frequencies, while a mid-frequency loss was apparent on the left side. Additionally, there was a decline in hearing loss at high frequencies (from 80 dB to 110 dB), specifically on the right side (Fig. 1B). Simultaneously, there was also a conductive hearing loss, primarily affecting high frequencies, attributed to a previously tympanic membrane perforation. After 3 years, there was further worsening of sensorineural hearing loss at high frequencies, especially on the right side (Fig. 1C).
Audiometric assessment of the proband at the ages of 5 (A), 26 (B), and 29 (C), years. O right airway, X left airway, > right bone conduction, < left bone conduction, □ without hearing aid, ■ with hearing aid. Free-field audiometry at the ages of 26 (D) and 29 (E) years, respectively. In Fig. 1F and G, a clear decline in hearing improvement resulting from the use of hearing aids can be observed, especially at higher frequencies.
Free-field audiometry performed at the age of 26 demonstrated that the patient could accurately perceive 60% of bisyllabic words at the 70 dB threshold (Fig. 1D), while at the age of 29, his correct perception of bisyllabic words at the same threshold was decreased to 30% (Fig. 1E). The use of hearing aids ensured a progressive slow decline in hearing loss, especially at high frequencies (Fig. 1F and G). In summary, the latest clinical assessment, compared to previous evaluations, revealed stability at low frequencies and progressive worsening at both mid and high frequencies. The thyroid ultrasound, which is often altered in Pendred syndrome, showed normal size, morphology, and structure. Auditory brainstem response (ABR) test demonstrated a destructured trace, while the distortion product otoacustic emissions (DPOAE) resulted in bilaterally evoked responses (Figs. 2 and 3).
Auditory brainstem response (ABR) test demonstrated a destructured trace with typical waves I-V not clearly identifiable.
Distortion product otoacustic emissions (DPOAE) resulted in bilaterally evoked responses.
CT scan revealed regular external auditory canals and tympanic membranes with tympanic cavities normally aerated as well as the middle ear ossicular chain. The bony structures of the inner ear and the bony canal of the facial nerve were normal bilaterally, despite the presence of bilateral EVA and a right bony canal of the chorda tympani thin and not recognizable in the most distal tract (Fig. 4).
CT scan showing bilateral engagement of the vestibular aqueduct (EVA) at the posterior and middle thirds with the anterior third of the normal diameter.
MRI showed a normal ventricular system and subarachnoid spaces without alterations of the gray and white matter. The midbrain axis was regular. The midline structures were aligned and a specific study on acoustic-facial bundles did not show any alterations.
The head shaking test (HST) and mastoid vibration did not elicit spontaneous or positional nystagmus, as observed through video-oculography. Bilaterally negative results were also noted in the video head impulse test (vHIT), whereas caloric testing using the Hallpike method with electronystagmography registration revealed bilateral vestibular hyporeflexia.
Given the features of the hearing loss, such as the bilaterality, the early occurrence without infections, and the contextual presence of bilateral EVA and vestibular system dysfunction, the molecular approach was principally focused on a panel of 59 genes (18 autosomal dominant, 37 autosomal recessive, and 4 X-linked genes) preferentially linked to sensorineural hearing loss as followed indicated: ACTG1, CCDC50, CEACAM16, CLDN14, COCH, COL11A2, COL4A6, DFNA5, DFNB59, DIAPH3, ESPN, EYA4, GIPC3, GJB2, GJB3, GJB6, GRHL2, GRXCR1, HGF, HOMER2, KCNQ4, LEDGF, MIR96, MYH14, MYH9, MYO15, MYO6, OSBPL2, OTOA, OTOF, OTOG, OTOGL, P2RX2, POU3F4, POU4F3, PRPS1, PTPRQ, RDX, SCL17A8, SLC26A4, SLC26A5, SMPX, STRC, TBC1D24, TECTA, TMC1, TMPRSS3, TPRN, CDH23, CIB2, CLRN1, GPR98, MYO7A, PDCH15, PDZD7, USH1C, USH1G, USH2A, WHRN.
From this genetic analysis, no likely pathogenic or pathogenic variants were revealed in the SLC26A4 POU3F4, GJB2, TMC1, and MYO6 genes, often associated with EVA syndrome. However, the analysis revealed a heterozygous missense variant, c.411A>C, in exon 4 of the DIAPH3 gene (OMIM#614567) that led to the substitution of an adenine with cytosine at position 411, resulting in the change of glutamic acid with aspartic acid at position 137 p.(Glu137Asp). The segregation analysis conducted on both parents and two siblings of the proband supported its de novo origin (Fig. 5).
Proband pedigree. The segregation study conducted on family members highlighted that the variant was present only in the affected proband, as indicated by the arrowhead.
According to the American College Medical Genetics (ACMG)/Association for Molecular Pathology (AMP) guidelines, this change affected a highly phylogenetically conserved amino acid residue in the functional Rho-GTPase-binding (GBD) domain (PM1) and it was absent in controls from specific databases such as the Exome Sequencing Project, 1000 Genomes Project, or Exome Aggregation Consortium (PM2) [6]. Since the maternity and paternity have not been confirmed, and the phenotype is consistent with the gene but not highly specific, a 0.5 point was assigned (PS2_supporting or PM6_supporting) [7].
This DIAPH3 gene variant was predicted as predominantly deleterious by SIFT (v6.2.0): deleterious (score: 0.00, median: 3.01); PolyPhen2 (v): HDivPred: probably damaging (score 0.992) and HVarPred: probably damaging (score 0.989), whereas it was considered as benign by MutationTaster (v2021): Benign; Tree vote: 30/70 (deleterious/benign); Align GVGD (v2007) Class C0 (GV: 222.52, GD: 11.33); CADD (v1.6): Phred: 26.0, Raw score 3.755266.
REVEL score of 0.460 (PP3, BP4) falling within the range of 0.15–0.7 did not allow applying either the PP3 or the BP4 criterion. Moreover, gain of function (GoF) variants, like the one reported, are less likely than loss of function (LoF) variants to receive the highest REVEL scores, which strongly indicate pathogenicity [8]. From an evolutionary standpoint, the missense variant involves a weakly conserved nucleotide (phyloP: 3.38 [10.9–19.0]), even though the glutamic acid in position 137 is a highly conserved amino acid.
Nonetheless, the final classification criteria (PM1, PM2, and PS2_supporting or PM6_supporting) confirm the likely pathogenicity of this specific variant.
Discussion
This investigation focused on a patient exhibiting a complex clinical phenotype characterized by sensorineural hearing loss coupled with bilateral EVA and vestibular system dysfunction that was strongly oriented toward a Pendred syndrome diagnosis. Unexpectedly, a missense variant in the DIAPH3 gene, resulting in the substitution of glutamic acid with aspartic acid at position 137, c.411A>C, p.(Glu137Asp), was identified.
These findings carry notable significance given that DIAPH3 protein is constituted by several domains that are pivotal for its activation and functionality [5,9]. Specifically, the interaction between the diaphanous inhibitory domain (DID) and the diaphanous autoregulatory domain (DAD) leads to protein’s inactivation, while the interaction of the GTPase binding domain (GBD) with the Rho family proteins induces a conformational shift that releases DIAPH3 from its self-induced inhibition, triggering protein homodimerization critical in the actin cytoskeleton dynamics [5,10].
Previously reported genetic variants in this DIAPH3 gene have been associated with ANUA1 [5], a condition characterized by compromised transmission of nerve impulses between the inner ear and the brain, which may be underdiagnosed in approximately one-third of patients [11]. In particular, the heterozygous G-to-A transition at position -172 and C-to-T transition at position -173 in the 5' UTR of the DIAPH3 gene have been found associated with DIAPH3 protein overexpression which resulted in the childhood development of a sensorineural hearing loss rapidly progressing to profound deafness, with OHC function preserving [5].
In the present clinical case, the c.411A>C gene variant was detected in the exon 4 coding region and substitutes glutamic acid (Glu) with aspartic acid (Asp) in position 137 resulting in a small physicochemical protein difference (Grantham dist: 45 [0–215]) since the both amino acids carrying the same negative charge. Indeed, on based on 3D reconstruction, the glutamic acid substitution, which forms hydrogen bonds with arginine, glycine, and lysine in position 144, 1056, and 140, respectively (Fig. 6), with aspartic acid does not produce significant changes in protein conformation. However, it is worth noting that glutamic acid at position 137 is a highly conserved phylogenetic residue (Fig. 7) located within the GBD domain critical for the interaction with the Rho-GTPase family proteins.
3D view of the DIAPH3 protein region, characterized by the substitution of glutamic acid with aspartic acid at position 137. The hydrogen bonds between glutamic acid with arginine 144 (1), glycine1056 (2), and lysine 140 (3) are shown.
Comparison across species reveals that glutamic acid is a highly conserved amino acid being present in 11 out of 12 species evaluated.
Dysregulation of the Diap3 protein may have important consequences as observed in transgenic mice model where its overexpression has been found to induce progressive increase in auditory thresholds, severe modifications in the IHC stereocilia, which appeared elongated, thickened, and reduced in numerosity but with OHC function preserving despite these structural and functional alterations [12]. Moreover, the introduction of a constitutively active form of Diaphanous protein in the auditory organ of Drosophila melanogaster confirmed a substantial reduction in sound-evoked potentials suggesting that an altered Diap3 protein expression/activity may have a notable impact on the structure of sensory cells [5]. In this context, although the missense variant c.411A>C, revealed in the proband, may not support a direct dysregulating effect on the interaction between GBD domain with the Rho-GTPase proteins or on the induction of IHC alterations and stress fiber formation, it could, however, affect protein expression by a genetic side effect.
An intriguing perspective arises when considering the replacement of GAA codon, responsible for encoding glutamic acid, with the GAC codon specific for aspartic acid. This change could have the potential to enhance protein translation speed by a greater availability of distinct tRNA molecules specifically tailored for particular codons [9,13]. Specifically, an elevated cell abundance of tRNA for the aspartic acid codon within IHCs might ultimately lead to an upsurge of DIAPH3 protein expression that, in turn, could contribute to the ANSD occurrence.
To date, only a limited number of DIAPH3 gene variants, often missense, have been identified in probands with sensorineural hearing loss or auditory neuropathy and the majority of these, according to the ACMG/AMP criteria, are classified as variants of uncertain significance (VUS) [9,14,15]. Apart from the well-characterized G-to-A transition at position -172, even for the c.2059del variant, reported as pathogenic in the ClinVar database, no assertion criteria were provided. Although the presence of EVA and vestibular system dysfunction may introduce elements of interpretative confusion regarding the occurrence of sensorineural hearing loss, the de novo origin of the c.411A>C variant, alongside other ACMG/ AMP criteria, suggests the need to reclassifying the variant as a likely pathogenic variant.
In summary, this is the first report in which a likely pathogenic DIAPH3 gene variant was found to be associated with a clinical phenotype closely resembling Pendred syndrome. The implications of this investigation underscore the imperative for further research, aimed at unraveling the intricate relationship between DIAPH3 gene variants and ANSD, thus contributing to a more comprehensive understanding of these genetic associations.
Notes
Ethics Statement
All patients consent for participation and publication of the paper. All procedures performed in studies involving human participants were in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Conflicts of Interest
The authors have no financial conflicts of interest.
Author Contributions
Conceptualization: Gianmaria Miolo. Data curation: Francesco Margiotta, Alessandra Murgia, Lara Della Puppa, Giuseppe Corona. Formal analysis: Alessandra Murgia. Methodology: Gianmaria Miolo. Investigation: Gianmaria Miolo, Francesco Margiotta, Alessandra Murgia. Project administration: Gianmaria Miolo. Software: Giuseppe Corona. Supervision: Gianmaria Miolo, Giuseppe Corona. Validation: all authors. Writing—original draft: Gianmaria Miolo, Giuseppe Corona. Writing—review & editing: all authors. Approval of final manuscript: all authors.
Funding Statement
None
Acknowledgments
The authors would like to thank our patient and his family for their prompt interest and cooperation in this report. Additionally, we thank MED-EL Implant Systems Australasia Pty Ltd for their support.