Nature Genetics
20, 319 - 320 (1998)
doi:10.1038/3781
One connexin, two diseasesKaren P SteelMRC Institute of Hearing Research, University of Nottingham,
University Park, Nottingham NG7 2RD, UK.
karen@ihr.mrc.ac.uk The complexity of human genetic disease continues to confuse, and it sometimes
seems remarkable that so much progress has been made in identifying disease
genes when subsequent work shows that the story is much more involved than
was at first imagined. In this issue, one group reports on the genetic cause
of a skin disease1 and another, of hearing loss2.
Each disorder is attributed to mutations in the same gene, which poses the
question: how can mutations in one gene lead to such diverse disorders?
The gene in question is GJB3, which encodes the connexin 31 component
of gap junctions. Gap junctions connect adjacent cells, allowing small molecules
to pass from one cell to the next and are believed to play an important role
in intercellular communication (Fig. 1). Members of
the connexin family have highly conserved sequences and four transmembrane
domains separating two extracellular loops and one cytoplasmic loop, with
cytoplasmic carboxy- and amino-terminal ends (Fig. 2;
Refs 3,4). Six connexin
molecules assemble to form one connexon, which docks with its counterpart
in the neighbouring cell to form the gap junction channel. A connexon composed
of one type of connexin may dock with a connexon of another type to form a
heterotypic channel, but connexin-31 connexons are unusual in that they only
form functional channels when docked with an identical connexin 31 connexon.
 | | |
| | |
Connexin genes have long been aetiologic candidates for skin disorders,
because many are expressed in the skin and some are upregulated in damaged
or psoriatic skin. On page 366, Gabriela Richard
and colleagues provide the first description of mutations in a connexin gene
(GJB3) causing a skin disorder—in this case erythrokeratodermia
variabilis1. This dominant disease is characterized by variable
regions of hyperkeratosis and transient red patches. GJB3 was a good
candidate gene, as its chromosomal position colocalizes with the disease locus
and it is expressed in hair follicles and skin, with high levels of expression
in differentiating keratinocytes5,
6,
7. Three missense mutations
were detected; two affect the same amino acid residue at position 12 at the
N terminus, while the third results in a substitution in the second transmembrane
domain (Fig. 2).
GJB3 mutations are also described in progressive, dominantly inherited
hearing loss by Jia-hui Xia and colleagues (see page370;
2). This is the third connexin
associated with hearing impairment. X-linked Charcot-Marie-Tooth syndrome,
characterized by progressive hearing loss along with other symptoms, is caused
by mutations in GJB1 (encoding connexin 32; 8) and mutations in GJB2 (encoding connexin 26) are a frequent
cause of recessive non-syndromic hearing impairment9,
10, and
occasionally, of dominant, progressive hearing loss11. Xia and
colleagues report two mutations, both affecting the second extracellular loop
domain of connexin 31 (Fig. 2). One is a nonsense mutation
(R180X), leading to a predicted truncation of the fourth transmembrane and
C-terminal cytoplasmic domains, and the other is a missense mutation, changing
a conserved glutamine to lysine (E183K; Fig. 2). As
with those effecting the skin disorder, these mutations exert a variable penetrance,
with some carriers having either subclinical hearing impairment or normal
hearing. This variability is probably due to variations in compensating pathways
or interacting molecules, and is perhaps not surprising given that the auditory
system seems to manage well for the first two or three decades of life in
all carriers, until the onset of hearing loss. GJB3 is expressed in
the inner ear, as demonstrated by RT-PCR analysis carried out by Xia et
al., and Stefan Heller and colleagues have noted that a connexin of the
same molecular weight is also expressed in the inner ear of the chicken12. Gap junctions are widespread within the supporting cells of the
cochlear duct of the inner ear13, but the precise locale of
connexin 31 expression in the mammalian cochlea remains to be determined;
informed speculation regarding its role in cochlear function must therefore
wait. Autosomal dominant, progressive hearing loss (locus DFNA2) has been
localized to the same chromosomal region as GJB3 (14), suggesting that GJB3 is a good candidate for involvement
in DFNA2. As linkage to this region has been found in 5 of 21 families with
autosomal dominant hearing loss14, mutations in GJB3
may be a relatively common cause of dominant hearing loss.
Different functions for different domains
How can the same gene underlie two such different diseases? Relatively
little is known about the function of the different domains of connexin 31,
but the high degree of sequence conservation between the connexin genes allows
us to infer some properties from in vitro studies carried out on other
connexins using specific mutations and chimaeric molecules3,
4,
15,
16,
17,
18.
The N-terminal domain harbouring two of the missense mutations that cause
the skin disorder is thought to be involved in determining the polarity of
the voltage gating (that is, it determines whether the channel opens when
the cytoplasm is at negative or positive potential15). The third
'skin' mutation affects a residue that sits next to a conserved proline residue
in the second transmembrane domain which is critical for voltage gating activity16. In contrast, both of the mutations associated with hearing loss
affect residues in an extracellular loop that is thought to be involved in
regulating the specificity of connexon-connexon interactions. One of these,
the nonsense mutation, should truncate the protein, eliminating the fourth
transmembrane domain and the C-terminal region with its four potential sites
for phosphorylation5, which are believed to have a role in controlling
gating of the whole channel. Thus, the mutations in GJB3 may affect
different aspects of channel function, which might explain the different phenotypes.
Cx32 throws a spanner in the works
Challenging this hypothesis, however, is the fact that mutations in
GJB1 (encoding connexin 32) which result in X-linked Charcot-Marie-Tooth
disease affect both the glycine at position 12, whose GJB3 counterpart
is mutant in keratodermia, and the glutamine at the position equivalent to
183, which is mutant in hearing loss. The effect of these variations (G12S
and E186K, respectively) on trafficking connexin 32 in rat pheochromocytoma
cells was explored through transfection experiments17. Both
resulted in very low levels of expression compared with that obtained using
wild-type GJB1, and the mutant proteins were retained in the Golgi
apparatus rather than appearing at the cell surface. If the mutations in
GJB3 have an effect similar to that of their counterparts in GJB1,
defective protein trafficking may lead to a similar aberration in the localization
of mutant proteins, making the different phenotypes difficult to explain.
It should be noted, however, that the E186K variant seems be localized to
the cell membrane (but fails to produce a functional channel) in a Xenopus
oocyte expression system18, emphasizing the importance
of characterizing expression in mammalian cells which may more closely resemble
the in vivo circumstances. Transfection experiments should provide
clues to the phenotypic differences arising from different mutations in
GJB3; differential trafficking of mutant proteins by different tissues
is one possible explanation. Targeted mutagenesis of mouse Gjb3 is
another obvious experimental route.
Finally, it should be noted that a number of syndromes involve both hearing
impairment and skin diseases, and at least two involve erythrokeratodermia19; GJB3 therefore represents a good candidate regarding aetiologic
role. One gene underlying diverse disease phenotypes is not a new observation,
but it is useful to be continually reminded of the diverse nature of genetic
aetiology, as we prioritize our investigations of candidate genes in the course
of positional cloning. One wonders if GJB3 would have been considered
a candidate for one disease if it had already been demonstrated to cause the
other?
REFERENCES
- Richard, G. et al. Nature Genet. 20, 366−369 (1998). | Article | PubMed | ISI | ChemPort |
- Xia, J. et al. Nature Genet. 20, 370−373 (1998). | Article | PubMed | ISI | ChemPort |
- Bruzzone, R., White, T.W. & Paul, D.L. Eur. J. Biochem. 238, 1−27 (1996). | Article | PubMed | ISI | ChemPort |
- Goodenough, D.A., Goliger, J.A. & Paul, D.L. Annu. Rev. Biochem. 65, 475−502 (1996). | Article | PubMed | ISI | ChemPort |
- Hoh, J.H., John, S.A. & Revel, J.P. J. Biol. Chem. 266, 6524−6531 (1991). | PubMed | ISI | ChemPort |
- Butterweck, A., Elfgang, C., Willecke, K. & Traub, O. Eur. J. Cell Biol. 65, 152−163 (1994). | PubMed | ISI | ChemPort |
- Brissette, J.L., Kumar, N.M., Gilula, N.B., Hall, J.E. & Dotto, G.P. Proc. Natl Acad. Sci. USA 91, 6453−6457 (1994). | PubMed | ChemPort |
- Bergoffen, J. et al. Science 262, 2039−2042 (1993). | PubMed | ISI | ChemPort |
- Kelsell, D.P. et al. Nature 387, 80−83 (1997). | Article | PubMed | ISI | ChemPort |
- Estivill, X. et al. Lancet 351, 394−398 (1998). | Article | PubMed | ISI | ChemPort |
- Denoyelle, F. et al. Nature 393, 319−320 (1998). | Article | PubMed | ISI | ChemPort |
- Heller, S., Sheane, C.A., Javed, Z. & Hudspeth, J. Proc. Natl Acad. Sci. USA 95, 11400−11405 (1998). | Article | PubMed | ChemPort |
- Kikuchi, T., Kimura, R.S., Adams, J.C. & Paul, J.C. Anat. Embryol. 191, 101−118 (1995). | Article | PubMed | ISI | ChemPort |
- Van Camp, G. et al. Genomics 41, 70−74 (1997). | Article | PubMed | ChemPort |
- Verselis, V.K., Ginter, C.S. & Bargiello, T.A. Nature 368, 348−351 (1994). | Article | PubMed | ISI | ChemPort |
- Suchyna, T.M. et al. Nature 365, 847−849 (1993). | Article | PubMed | ISI | ChemPort |
- Deschênes, S.M. et al. J. Neurosci. 17, 9077−9084 (1997). | PubMed | ISI |
- Bruzzone, R. et al. Neuron 13, 1253−1260 (1994). | Article | PubMed | ISI | ChemPort |
- Gorlin, R.J., Toriello, H.V. & Cohen, M.M. Hereditary Hearing Loss and its Syndromes (Oxford University Press, Oxford, 1995).
|
