Closing the Gap on Autosomal Dominant Connexin-26 and Connexin-43 Mutants Linked to Human Disease

Cells within the vast majority of human tissues communicate directly through clustered arrays of intercellular channels called gap junctions. Gene ablation studies in mouse models have revealed that these intercellular channels are necessary for a variety of organ functions and that some of these genes are essential for survival. Molecular genetics has uncovered that germ line mutations in nearly half of the genes that encode the 21-member connexin family of gap junction proteins are linked to one or more human diseases. Frequently, these mutations are autosomal recessive, whereas in other cases, autosomal dominant mutations manifest as disease. Given the broad and overlapping distribution of connexins in a wide arrangement of tissues, it is hard to predict where connexin-linked diseases will clinically manifest. For instance, the most prevalent connexin in the human body is connexin-43 (Cx43), yet autosomal dominant mutations in the GJA1 gene, which encodes Cx43, exhibit modest developmental disorders resulting in a disease termed oculodentodigital dysplasia. Autosomal recessive mutations in the gene encoding Cx26 result in moderate to severe sensorineural hearing loss, whereas autosomal dominant mutations produce hearing loss and a wide range of skin diseases, including palmoplantar keratoderma. Here, we will focus on autosomal dominant mutations of the genes encoding Cx26 and Cx43 in relation to models that link genotypes to phenotypic outcomes with particular reference to how these approaches provide insight into human disease. Cells within the vast majority of human tissues communicate directly through clustered arrays of intercellular channels called gap junctions. Gene ablation studies in mouse models have revealed that these intercellular channels are necessary for a variety of organ functions and that some of these genes are essential for survival. Molecular genetics has uncovered that germ line mutations in nearly half of the genes that encode the 21-member connexin family of gap junction proteins are linked to one or more human diseases. Frequently, these mutations are autosomal recessive, whereas in other cases, autosomal dominant mutations manifest as disease. Given the broad and overlapping distribution of connexins in a wide arrangement of tissues, it is hard to predict where connexin-linked diseases will clinically manifest. For instance, the most prevalent connexin in the human body is connexin-43 (Cx43), yet autosomal dominant mutations in the GJA1 gene, which encodes Cx43, exhibit modest developmental disorders resulting in a disease termed oculodentodigital dysplasia. Autosomal recessive mutations in the gene encoding Cx26 result in moderate to severe sensorineural hearing loss, whereas autosomal dominant mutations produce hearing loss and a wide range of skin diseases, including palmoplantar keratoderma. Here, we will focus on autosomal dominant mutations of the genes encoding Cx26 and Cx43 in relation to models that link genotypes to phenotypic outcomes with particular reference to how these approaches provide insight into human disease. A little over a decade ago, the first mutations in the gene encoding Cx32 2The abbreviations used are: CxconnexinODDDoculodentodigital dysplasia. were found to be linked to the X-linked form of Charcot-Marie-Tooth disease (1Ressot C. Latour P. Blanquet-Grossard F. Sturtz F. Duthel S. Battin J. Corbillon E. Ollagnon E. Serville F. Vandenberghe A. Dautigny A. Pham-Dinh D. Hum. Genet. 1996; 98: 172-175Crossref PubMed Scopus (28) Google Scholar, 2Scherer S.S. Bone L.J. Deschenes S.M. Abel A. Balice-Gordon R.J. Fischbeck K.H. Novartis Found. Symp. 1999; 219 (Discussion 185–177): 175-185PubMed Google Scholar, 3Ressot C. Bruzzone R. Brain Res. Rev. 2000; 32: 192-202Crossref PubMed Scopus (67) Google Scholar). Now, numerous mutations in the gene encoding Cx32 have been linked to this neuropathy associated with the demyelination of peripheral axons (4Kleopa K.A. Scherer S.S. Neuromol. Med. 2006; 8: 107-122Crossref PubMed Scopus (105) Google Scholar). Once it was established that connexin gene mutations could lead to disease, this brought in a new era of connexin studies that included the structure and function analysis of disease-linked mutants. Mutations in the gene encoding Cx47 have been attributed to Pelizaeus-Merzbacher-like disease, which is another disease characterized by defective myelination (5Bugiani M. Al Shahwan S. Lamantea E. Bizzi A. Bakhsh E. Moroni I. Balestrini M.R. Uziel G. Zeviani M. Neurology. 2006; 67: 273-279Crossref PubMed Scopus (93) Google Scholar, 6Orthmann-Murphy J.L. Enriquez A.D. Abrams C.K. Scherer S.S. Mol. Cell. Neurosci. 2007; 34: 629-641Crossref PubMed Scopus (94) Google Scholar). Sensory organ diseases, including congenital cataracts, have been attributed to mutations in the genes encoding Cx46 and Cx50 (7Gerido D.A. White T.W. Biochim. Biophys. Acta. 2004; 1662: 159-170Crossref PubMed Scopus (138) Google Scholar), whereas moderate to severe sensorineural hearing loss has been assigned to mutations in upwards of five different connexin family members, including Cx26, Cx30, Cx30.3, Cx31, and Cx43 (8Petit C. Levilliers J. Hardelin J.P. Annu. Rev. Genet. 2001; 35: 589-646Crossref PubMed Scopus (260) Google Scholar, 9Yang J.J. Huang S.H. Chou K.H. Liao P.J. Su C.C. Li S.Y. Audiol. Neurotol. 2007; 12: 198-208Crossref PubMed Scopus (74) Google Scholar, 10Liu X.Z. Xia X.J. Adams J. Chen Z.Y. Welch K.O. Tekin M. Ouyang X.M. Kristiansen A. Pandya A. Balkany T. Arnos K.S. Nance W.E. Hum. Mol. Genet. 2001; 10: 2945-2951Crossref PubMed Scopus (120) Google Scholar). Interestingly, mutations in Cx26 alone are thought to account for 35–45% of all congenital sensorineural hearing loss in some populations (8Petit C. Levilliers J. Hardelin J.P. Annu. Rev. Genet. 2001; 35: 589-646Crossref PubMed Scopus (260) Google Scholar). The same subset of connexin family members linked to sensorineural hearing loss is, in some cases, responsible for a wide array of skin diseases (11Richard G. Clin. Exp. Dermatol. 2003; 28: 397-409Crossref PubMed Scopus (57) Google Scholar, 12Richard G. Clin. Dermatol. 2005; 23: 23-32Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar, 13van Steensel M.A. Am. J. Med. Genet. 2004; 131C: 12-19Crossref PubMed Scopus (48) Google Scholar). Until ODDD was linked to mutations in the GJA1 gene encoding Cx43 (14Paznekas W.A. Boyadjiev S.A. Shapiro R.E. Daniels O. Wollnik B. Keegan C.E. Innis J.W. Dinulos M.B. Christian C. Hannibal M.C. Jabs E.W. Am. J. Hum. Genet. 2003; 72: 408-418Abstract Full Text Full Text PDF PubMed Scopus (542) Google Scholar), it was expected that germ line mutations in this connexin would likely be lethal given the critical role Cx43 plays in cardiac function and its widespread expression in >35 distinct tissue environments (15Laird D.W. Biochim. Biophys. Acta. 2005; 1711: 172-182Crossref PubMed Scopus (245) Google Scholar, 16Laird D.W. Biochem. J. 2006; 394: 527-543Crossref PubMed Scopus (663) Google Scholar). In addition to reports linking connexin gene germ line mutations to disease, other studies have suggested that somatic mutations in the gene encoding Cx40 may be linked to atrial fibrillation (17Gollob M.H. Jones D.L. Krahn A.D. Danis L. Gong X.Q. Shao Q. Liu X. Veinot J.P. Tang A.S. Stewart A.F. Tesson F. Klein G.J. Yee R. Skanes A.C. Guiraudon G.M. Ebihara L. Bai D. N. Engl. J. Med. 2006; 354: 2677-2688Crossref PubMed Scopus (461) Google Scholar), and recently, polymorphisms in Cx50 have been proposed to be linked to schizophrenia (18Ni X. Valente J. Azevedo M.H. Pato M.T. Pato C.N. Kennedy J.L. J. Med. Genet. 2007; 44: 532-536Crossref PubMed Scopus (31) Google Scholar). These latter associations need to be interpreted cautiously and will likely require the study of large cohorts of patients before they can unequivocally be considered causal of human disease. connexin oculodentodigital dysplasia. Once molecular geneticists had identified mutations in the genes encoding connexins and linked these results to a cohort of patients, these findings inevitability led to the need for a rigorous set of studies to establish how these genotypic changes lead to disease outcomes. This process is particularly intriguing when considering autosomal dominant mutants, in which both wild-type and mutant proteins are predicted to be produced in equal quantities in all cells where the promoter is temporally and spatially activated. Notably, not only do these molecular studies provide reference points to understanding the manifestations of the disease but offer tremendous potential to assign functional importance to specific domains and motifs that house the mutation. In fact, the use of disease-causing Cx43 mutants provides an experimental setting that cannot be recapitulated by gene ablation or replacement approaches. Although deletion of a connexin gene (knock-out) or substitution of one connexin gene for another (knock-in) in the mouse is an excellent means to define the essential properties of a specific connexin, it does not provide a realistic representation of connexins in human health and disease, where patients rarely suffer from a condition in which a connexin is ablated. The closest comparison with connexin ablation is found in a few cases in which the connexin gene harbors a nucleotide duplication or deletion resulting in a severe truncation or deletion of the protein (19Ainsworth P.J. Bolton C.F. Murphy B.C. Stuart J.A. Hahn A.F. Hum. Genet. 1998; 103: 242-244Crossref PubMed Scopus (43) Google Scholar, 20Rabionet R. Lopez-Bigas N. Arbones M.L. Estivill X. Trends Mol. Med. 2002; 8: 205-212Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar). There are at least 14 autosomal dominant mutations resulting in 10 amino acid substitutions or deletion sites in the GJB2 gene encoding Cx26 linked to conditions of moderate to severe hearing loss and skin diseases that include Vohwinkel syndrome, keratitis-ichthyosis- deafness syndrome, hystrix-like ichthyosis-deafness syndrome, Bart-Pumphrey syndrome, and palmoplantar keratodermas (12Richard G. Clin. Dermatol. 2005; 23: 23-32Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). Only seven autosomal dominant mutations have been associated with hearing loss without the added skin disease burden. Thus, the autosomal dominant nature of these mutations tends to be syndromic, whereas >100 recessive mutations in the same gene typically manifest as sensorineural hearing loss only (12Richard G. Clin. Dermatol. 2005; 23: 23-32Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). Thus, one could argue that a mixed background of wild-type and mutant proteins increases the spectrum of disease load if not the severity of the disease. There are 39 autosomal dominant mutations in the GJA1 gene encoding Cx43 (Fig. 1A) (16Laird D.W. Biochem. J. 2006; 394: 527-543Crossref PubMed Scopus (663) Google Scholar) with the resulting disease outcome being classified as ODDD highlighted by developmental defects in the craniofacial bones around the eyes and nose, loss of enamel resulting in early destruction of the teeth, and lack of soft tissue separation of two or three digits (14Paznekas W.A. Boyadjiev S.A. Shapiro R.E. Daniels O. Wollnik B. Keegan C.E. Innis J.W. Dinulos M.B. Christian C. Hannibal M.C. Jabs E.W. Am. J. Hum. Genet. 2003; 72: 408-418Abstract Full Text Full Text PDF PubMed Scopus (542) Google Scholar). In addition to these commonly found developmental disorders, patients often exhibit an assortment of conditions that range broadly from neurological to cardiac disorders (14Paznekas W.A. Boyadjiev S.A. Shapiro R.E. Daniels O. Wollnik B. Keegan C.E. Innis J.W. Dinulos M.B. Christian C. Hannibal M.C. Jabs E.W. Am. J. Hum. Genet. 2003; 72: 408-418Abstract Full Text Full Text PDF PubMed Scopus (542) Google Scholar, 21Loddenkemper T. Grote K. Evers S. Oelerich M. Stogbauer F. J. Neurol. 2002; 249: 584-595Crossref PubMed Scopus (151) Google Scholar, 22Debeer P. Van Esch H. Huysmans C. Pijkels E. De Smet L. Van de Ven W. Devriendt K. Fryns J.P. Eur. J. Med. Genet. 2005; 48: 377-387Crossref PubMed Scopus (38) Google Scholar, 23Kjaer K.W. Hansen L. Eiberg H. Leicht P. Opitz J.M. Tommerup N. Am. J. Med. Genet. 2004; 127A: 152-157Crossref PubMed Scopus (83) Google Scholar, 24Vreeburg M. de Zwart-Storm E.A. Schouten M.I. Nellen R.G. Marcus-Soekarman D. Devies M. van Geel M. van Steensel M.A. Am. J. Med. Genet. 2007; 143A: 360-363Crossref Scopus (56) Google Scholar). It is possible that epigenetic effects may also contribute to these disease states. Interestingly, in cases in which the patient harbors a frameshift mutation (fs230 (24Vreeburg M. de Zwart-Storm E.A. Schouten M.I. Nellen R.G. Marcus-Soekarman D. Devies M. van Geel M. van Steensel M.A. Am. J. Med. Genet. 2007; 143A: 360-363Crossref Scopus (56) Google Scholar) or fs260 (25van Steensel M.A. Spruijt L. van der Burgt I. Bladergroen R.S. Vermeer M. Steijlen P.M. van Geel M. Am. J. Med. Genet. 2005; 132A: 171-174Crossref PubMed Scopus (74) Google Scholar)) resulting in a gross deletion of the C terminus of Cx43, these individuals have an increased disease load of palmoplantar keratodermas or palmar hyperkeratosis not unlike what is found in some patients harboring dominant GJB2 mutations. The fact that one point mutation in Cx43 (L11P) (26Kelly S.C. Ratajczak P. Keller M. Purcell S.M. Griffin T. Richard G. Eur. J. Dermatol. 2006; 16: 241-245PubMed Google Scholar) has been associated with hyperkeratosis suggests that the added disease load in the skin is not restricted solely to the events linked to the C terminus of the molecule. A common approach to assess any new family or class of disease-linked mutants is to express cDNA constructs that encode the mutant protein in reference cell models that lack endogenous connexins and compare and functional properties with wild-type half of the autosomal dominant Cx26 mutants G. S. T. S. S. L. M. D. M. 2003; 10: PubMed Scopus Google Scholar, F. White T.W. N. T.W. D.L. J. M.B. Richard G. J. 2001; PubMed Google Scholar, T. D. D.W. J. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar) and Cx43 mutants X.Q. Shao Q. Bai D. D.W. J. 2006; Full Text Full Text PDF PubMed Scopus Google Scholar, W. Gong X.Q. E. Bai D. E. Shao Q. G.M. D.W. J. 2005; Full Text Full Text PDF PubMed Scopus Google Scholar, J. W. A. S. Jabs E.W. M. H. Res. 2005; PubMed Google Scholar, A. W.A. Jabs E.W. A.C. J. 2006; PubMed Scopus Google Scholar) have been this approach and found to exhibit two distinct the vast majority of mutants the cell and into gap as by on of and these same mutants are of functional gap junction channels in gap intercellular reference cells or form channels that function in comparison with wild-type The of these of mutants is that to to the cell is but they suffer from or resulting in the or of the gap junction Although is the most common of Cx26 and Cx43 mutants, a class of mutants G. S. T. S. S. L. M. D. M. 2003; 10: PubMed Scopus Google Scholar) and (25van Steensel M.A. Spruijt L. van der Burgt I. Bladergroen R.S. Vermeer M. Steijlen P.M. van Geel M. Am. J. Med. Genet. 2005; 132A: 171-174Crossref PubMed Scopus (74) Google Scholar)) exhibit protein In the of the it may the to the cell but to and to the where it is most prevalent T. D. D.W. J. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). the other the fs260 mutant to have in associated with the (Fig. and X.Q. Shao Q. Bai D. D.W. J. 2006; Full Text Full Text PDF PubMed Scopus Google Scholar). are also a few mutants that exhibit properties of being to the cell and into gap with J. W. A. S. Jabs E.W. M. H. Res. 2005; PubMed Google Scholar). both mutant and wild-type connexin are within the same the is these will if does the mutant the function of the wild-type connexin or does the wild-type connexin the function of the for mutant and wild-type Cx43 in some of was by in which wild-type and mutant and Cx43 proteins were found to X.Q. Shao Q. S. Bai D. D.W. J. 2007; Full Text Full Text PDF PubMed Scopus Google Scholar). Although this is likely to if not most of the autosomal dominant Cx43 and Cx26 mutants, this has yet to be studies in which wild-type Cx26 or Cx43 was with mutant have that the mutants typically the function of wild-type F. White T.W. N. T.W. D.L. J. M.B. Richard G. J. 2001; PubMed Google Scholar, T. D. D.W. J. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). this to be as mutant connexins are often in to wild-type connexin to this have equal quantities of encoding wild-type and mutant and the of and proteins found at gap junction X.Q. Shao Q. S. Bai D. D.W. J. 2007; Full Text Full Text PDF PubMed Scopus Google Scholar). this more both the and mutants were found to be to Cx43 when predicted to be at a In these two mutants different that not all mutants are likely to be in the function of wild-type Cx43 X.Q. Shao Q. S. Bai D. D.W. J. 2007; Full Text Full Text PDF PubMed Scopus Google Scholar). Given that autosomal dominant Cx26 mutants and a of Cx43 mutants are syndromic, it was that one or more of these mutants may have effects on other members of the connexin The for this is on the fact that connexins can with other members of the connexin family in to form a variety of For it is to predict that Cx26 mutants directly with Cx32 both wild-type Cx32 and Cx26 form K.A. J. Full Text Full Text PDF PubMed Scopus Google Scholar). analysis in which Cx26 mutants were with Cx32 revealed that the mutants were dominant to this connexin F. White T.W. N. T.W. D.L. J. M.B. Richard G. J. 2001; PubMed Google Scholar, T. D. D.W. J. 2004; Full Text Full Text PDF PubMed Scopus Google Scholar). The of this but may express upwards of different connexin family members, may result in a of gap intercellular skin Although reference can provide into how mutants are and within a they do not the of and that in or autosomal dominant Cx26 and Cx43 mutants clinically skin these mutants in and the and of the skin offer a potential that both Cx26 and Cx43 mutants have been in and into an has been for defects in and cell as as for the to express molecular of S. A.C. J.L. Shao Q. G.M. D.W. J. 2007; Full Text Full Text PDF PubMed Scopus Google Scholar, A.C. T. J.L. G. Shao Q. D.W. 2005; 12: PubMed Scopus Google Scholar, T. Shao Q. D.W. J. 2007; PubMed Scopus Google Scholar). studies have revealed that Cx26 and Cx43 mutants that are to hyperkeratosis and other skin defects to any changes in the or of (Fig. and It is that these mutants typically manifest effects in the skin of the and of patients, the that an will not in to In it is possible that to human and that more approaches human are Thus, the of mouse models of human diseases may to be necessary to Cx26 and Cx43 mutant effects on organ and to disease. In and G. R. T. L. K. S. S. L. M. D. M. Hum. Mol. Genet. 2003; 12: PubMed Scopus Google Scholar) on the first mouse of a human connexin disease. In this the a mouse that the mutant by the promoter and that the to moderate sensorineural hearing loss and Vohwinkel G. R. T. L. K. S. S. L. M. D. M. Hum. Mol. Genet. 2003; 12: PubMed Scopus Google Scholar). This mouse was the of patients that this mutation and suffer from Vohwinkel a mutant mouse of by a mutant was and characterized M. P. R. C. M. G. E. K. Hum. Mol. Genet. 2007; 16: PubMed Scopus Google Scholar). Interestingly, this mutant mouse recapitulated the of this autosomal dominant the of of the found in was not in these These findings that the in human connexin-linked diseases may not be in mouse distinct and For Cx43 and Cx26 exhibit distinct expression within the mouse in comparison with the found in human In another a mouse was by that a mutation in the gene encoding Cx43 that in being with as the N. N. C. Gong X.Q. C. L. K. D. W. Shao Q. D. I. I. Bai D. M. C. D.W. G.M. J. 2005; PubMed Scopus Google Scholar). Although this mutation has not been identified within the human these were characterized to exhibit in the and (Fig. craniofacial and loss of to the most developmental found in ODDD In this had a mouse of the molecular these Cx43 and gap junction in tissues, including the and skin (Fig. with a for the loss of the most of Cx43, which are associated with Cx43 into gap junction S. A.C. J.L. Shao Q. G.M. D.W. J. 2007; Full Text Full Text PDF PubMed Scopus Google Scholar, N. N. C. Gong X.Q. C. L. K. D. W. Shao Q. D. I. I. Bai D. M. C. D.W. G.M. J. 2005; PubMed Scopus Google Scholar). when the functional of Cx43 was in these it was found to be at the fact that the mutant protein was dominant to the wild-type connexin both are predicted to be at or N. N. C. Gong X.Q. C. L. K. D. W. Shao Q. D. I. I. Bai D. M. C. D.W. G.M. J. 2005; PubMed Scopus Google Scholar). In a more and N. J. N. L. J. S. A. 2007; PubMed Scopus Google Scholar) a mutant mouse to express a human this mutant mouse as for the as as molecular that included Cx43 and to intercellular N. J. N. L. J. S. A. 2007; PubMed Scopus Google Scholar). both mouse models to the human ODDD disease at least with to the mutant defects of the these a and N. J. N. L. J. S. A. 2007; PubMed Scopus Google Scholar), whereas of and defects were in the N. N. C. Gong X.Q. C. L. K. D. W. Shao Q. D. I. I. Bai D. M. C. D.W. G.M. J. 2005; PubMed Scopus Google Scholar). Interestingly, these latter defects are rarely in the cohort of human ODDD patients, some that these conditions may be more it will be to these mouse models can be used to provide as to what disease may be in the human ODDD patient that typically but may manifest or disease burden. For instance, the with the as to ODDD patients in the molecular the as to the in Cx43, in both reference mouse would also in ODDD In it that functional Cx43 can be to a of its in a wide array of tissues before for Cx43 being produced at essential for most organ given the critical role of Cx43 in cardiac it to be ODDD patients are more at for as in some mutant This will some to in large to the fact that the of ODDD in the human to are in the It is possible that as a cohort of patients is these of studies will more The of autosomal dominant connexin gene mutations linked to human diseases has the of connexins in health and disease. The fact that amino acid substitutions are of Cx26 and Cx43 function suggests that the majority of the motifs of these connexins are essential and to Although it is intriguing to that mouse models in which specific mutations to of the mutant and wild-type connexins are into the mouse the reference to the and organ of the disease, this may be to connexin and that and that it will to be necessary and to the and functional of connexin mutants in and and It would also be to establish cell and from ODDD patients in which the and functional properties of the connexin mutants can be in yet another reference that to the human In the into how connexin mutants developmental defects or organ will likely to require the of results from reference in and in

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.