| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
From the Department of Dermatology,* Hokkaido University Graduate School of Medicine, Sapporo, Japan; Laboratory of Structural Bio-Macromolecular Science III, Division of Biological Science, Hokkaido University, Sapporo, Japan; and the Department of Dermatology, Iwate Prefectural Central Hospital, Iwate, Japan
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
The importance of gap junctional intercellular communication in the function of several tissues or organs is demonstrated by the presence of Cx gene mutations in several congenital disorders.1,2 For example, Cx26 mutations are a major cause of nonsyndromic congenital sensorineural deafness (DFNB1: MIM no. 220290). The Cx-related deafness is sometimes associated with congenital skin disorders, such as Vohwinkel’s syndrome (MIM no. 124500)3 and keratitis-ichthyosis-deafness (KID) syndrome (MIM no. 148210).4 These syndromic deafness syndromes are autosomal dominant diseases in which it is assumed that the mutated Cx26 protein inhibits normal gap junction function by a dominant-negative effect.5
Here, we report the case of a Japanese girl with KID syndrome. The mutation analysis of GJB2 (the coding region of Cx26 gene) revealed a novel missense mutation, N14Y. This mutation is in the N-terminal domain of Cx26 where other mutations in KID syndrome have previously been reported; therefore, it is assumed that the N-terminal domain of Cx26 should be necessary for the proper function of the protein. To understand the function of this domain, it was important to clarify the relation between the N14Y mutation and the altered channel function of the gap junction. For this, we performed the following experiments: 1) ultrastructural examination of gap junctions and immunohistological study for Cx26 expression in the patient’s skin was performed; 2) we investigated the effect of N14Y mutation on gap junctional intercellular communication by a dye transfer assay; and 3) we studied the structural changes in the N-terminal domain of Cx26 by molecular structural analysis using nuclear magnetic resonance (NMR).
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Skin biopsies were taken from the skin lesion on the left foot of the patient after informed consent. Genomic DNA samples from peripheral blood were obtained from the family members including the patient and her parents after informed consent.
Mutation Analysis
Genomic DNA was extracted from peripheral blood and used as a template of gene amplification. The coding region of GJB2 (GenBank accession no. NM 004004) was amplified by polymerase chain reaction (PCR), as previously described.5 DNA sequencing of the PCR product was performed with an ABI Prism 3100-Avant genetic analyzer (Perkin Elmer-ABI, Foster City, CA).
Electron Microscopy
The skin sample was fixed in one-half strength Karnovsky’s fixative or 2% glutaraldehyde solution, postfixed in 1% OsO4, dehydrated, and embedded in Epon 812. The sample was ultrathin-sectioned at a thickness of 70 nm and stained with uranyl acetate and lead citrate. Photographs were taken using a Hitachi H-7100 transmission electron microscope (Hitachi High-Technologies Corporation, Tokyo, Japan).
Immunofluorescence Labeling
The patient’s skin sample was snap-frozen in isopentane, and 6-µm-thick sections were cut using a cryostat. The sections were washed with 0.01 mol/L phosphate-buffered saline (PBS) for 10 minutes and then incubated in rabbit polyclonal anti-Cx26 antibody—the epitope is a portion of the cytoplasmic loop of Cx26—(Zymed Laboratories, San Francisco, CA) or mouse monoclonal anti-Cx43 antibody (clone 4E6.2; Chemicon International, Temecula, CA) solution for 1 hour at 37°C. Antibody dilutions were 1/10 for Cx26 antibody and 1/200 for Cx43 antibody. The sections were then incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit immunoglobulins for Cx26 and fluorescein isothiocyanate-conjugated goat anti-mouse immunoglobulins for Cx43 (Jackson Immunoresearch Laboratories, West Grove, PA) solution for 30 minutes at room temperature, followed by 10 µg/ml propidium iodide solution as a nuclear counterstain (Sigma Chemical Co., St. Louis, MO) for 10 minutes. The sections were extensively washed with 0.01 mol/L PBS between incubations. The stained sections were mounted using a glycerol-based mounting medium (Permafluor, Shandon, PA) and stored in the refrigerator in the dark. Immunostaining was detected as green (fluorescein isothiocyanate), and nuclear staining was observed as red (propidium iodide). Overlap of both fluorescein isothiocyanate and propidium iodide was demonstrated as yellowish color. Fluorescence images were observed using an Olympus IX70 confocal laser-scanning microscope. Image collection was performed by software Fluoview version 2.0 (Olympus America Inc., Melville, NY).
Cell Culture
Cell culture was performed with slight modifications to the methods previously described.6 A biopsy was taken from a hyperkeratotic plaque on the dorsum of the patient’s left foot. Biopsy samples were kept in ice-cold PBS containing antibiotics (100 U/ml penicillin, 100 µg/ml streptomycin, 50 µg/ml gentamicin, 2.5 µg/ml amphotericin B, and 0.4% neomycin). After trimming subepidermal connective tissue, the samples were placed (overnight, 4°C) in dispase in PBS. The epidermis was peeled off from the tissue and placed in trypsin-ethylenediaminetetraacetic acid solution (0.25% trypsin in 0.05% ethylenediaminetetraacetic acid). Single cells were suspended in keratinocyte seeding medium.7 Cells were seeded on mitomycin C-treated feeder layer of 3T3 cells in 35-mm-diameter tissue culture plates. Cultures were fed with Dulbecco’s modified Eagle’s medium, incubated at 37°C, 5% CO2.
Dye Transfer Assay
The dye transfer assay was performed by scrape-loading methods introduced by el-Fouly and colleagues8 with slight modification. Briefly, epidermal keratinocytes from the patient, normal human epidermal keratinocyte (NHEK), and HaCaT cells (human keratinocyte cell line) were grown to confluency on 35-mm plastic plates in Dulbecco’s modified Eagle’s medium. After removal of medium, the cells were scraped using a plastic eraser, and 0.125% Lucifer yellow and 0.05% rhodamine dextran (Molecular Probes, Eugene, OR) dissolved in PBS were immediately added to the cells. Two minutes later, the dye solution was discarded, and the plates were rinsed with adequate amounts of PBS to remove detached cells and background fluorescence. Cells were examined under an Olympus IX70 confocal laser-scanning microscope (Olympus America Inc., Melville, NY) at the time of 20 minutes after scrape loading. The degree of dye transfer was quantified by counting the Lucifer yellow-positive cells per unit length (50 µm) of scratched plane. We counted only the Lucifer yellow-positive cells, and the cells that were both Lucifer yellow- and rhodamine-positive were excluded. The gap junction-mediated dye transfer was confirmed by the presence of red staining of rhodamine dextran at the edge of the scratched plane, because rhodamine dextran is 10 kd and cannot pass through the gap junction. Lucifer yellow, on the other hand, is 0.5 kd and can be transferred to other cells through gap junctions; therefore, it diffuses from the edge of the scratched plane to inner intact cells. Statistical analysis was performed using Student’s t-test.
Peptide Synthesis
Wild-type and mutant peptides containing the 20 N-terminal amino acids of Cx26 (MDWGTLQTILGGVNKHSTSI and MDWGTLQTILGGVYKHSTSI, respectively) were synthesized by Sigma Genosys (Japan).
NMR Spectroscopy
The peptides were dissolved to a final concentration of 2 mmol/L in 350 µl of 90% H2O/10% D2O or 99.9% D2O at pH 4.0. The pH was adjusted by adding µl increments of HCl and NaOH. The NMR experiments were performed on JEOL ECA 600 or Bruker DMX 500 spectrometers. The NMR spectra, DQF-COSY,9 TOCSY,10 and NOESY11 were recorded at 10°C, and some experiments were also recorded at 20 and 30°C to resolve ambiguities. TOCSY spectra with a MLEV-17 sequence were collected with spin-lock times of 90 ms, and NOESY spectra were obtained with mixing times of 100 to 300 ms. The chemical shifts were measured from the internal standard of sodium 2,2-dimethyl-2-silapentane-5-sulfate. All two-dimensional spectra were processed using NMRPipe software.12
Circular Dichroism (CD) Measurement
All measurements were performed on a Jasco J-725 spectropolarimeter (Tokyo, Jasco, Japan). Sample solutions were buffered with 10 mmol/L potassium phosphate buffer (pH 7.0) in various 2,2,2-trifluoroethanol (TFE) concentrations. Spectra were recorded at 25°C and at peptide concentrations of 0.2 mmol/L using a quartz cell with a path length of 1 mm. All of the spectra were baseline-corrected by subtracting buffer spectra.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The patient was a 4-year-old Japanese girl with no family history of skin disorders or auditory dysfunction. From birth, impetiginous erythema had been observed on her neck, axilla, and perianal areas. At the age of 4 months, spiky white keratotic papules appeared on her palms and soles, and impetiginous plaques were also noted on her occipital area. At the age of 2 years, she was referred to the otolaryngological clinic, and profound sensorineural deafness was noted. At the age of 4 years, keratitis was found on both eyes. She was diagnosed with KID syndrome at this time. By the age of 4 years, hyperkeratotic plaques were scattered on her scalp and extremities (Figure 1A)
. Palms and soles were severely hyperkeratotic (Figure 1, B and C)
.
|
Direct sequencing of the patient’s genomic DNA, amplified by PCR, revealed that the patient was heterozygous for a novel missense mutation NM_004004:c.40A>C resulting in the amino acid substitution asparagine to tyrosine (N14Y). This mutation results in the gain of a recognition site for the restriction enzyme Bsp1470I (Figure 2)
. N14Y was not found in her parents (Figure 2)
and was thought to be a de novo mutation. In addition, this mutation was not found in 50 normal unrelated Japanese alleles (25 normal unrelated Japanese individuals) and was unlikely to be a polymorphism. Direct sequencing of the entire coding region and borders of GJB2 failed to detect any other pathogenic mutation in the patient’s DNA.
|
Skin biopsy obtained from the lesional skin revealed hyperkeratosis with focal parakeratosis and regular acanthosis with broad rete ridges (Figure 3A)
. Granular layer was absent, and vacuolar change of cytoplasm was observed in the upper spinous layer. Electron microscopy demonstrated gap junctions in all epidermal layers, with normal morphology showing typical pentalaminar structures 20 nm in width (Figure 3B)
. Immunofluorescence showed Cx26 expression in the upper layer of wide rete ridge in the patient’s epidermis (Figure 3C)
, compared with the normal epidermis that does not express Cx26 (Figure 3G)
. The staining of Cx26 was more cytoplasmic than membranous in the keratinocytes of the upper layer, although punctate staining on cellular interface was also observed in the acrosyringium cells (Figure 3D)
. We also examined Cx43 expression in the patient’s epidermis because Cx43 is the major Cx expressed in the epidermis. Cx43 staining was seen in the middle and upper epidermal layer, similar to the expression of Cx26 in the patient’s skin (Figure 3E)
. The expression of Cx43 was almost completely membranous (Figure 3F)
and seemed to be the same as in normal control skin (Figure 3H)
. The antibody used in the present study binds to both normal and mutant Cx26 peptides; thus, it was not clear whether the overexpressed Cx26 in the patient’s epidermis was normal and/or mutated.
|
The transfer of dye between cells was observed in all cell types, although the degree of dye transfer of cultured patient’s keratinocytes was less than in NHEK and HaCaT cells, indicating abnormality of gap junctional intercellular communication in the patient (Figure 4, A–C)
. The number of Lucifer yellow-positive keratinocytes per unit length of scratched plane was counted. The number of positive cells in cultured patient’s keratinocytes was significantly smaller than that in NHEK and HaCaT cells (P < 0.01) (Figure 4G)
.
|
The chemical shift assignments of wild-type and mutant peptides were performed according to standard procedure of sequential assignment.13
An almost complete assignment of the proton NMR signals of wild-type and mutant peptides was obtained. A comparison between the DQF-COSY spectra of the fingerprint regions of wild type and mutant, with the results of assignment, is shown in Figure 5
.
|
-protons and amide protons are plotted in Figure 6
. Figure 6A
shows that the patterns of chemical shift deviation of the mutant peptide from random coil structure resemble that of wild-type peptide in both CH and NH. In addition, continuous negative values of chemical shift deviations of CH in the N-terminal region (Trp3-Leu10) suggest formation of helical conformation in both wild type and mutant. Thus, the three-dimensional structural properties of these two peptides are thought to be basically the same. This result is further confirmed by the data that large changes in shift between wild type and mutant are only observed for resonances around the mutated residue N14Y (Figure 6B)
. It is reasonable to think that the conformation of mutant peptide maintains its wild-type form because the chemical shifts of -protons and amide protons are very sensitive to backbone conformation and secondary structure.
|
-helical conformation, especially in the N-terminal region. Figure 7
shows the NH-NH region from NOESY spectra of wild-type and mutant peptide. In both spectra, a number of sequential NH-NH NOEs were observed and these results indicated that these peptides have propensity to adopt helical structure. Although the basic NOE pattern of the mutant peptide agreed with that of wild-type one, some novel NOEs were observed from Tyr14 (Figure 8)
. In particular, a number of medium-range NOEs from the side chain of Tyr14, such as between 
H of Tyr14 and H of Gly12, 
H of Tyr14. and H of Gly 12, H of Tyr14 and H of His16 were observed. In contrast, no medium-range NOEs were observed in this C-terminal region in the case of wild-type peptide.
|
|
The far-UV CD spectra were recorded for wild-type and mutant peptides derived from the N-terminal 20 amino acids of Cx26 (Figure 9)
. In water, the CD spectra of both wild-type and mutant peptides showed random-like conformational features (Figure 9A)
. The -helix contents of these peptides are estimated to be considerably low judged from the value at 222 nm. On the other hand, in the case of CD spectra in TFE solution, the -helical contents increased with increasing TFE concentration (Figure 9B)
. Although the conformational changes were induced by TFE in both wild-type and mutant peptides, -helical structure was induced more easily and strongly in the case of mutant one.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
.4,14-23
To date, five Cx26 mutations resulting in KID syndrome have been reported in the literature.4,15,16
Among them, two mutations, G12R and S17F, exist in the N-terminal domain of Cx26, similar to our case (N14Y).
|
Despite the mutation of Cx26 in our patient, the morphology of gap junctions observed by electron microscopy seemed to be normal, and disturbed adhesional structures were not observed in the specimens. In a report of erythrokeratoderma variabilis with Cx31 mutation, morphological changes of gap junctions were also not observed.25 In our case, the mutation was in the cytoplasmic side; therefore, the coupling of connexons with those of opposite cells at the extracellular site might not be affected, resulting in normal morphology of gap junctions. Otherwise, the normal gap junction structures seen in the patient’s skin sample may consist of Cx43, which was expressed normally in the patient’s skin.
By immunofluorescence study, the expression of Cx26 was observed in the upper layer of the epidermis in our patient. In normal skin, Cx26 expression is not seen in the epidermis, although strong expression is observed in the hair follicle and the sweat duct and gland.26 However, in some of the hyperkeratotic skin disorders such as psoriasis, Cx26 expression is seen in the upper layer of the epidermis.27 Cx26 expression in the epidermis of patients with KID syndrome has also been previously reported.4 The Cx26 staining in the epidermis of our patient confirmed in vivo expression of Cx26 in the patient’s keratinocytes and supported the pathological significance of this protein for the hyperkeratotic changes of the epidermis.
The transfer of dye between cells was reduced in the patient’s keratinocytes compared to NHEK and HaCaT cells. This result directly proved that the channel function of gap junctions was affected by the Cx26 mutation, N14Y. The dye transfer was not completely impaired in the patient’s keratinocytes, probably because other Cx molecules, such as Cx43, can work partly independently of mutated Cx26. However, certain Cx26 mutations were reported to dominantly inhibit normal Cx43 function, demonstrated by coupling transfection of the mRNA of wild-type and mutant Cxs into Xenopus oocytes.28 Therefore, a part of impairment of gap junctional intercellular communication may be because of trans-dominant inhibition of mutated Cx26 to functions of other intact Cxs.
Our NMR data of N-terminal peptides of human Cx26 in water suggested that both wild-type and mutated peptides have basically the same conformational feature. The chemical shift differences of CH between the experimental shifts for the peptides and random coil shifts (Figure 6)
clearly suggested that these peptides have a tendency to form -helical structures, except for a few residues in the C termini. In addition, NOE data also showed formation of -helix-like conformation in some N-terminal residues. However, the number of NOEs observed in our studies was not sufficient to confirm that these peptides form rigid -helical conformation in water. In addition, CD data also suggested that helical contents of these peptides were low in water although -helical structures were induced by considerably low TFE concentrations. Thus, these peptides are likely to have relatively high flexibility in water even though rigid helical structures are easily induced by nonpolar environments, such as in TFE solvent. In water, there were no significant differences between CD spectra of these peptides. However, the secondary structure of mutant peptide is susceptible to induction of TFE. Thus, the mutation may change the conformational flexibility of the peptide and the helical propensity.
Recently, sequence-specific 1H NMR resonance assignment for the N-terminal 15-amino acid peptide derived from rat Cx26 have been reported.29 The primary structures of the N-terminal domain of Cx26 are highly conserved in mammals, and the difference between human and rat is only one residue (Thr8 in human to Ser8 in rat) in this N-terminal region. Our peptides synthesized from human Cx26 sequence are slightly longer than the peptide derived from rat (15 residues long). The results of NMR studies from rat Cx26 showed that the peptide has a more highly ordered structure than our human Cx26 peptide although the conformational features of rat Cx26 peptides basically agree with our human one. These peptides are highly homologous except in their length. Thus, the difference in flexibly may be attributable to the extension of C-terminal residues. Previous analysis of the rat Cx26 peptide suggested the importance of flexibility in the hinge region (Gly12 and Gly13) in the placement of the N-terminal residues within the channel pore. In the present study, some NOEs to these glycine residues, which were not observed in the wild-type peptide, appeared in the N14Y mutant. Thus, it is likely that the N14Y mutation induced a change in local flexibility and that the motion of the N-terminal residues, which are important for the channel function of the protein, was altered profoundly.
|
Footnotes |
|---|
Supported in part by the Ministry of Education, Science, Sports, and Culture of Japan (Kiban B grant-in-aid 16390312 to M.A.).
Accepted for publication April 25, 2006.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  E A de Zwart-Storm, H Hamm, J Stoevesandt, P M Steijlen, P E Martin, M van Geel, and M A M van Steensel A novel missense mutation in GJB2 disturbs gap junction protein transport and causes focal palmoplantar keratoderma with deafness J. Med. Genet., March 1, 2008; 45(3): 161 - 166. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
