Originally published online as doi:10.2353/ajpath.2008.080049 on September 11, 2008
Published online before print September 11, 2008
(American Journal of Pathology. 2008;173:1113-1119.)
© 2008 American Society for Investigative Pathology
DOI: 10.2353/ajpath.2008.080049
A Novel Missense Mutation in the Second Extracellular Domain of GJB2, p.Ser183Phe, Causes a Syndrome of Focal Palmoplantar Keratoderma with Deafness
Eugene A. de Zwart-Storm*
,
Michel van Geel*,
Pierre A.F.A. van Neer
,
Peter M. Steijlen*,
Patricia E. Martin and
Maurice A.M. van Steensel*
From the Department of Dermatology,* University Medical Centre Maastricht, Maastricht, The Netherlands; GROW-School for Oncology and Developmental Biology, the Department of Biological and Biomedical Sciences, Glasgow Caledonian University, Glasgow, United Kingdom and the University Maastricht, The Netherlands; the Department of Dermatology, Laurentius Hospital, Roermond, The Netherlands
 |
Abstract
|
|---|
Gap junctions, which consist of connexins, are intercellular channels that mediate rapid intercellular communication. In the skin, connexins are involved in the regulation of epidermal growth and differentiation. GJB2 encodes connexin26, which is an important skin-expressed gap junction protein. Mutations in GJB2 cause a wide variety of unique disorders, but despite extensive research, their mechanisms of action are poorly understood. The identification of novel diseases caused by mutations in GJB2 may help to illuminate the genotype-phenotype correlation and elucidate the function of different regions of the protein. Here, we report the first account of a family with a GJB2 missense mutation in the second extracellular domain (p.Ser183Phe) that causes skin abnormalities in addition to sensorineural hearing loss. Using fluorescent connexin26-EGFP fusion proteins, we showed that the mutation induces a partial protein transport defect that cannot be rescued by wild-type protein. Dye-transfer experiments using a parachute assay revealed channel functionality. Although p.Ser183Phe affects the second extracellular domain, mutations in the first extracellular domain also lead to focal palmoplantar keratoderma and likewise perturb protein transport in a dominant-negative manner. Therefore, we hypothesize that focal palmoplantar keratoderma in gap junction skin disease may be specifically associated with connexin trafficking defects as well as with mutations affecting its extracellular domains, thus broadening the spectrum of GJB2-associated diseases.
Connexins are small transmembrane proteins that belong to an extensive protein family found in most metazoans.1
Six connexins can oligomerize to form a hemichannel (connexon) in the cell membrane. Two opposite channels connect, or dock to form an aqueous channel that is called a gap junction and facilitates intercellular communication. In the skin, gap junction intercellular communication is essential for regulation of its growth and differentiation.2
Many different skin diseases of widely varying severity are associated with a large number of gap-junction gene mutations. In the case of the GJB2 gene, which codes for connexin26, the list of mutations is still growing. Sensoneurinal deafness is common to all GJB2-associated phenotypes, but the skin symptoms are quite distinct and specific to the mutation.
Why the different mutations cause such widely varying cutaneous manifestations is poorly understood. Some mutations can form heteromeric connexins with wild-type counterparts causing a gain of function of the mutant connexin, whereas other mutations give rise to transport defects, which may affect gap junction stoichiometry and lead to a loss of function.1-3
How these functional consequences relate to the skin and hair abnormalities is unclear, although mutations associated with skin symptoms, in particular palmoplantar keratoderma, so far seem to cluster in the first cytoplasmic and extracellular loop of connexin26 (for an overview see de Zwart-Storm et al4
). Thus, one might infer that these domains are of particular importance for gap junction function in the skin.
In the present study we report on the first missense mutation in the second extracellular domain (E2) of GJB2 that causes cutaneous symptoms in addition to hearing impairment. Functional analysis of the mutation suggests that it leads to a partial protein transport defect that is dominant-negative with respect to the wild-type protein. Dye transfer assays show permeability of mutant gap junction channels to the fluorescent dye calcein. With these new data, the spectrum of GJB2-associated disease continues to broaden.
|
Case Reports
|
|---|
The proposita, a 43-year-old woman of Dutch descent, consulted the Department of Dermatology of the Laurentius Hospital (Roermond, The Netherlands) for a long-standing complaint of palmoplantar keratoderma. In addition, she had sensory hearing loss with preferential high-tone loss. The patient’s youngest daughter had sensory hearing loss that had been diagnosed at 5 years of age and was said to have similar skin changes (Figure 1A)
. As far as our patient knew, no other members of the family were deaf or had skin abnormalities. The proposita and her daughter were available for physical examination. In the proposita, flat, hyperkeratotic translucent plaques consisting of confluent hyperkeratotic papules each with a diameter of a few millimeters were present on the palms, in particular on the border of the thenar and wrist (Figure 2, A–C)
. On the feet we likewise saw hyperkeratotic plaques (Figure 2, D–F)
. Most were seen in areas exposed to mechanical stress, such as heels, balls of feet, knuckles, and wrists. At the proximal interphalangeal joints of the hands, we saw a more pronounced hyperkeratosis, sometimes in the form of knuckle pads. Knees and elbows were not involved, and other skin abnormalities were not found. In particular, there were no hypotrichosis, mucositis, nail abnormalities, or pseudo-ainhum. Hair and teeth were normal. Audiological examination showed preferential sensory high-tone hearing loss (Figure 3)
. Similar but less pronounced skin alterations and high-tone sensory hearing loss were found in the affected daughter. The coincidence of hearing impairment and palmoplantar hyperkeratosis prompted us to look for mutations in the GJB2 gene.
View larger version (29K):
[in this window]
[in a new window]
|
Figure 1. Presence of the mutation within the index family. A: Pedigree of the family, with the filled symbols representing the affected members and the proband indicated by an arrow. B: Restriction analysis with NciI; mutation present in the mother and daughter (lanes 1 and 3, respectively; 535, 440, 318, and 122 bp), not in the father or a son (lanes 2 and 4, respectively; 535, 318, and 122 bp).
|
|
View larger version (109K):
[in this window]
[in a new window]
|
Figure 2. Clinical phenotype of the index patient. A: Diffuse, flat, hyperkeratotic translucent plaques were present on the palms, in particular on the border of the thenar and wrist. B: Close-up view of the mild palmar hyperkeratosis. C: Hyperkeratotic plaques were seen in areas exposed to mechanical stress, such as heels, balls of feet, knuckles, and wrists, sometimes in the form of knuckle pads. D: Hyperkeratotic plaques (knuckle pads) on the toes. E: Mild diffuse plantar hyperkeratosis. F: Close-up of the plantar surface, where areas with hyperkeratosis are clearly visible. Note that the plaques seem to actually be formed by confluence of smaller hyperkeratotic papules.
|
|
View larger version (65K):
[in this window]
[in a new window]
|
Figure 3. Pronounced sensory hearing loss, in particular for higher frequencies (vertical: decibels; horizontal: the frequency in KHz). Air conduction: cross/round marks; bone conduction: triangles, arrows pointing downwards indicate the loudest tone was not heard). The curves of a normal-hearing person are almost horizontal at  10 to 15 decibels.
|
|
|
Materials and Methods
|
|---|
Informed consent was obtained from the patient and her family. DNA was isolated from peripheral blood leukocytes using a simple salt precipitation method.5
GJB2 Mutation Analysis
The GJB2 gene was amplified by standard polymerase chain reaction (PCR) using primers Cx26F 5'-GCATGCTTGCTTACCCAGACTC-3' and Cx26R 5'-AGGGGAGCAGAGCTCCATTG-3'. The gene was analyzed using the PCR primers and the sequencing primers: Cx26FS 5'-CAGAAGGTCCGCATCGAAGG-3', Cx26RS 5'-GCTTCGAAGATGACCCGGAAG-3', using the BigDye Deoxy terminator system and an ABI 3100 capillary sequencer (Applied Biosystems, Foster City, CA). The mutation eliminates an NciI (New England Biolabs, Frankfurt am Main, Germany) restriction site (the wild type contains two NciI sites). With this enzyme, we checked the patient’s family and 100 unaffected controls by restriction analysis.
GJB2 Cloning Strategy
The wild-type GJB2/EGFP-fusion construct coding for 54 kDa was produced as previously described (GJB2, 26.2 kDa).4
The GJB2 c.548C>T (p.Ser183Phe) mutation was introduced into the wild-type GJB2/EGFP construct with the Quick-Change site-directed mutagenesis (SDM) kit (Stratagene, La Jolla, CA) according to the manufacturer’s recommendations. Clones were sequenced to confirm that they contained the correct insert and mutation. The complete insert was sequenced to verify if no additional mutations had occurred. Plasmid DNA for transfection was isolated using Nucleobond PC 100 (Machery-Nagel GmbH & Co. KG, Düren, Germany).
Cell Culture
HeLa Ohio wild-type (wt) cells, that do not endogenously express Cx26, Cx30, or Cx43 and are communication-deficient, were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% v/v fetal bovine serum, 100 µg/ml penicillin/streptomycin, and 2 mmol/L L-glutamine (cDMEM; Lonza Biologics, Cambridge, UK). HeLa cells stably expressing Cx26 were maintained in cDMEM supplemented with 0.5 µg/ml puromycin (Sigma-Aldrich, Gillingham, UK) selective antibiotic.6
Immunofluorescence
Wild-type or Cx26-expressing HeLa Ohio cells (1 x 105 cells) were plated onto 16-mm2 glass coverslips in a 12-well plate. After 24 hours, at 70% confluency, cells were transfected with 0.5 µg of mutant p.Ser183Phe-GJB2/EGFP fusion construct or GJB2/EGFP using Lipofectamine 2000 (Invitrogen Ltd., Paisley, UK) according to the manufacturer’s protocol. After 5 hours the medium was replaced and 24 hours later the cells were fixed with ice-cold methanol (–20°C).7
Briefly, cells were permeabilized with phosphate-buffered saline (PBS) + 0.1% Triton X-100 (v/v), blocked with 5% (w/v) powdered milk-PBS solution followed by staining with appropriate primary antibody for 1 hour at 37°C. Connexin 26wt localization was assessed using a mouse anti-Cx26 (1:50 dilution, Invitrogen). Cells were then washed in PBS and incubated with anti-mouse secondary antibody conjugated to Alexa 594 as previously described.8
Finally, cell nuclei were counterstained with 4,6-diamidino-2-phenylindole (10 ng/ml, w/v, in PBS). Cells with intermediate expression levels of the mutant protein were selected for study.
Parachute Assay
A co-culture system was used to examine the ability of p.Ser183Phe-GJB2/EGFP to form functional channels. Briefly, 24 hours after transfection of wild-type HeLa cells with p.Ser183Phe-GJB2/EGFP or GJB2/EGFP, these donor cells were loaded with the fluorescent tracer calcein AM (2.5 µmol/L, Mw 623 Da, charge –4) and the cell membrane marker CM-DiI (Mw 1053 Da, 1 µg/µl) as previously described.9
HeLa cells stably expressing wild-type Cx26 (acceptor cells, 1 x 105) were seeded onto 24-mm2 cover chambers. Donor cells were subsequently washed in PBS, trypsinized, resuspended, and centrifuged at 1500 rpm for 5 minutes before final resuspension in cDMEM and seeding onto the acceptor cells at a ratio of 1 donor to 50 acceptor cells. Co-cultured cells were then incubated for a minimum of 4 hours at 37°C to allow the donor cells to settle onto the acceptor cells and gap junctions to form. After 4 hours cells were examined on a confocal microscope (Zeiss, Oberkochen, Germany) using dual wavelengths to examine transfer of calcein from the CM-DiI loaded donor cells to the acceptor cells.
Microscopy
Fluorescent samples were viewed using a Zeiss Axiovert 200 confocal microscope linked to a LSM510 META laser-scanning system. Images were acquired under similar conditions and magnifications using either an argon (488-nm excitation, to detect calcein and GFP) or a HeNe laser (543-nm excitation, to detect CM-DiI and Alexa 594) as appropriate. Optical sections were recorded in the Kalman filtering mode using four scans for each picture. Images were further processed using Adobe Photoshop software (Adobe, San Jose, CA).
|
Results
|
|---|
In the patient and her daughter (Figure 1)
, we found a heterozygous c.548C>T transition at codon 183 of the GJB2 gene, resulting in a substitution of a serine by a phenylalanine (p.Ser183Phe, Figure 4
). The mutation was not found in unaffected family members or in 100 unrelated controls from the Dutch population, using restriction analysis (Figure 2B)
, nor was the mutation found in the SNP database (NCBI dbSNP Build 128).
View larger version (23K):
[in this window]
[in a new window]
|
Figure 4. GJB2 sequence chromatograms of an unaffected family member (Wt) and a patient (Mt) at the mutated location (arrow).
|
|
When HeLa Ohio cells were transiently transfected with GJB2/EGFP, intercellular gap junction plaques were seen at points of cell-to-cell contact with limited intracellular staining (Figure 5A)
. By contrast, in HeLa Ohio cells transiently transfected with p.Ser183Phe-GJB2/EGFP, the mutant fusion protein had a predominantly cytoplasmic localization, where it accumulated diffusely (Figure 5B)
. Brighter fluorescence was seen in the endoplasmic reticulum. Some punctate fluorescence, suggestive of membrane localization was also present, as were small intercellular gap junction plaques (Figure 5B)
. The cellular localization of p.Ser183Phe-GJB2/EGFP was not significantly altered after transfection into HeLa Ohio cells stably expressing Cx26, suggesting a dominant-negative effect of the mutation, although almost no protein was retained in the endoplasmic reticulum (Figure 5D)
. A control experiment showed that GJB2/EGFP and wild-type Cx26 detected with a Cx26-specific monoclonal antibody (red) co-localized at points of cell-to-cell contact in gap junction plaques in HeLa Ohio cells stably transfected with Cx26wt (Figure 5C)
.
View larger version (48K):
[in this window]
[in a new window]
|
Figure 5. Results of GJB2/EGFP fusion protein (green: Cx26wt/mt, red: stably expressed Cx26wt) detection in HeLa Ohio cells. A: Gap junction plaque formation by wild-type GJB2/EGFP fusion protein (arrows). B: Some gap junction plaque formation is visible with p.Ser183Phe-GJB2/EGFP mutant fusion protein (arrows), but most of the protein accumulates in the cytoplasm and endoplasmic reticulum. C: Co-localization of stably expressed Cx26wt (red) and transfected Cx26wt (green) at the membrane and in intercellular gap junction plaques (merge = yellow, arrow gap junction plague). D: There is no rescue of the mutant protein with Cx26wt; its primarily cytoplasmatic distribution is suggestive for ER localization. Note that most of the wild-type protein co-localizes with the mutant, suggesting a dominant-negative effect on transport of the latter (arrows, gap junction plaques).
|
|
To determine whether this novel Cx26 mutation formed functional gap junction channels, we performed a parachute assay and showed that HeLa Ohio cells, transiently transfected with GJB2/EGFP or p.Ser183Phe-GJB2/EGFP, could transfer the tracer dye calcein (Mw 623 Da, charge –4) to a monolayer of HeLa Ohio cells stably expressing Cx26wt, suggesting that channels containing mutant protein are at least partly functional (Figure 6, A and B)
. Control, nontransfected wild-type HeLa Ohio cells were unable to communicate with the stably Cx26wt-expressing HeLa cells (Figure 6C)
.
View larger version (17K):
[in this window]
[in a new window]
|
Figure 6. HeLa Ohio cells (marked with an asterisk) transfected with GJB2/EGFP, p.Ser183Phe-GJB2/EGFP, or untransfected, loaded with CM-DiI (red) and calcein (MW622 Da, charge –4) parachuted onto a monolayer of HeLa Ohio cells stably expressing Cx26wt. A: HeLa Ohio cells transfected with GJB2/EGFP show transfer of calcein to a monolayer of HeLa Ohio cells stably expressing Cx26wt. B: HeLa Ohio cells transfected with p.Ser183Phe-GJB2/EGFP show transfer of calcein to a monolayer of HeLa Ohio cells stably expressing Cx26wt. C: HeLa Ohio cells, untransfected, show no transfer of calcein to a monolayer of HeLa Ohio cells stably expressing Cx26wt.
|
|
|
Discussion
|
|---|
We report a novel missense mutation, p.Ser183Phe, in the second extracellular loop (E2) of connexin26. To the best of our knowledge, this is the first report of an E2 mutation causing skin symptoms. The pathogenesis of the keratoderma is not clear, but our results may offer a clue. We recently reported that a dominant E1 mutation, p.His73Arg, associated with a quite similar palmoplantar keratoderma, causes defective protein transport.4
We therefore looked for a similar effect, which we confirmed in this study. We also demonstrated that the mutant protein cannot be completely rescued by its wild-type counterpart, although when co-expressed there is less accumulation of mutant protein in the endoplasmatic reticulum. The parachute assay demonstrates residual channel activity, which might explain the mildness of the phenotype. Mutations causing more severe phenotypes such as p.Asp66His (Vohwinkel’s syndrome) are reported to show less dye transfer than p.Ser183Phe.10
However, other pathogenic mechanisms must also be considered. The p.Ser183Phe mutation occurs in the highly conserved second extracellular loop of Cx26. The serine at position 183 of connexin26 is highly conserved between species and even between connexins, suggesting that this residue is critical for protein function (Figure 7)
. Because the p.Ser183Phe change replaces a polar with an aromatic amino acid, this substitution can be expected to modify local conformation. The two extracellular domains play a major role in connexon assembly and docking.11
Serine 183 is the first amino acid of the Gap27 mimetic peptide, which can inhibit intercellular coupling, lowers channel conductance, and may prevent gap junction assembly.12
Thus, p.Ser183Phe may influence assembly and the interconnexon docking process, which might lead to the skin phenotype. It is of interest to note in this regard, that connexon assembly takes place, at least partly, in the endoplasmic reticulum.11
If the mutation interferes with assembly and correct protein folding, it might interfere with transport to the cell membrane.
View larger version (103K):
[in this window]
[in a new window]
|
Figure 7. Conservation throughout species and within connexins of E2 domain sequence. No Cx preface is Homo sapiens. Abbreviations: Mmul, Macaca mulatta; Bt, Bos taurus; Cf, Canis familiaris; Ec, Equus caballus; Mm, Mus musculus; Md, Monodelphis domestica; Oa, Ornithorhynchus anatinus; Gg, Gallus gallus; Xl, Xenopus laevis.
|
|
Considering that we consistently find dominant-negative transport defects for mutations associated with keratoderma,4,10,13
we hypothesize that disturbed trafficking has an important role in the pathogenesis of gap junction diseases of the skin. One issue with this line of reasoning is that a transport defect per se is not typical for mutations associated with PPK because it has also been observed for mutations causing deafness only.14,15
Furthermore, dominant mutations apparently without skin symptoms have been reported in E2; p.Met163Leu, p.Asp179Asn, and p.Arg184Gln.16-18
Besides P.Met163Leu is, however, not known whether these mutations affect protein transport. As a more parsimonious explanation, we suggest that skin symptoms have been overlooked in these patients. It would be of interest to re-examine them with particular attention to signs of keratoderma.
In conclusion, we describe the first skin phenotype caused by a mutation in the second extracellular domain of connexin26, p.Ser183Phe. Using Cx26/EGFP fusion proteins, we show that the mutation leads to a partial transport defect with the mutant protein accumulating in the cytoplasm. Still, transport to the membrane does occur as evident from the ability to form gap junction plaques. These are (partially) functional with respect to transfer of calcein (Mw 623 Da, charge –4), which might explain the relatively mild phenotype compared to other mutations such as p.Asp66His that show very limited dye transfer. Our results suggest that, in light of the existing genotype-phenotype correlation for GJB2, mutations in E1 and E2 that cause dominant-negative transport defects with residual channel function lead to a focal palmoplantar keratoderma and sensory hearing impairment.
|
Acknowledgements
|
|---|
We thank the family for their kind cooperation.
|
Footnotes
|
|---|
Address reprint requests to E.A. de Zwart-Storm, M.D., Department of Dermatology, University Hospital Maastricht, PO Box 5800, 6202 AZ Maastricht, The Netherlands. E-mail: edz{at}sder.azm.nl
Supported by Geneskin (grant to E.A.d.Z.-S.), The Netherlands Organization for Scientific Research (ZonMw grant 907-00-202 to M.A.M.vS.), Barrier Therapeutics (to M.A.M.vS.), the GROW Research School for Oncology and Developmental Biology (to M.A.M.vS.), the University Hospital Maastricht (to M.A.M.vS.), the Stichting de Drie Lichten (to M.A.M.vS.), the British Skin Foundation (to P.M.), and Medical Research Scotland (to P.M.).
Accepted for publication June 24, 2008.
|
References
|
|---|
- van Steensel MA: Gap junction diseases of the skin. Am J Med Genet C Semin Med Genet 2004, 131C:12-19
- Meçse G, Richard G, White TW: Gap junctions: basic structure and function. J Invest Dermatol 2007, 127:2516-2524[CrossRef][Medline]
- Cottrell GT, Burt JM: Functional consequences of heterogeneous gap junction channel formation and its influence in health and disease. Biochim Biophys Acta 2005, 1711:126-141[Medline]
- de Zwart-Storm EA, Hamm H, Stoevesandt J, Martin P, Steijlen PM, van Geel M, van Steensel MA: A novel missense mutation in GJB2 disturbs gap junction protein transport and causes focal palmoplantar keratoderma with deafness. J Med Genet 2008, 45:161-166[Abstract/Free Full Text]
- Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988, 16:1215[Free Full Text]
- Mesnil M, Krutovskikh V, Piccoli C, Elfgang C, Traub O, Willecke K, Yamasaki H: Negative growth control of HeLa cells by connexin genes: connexin species specificity. Cancer Res 1995, 55:629-639[Abstract/Free Full Text]
- Martin PE, Blundell G, Ahmad S, Errington RJ, Evans WH: Multiple pathways in the trafficking and assembly of connexin 26, 32 and 43 into gap junction intercellular communication channels. J Cell Sci 2001, 114:3845-3855[Abstract/Free Full Text]
- Kandyba EE, Hodgins MB, Martin PE: A murine living skin equivalent amenable to live-cell imaging: analysis of the roles of connexins in the epidermis. J Invest Dermatol 2008, 128:1039-1049[CrossRef][Medline]
- Martin PE, Wall C, Griffith TM: Effects of connexin-mimetic peptides on gap junction functionality and connexin expression in cultured vascular cells. Br J Pharmacol 2005, 144:617-627[CrossRef][Medline]
- Marziano NK, Casalotti SO, Portelli AE, Becker DL, Forge A: Mutations in the gene for connexin 26 (GJB2) that cause hearing loss have a dominant negative effect on connexin 30. Hum Mol Genet 2003, 12:805-812[Abstract/Free Full Text]
- Laird DW: Life cycle of connexins in health and disease. Biochem J 2006, 394:527-543[CrossRef][Medline]
- Evans WH, De Vuyst E, Leybaert L: The gap junction cellular internet: connexin hemichannels enter the signalling limelight. Biochem J 2006, 397:1-14[CrossRef][Medline]
- Welch KO, Marin RS, Pandya A, Arnos KS: Compound heterozygosity for dominant and recessive GJB2 mutations: effect on phenotype and review of the literature. Am J Med Genet A 2007, 143:1567-1573
- Melchionda S, Bicego M, Marciano E, Franze A, Morgutti M, Bortone G, Zelante L, Carella M, D'Andrea P: Functional characterization of a novel Cx26 (T55N) mutation associated to non-syndromic hearing loss. Biochem Biophys Res Commun 2005, 337:799-805[CrossRef][Medline]
- Oshima A, Doi T, Mitsuoka K, Maeda S, Fujiyoshi Y: Roles of Met-34, Cys-64, and Arg-75 in the assembly of human connexin 26. Implication for key amino acid residues for channel formation and function. J Biol Chem 2003, 278:1807-1816[Abstract/Free Full Text]
- Primignani P, Castorina P, Sironi F, Curcio C, Ambrosetti U, Coviello DA: A novel dominant missense mutation—D179N—in the GJB2 gene (connexin 26) associated with non-syndromic hearing loss. Clin Genet 2003, 63:516-521[CrossRef][Medline]
- Hamelmann C, Amedofu GK, Albrecht K, Muntau B, Gelhaus A, Brobby GW, Horstmann RD: Pattern of connexin 26 (GJB2) mutations causing sensorineural hearing impairment in Ghana. Hum Mutat 2001, 18:84-85[Medline]
- Matos TD, Caria H, Simoes-Teixeira H, Aasen T, Dias O, Andrea M, Kelsell DP, Fialho G: A novel M163L mutation in connexin 26 causing cell death and associated with autosomal dominant hearing loss Hear Res 2008, 240:87-92[CrossRef][Medline]
Top
Abstract
Case Reports
