A novel N14Y mutation in Connexin26 in keratitis-ichthyosis-deafness syndrome: analyses of altered gap junctional communication and molecular structure of N terminus of mutated Connexin26

Abstract

Connexins (Cxs) are transmembranous proteins that connect adjacent cells via channels known as gap junctions. The N-terminal 21 amino acids of Cx26 are located at the cytoplasmic side of the channel pore and are thought to be essential for the regulation of channel selectivity. We have found a novel mutation, N14Y, in the N-terminal domain of Cx26 in a case of keratitis-ichthyosis-deafness syndrome. Reduced gap junctional intercellular communication was observed in the patient's keratinocytes by the dye transfer assay using scrape-loading methods. The effect of this mutation on molecular structure was investigated using synthetic N-terminal peptides from both wild-type and mutated Cx26. Two-dimensional (1)H nuclear magnetic resonance and circular dichroism measurements demonstrated that the secondary structures of these two model peptides are similar to each other. However, several novel nuclear Overhauser effect signals appeared in the N14Y mutant, and the secondary structure of the mutant peptide was more susceptible to induction of 2,2,2-trifluoroethanol than wild type. Thus, it is likely that the N14Y mutation induces a change in local structural flexibility of the N-terminal domain, which is important for exerting the activity of the channel function, resulting in impaired gap junctional intercellular communication.

Figure 1-6927

Clinical figures of the patient at the age of 4. A: Hyperkeratotic plaques on her scalp. She wears a hearing aid on the left side because of sensorineural deafness. B and C: Hyperkeratosis of her palms and soles.

Figure 2-6927

DNA sequences of GJB2. A heterozygous A to T transition at codon 40 is detected in the patient’s gene (NM_004004:c.40A>T). This mutation leads to the amino acid substitution N14Y. This missense mutation is de novo because it is not found in the parent’s DNA. N14Y results in the gain of a Bsp1407I restriction site. PCR products from the patient (Pt) are digested into fragments of 733 and 943 bp. In contrast, the digested PCR products from the father (Fa) show only the 943-bp band.

Figure 3-6927

Skin morphology and immunofluorescence study. A: H&E staining. Hyperkeratosis with focal parakeratosis and regular acanthosis with broad rete ridges are observed. Granular layer was lost and vacuolar change of cytoplasm was seen in the upper spinous layer. B: Electron microscopy in the patient’s skin (granular layer of the epidermis). The gap junctions have a typical pentalaminar structure, 20 nm in width. Abnormal junctional structures were not found. D: Desmosome, G: gap junction. C–F: Cx26 and Cx43 expression of the patient’s epidermis. Green (fluorescein isothiocyanate) indicates Cx26 expression. Red (propidium iodide) is nuclear staining. C: Cx26 was expressed in the keratinocytes in the widened rete ridge. D: The staining of the Cx26 in the keratinocytes was more cytoplasmic than membranous, although punctate membranous staining was also seen in the acrosyringium cells. Cx43 expression was observed in the keratinocytes in the middle and upper epidermal layers (E) and was mostly membranous (F). G: Cx26 staining in normal skin. There was no expression of Cx26 in the normal epidermis. The staining at the corneum seemed to be nonspecific. H: Cx43 staining in normal skin. The staining was observed at the membrane of epidermal keratinocytes. Scale bar = 100 nm.

Figure 4-6927

Dye transfer assay of cultured keratinocyte. A: NHEK. B: HaCaT. C: Keratinocytes of KID syndrome. The diffusion of Lucifer yellow of patient’s keratinocytes was less than in NHEK and HaCaT cells. Rhodamine dextran (red) is observed at the edge of the scratched plane because the rhodamine dextran molecule is too bulky to pass through gap junctions (D–F: NHEK, HaCaT, and patient’s keratinocyte, respectively). G: The number of Lucifer yellow-positive cells per unit length of scratched plane. The number of positive cells of the patient’s keratinocytes was significantly smaller than that of NHEK and HaCaT. NHEK: 7.17 ± 1.14 (average ± SD); HaCaT: 7.43 ± 1.07; patient’s cell: 4.34 ± 0.75. *P < 0.01 by Student’s t-test.

Figure 5-6927

Comparison of the fingerprint region from DQF-COSY spectra of wild type (A) and N14Y mutant (B). The mutation does not cause significant change in chemical shifts when compared to wild type.

Figure 6-6927

Chemical shift differences of amino acid residues in wild type and N14Y mutant. A: The differences in chemical shift defined as δ (wild-type or mutant peptide)-δ (random coil) are plotted. B: The differences in chemical shift defined as wild-type peptide-mutant are plotted.

Figure 7-6927

NH-NH regions from NOESY spectra of wild-type (A) and N14Y mutant (B) peptides. Sequential NH-NH cross-peaks are annotated.

Figure 8-6927

Novel NOE signals in N14Y mutant.

Figure 9-6927

Far-UV CD spectra of wild-type and mutant peptides. Wild-type (blue) and mutant peptide (red) were measured in water (A) and in various TFE concentrations (B, C). The numbers indicate the percentage of TFE.

Figure 10-6927

The Cx26 mutations of syndromic sensorineural deafness reported in the literature. The mutations of KID syndrome are aggregated to the N-terminal side of Cx26, especially in the cytoplasmic portion.