Srinivas M et al. (2019), Connexin43 mutations linked to skin disease hav...

XB-ART-55617

Sci Rep 2019 Jan 10;91:19. doi: 10.1038/s41598-018-37221-2.

Show Gene links Show Anatomy links

Connexin43 mutations linked to skin disease have augmented hemichannel activity.

Srinivas M , Jannace TF , Cocozzelli AG , Li L , Slavi N , Sellitto C , White TW .

???displayArticle.abstract???
Mutations in the gene (GJA1) encoding connexin43 (Cx43) are responsible for several rare genetic disorders, including non-syndromic skin-limited diseases. Here we used two different functional expression systems to characterize three Cx43 mutations linked to palmoplantar keratoderma and congenital alopecia-1, erythrokeratodermia variabilis et progressiva, or inflammatory linear verrucous epidermal nevus. In HeLa cells and Xenopus oocytes, we show that Cx43-G8V, Cx43-A44V and Cx43-E227D all formed functional gap junction channels with the same efficiency as wild-type Cx43, with normal voltage gating and a unitary conductance of ~110 pS. In HeLa cells, all three mutations also localized to regions of cell-cell contact and displayed a punctate staining pattern. In addition, we show that Cx43-G8V, Cx43-A44V and Cx43-E227D significantly increase membrane current flow through formation of active hemichannels, a novel activity that was not displayed by wild-type Cx43. The increased membrane current was inhibited by either 2 mM calcium, or 5 µM gadolinium, mediated by hemichannels with a unitary conductance of ~250 pS, and was not due to elevated mutant protein expression. The three Cx43 mutations all showed the same gain of function activity, suggesting that augmented hemichannel activity could play a role in skin-limited diseases caused by human Cx43 mutations.

???displayArticle.pubmedLink??? 30631135
???displayArticle.pmcLink??? PMC6328547
???displayArticle.link??? Sci Rep
???displayArticle.grants??? [+]

Species referenced: Xenopus
Genes referenced: gja1

???attribute.lit??? ???displayArticles.show???

	Figure 1. Wild-type and mutant forms of Cx43 are equivalently expressed and form functional gap junctions in paired Xenopus oocytes. (a) Equal amounts of membrane extracts were frst probed with an antibody that recognized Cx43. H2O injected controls did not express Cx43 as expected. Wild-type Cx43, Cx43-G8V, Cx43- A44V, and Cx43-E227D were readily detected in lanes corresponding to each injection condition with similar band intensities. (b) To confrm equal sample loading, blots were stripped and reprobed with an antibody against β-tubulin, which was present at comparable levels in all lanes. (c) Gap junctional conductance levels for cell pairs expressing either wild-type, or mutant, forms of Cx43 were all signifcantly greater than the H2O injected negative controls (p<0.05, one way ANOVA), but not signifcantly diferent from each other. Values are the mean±SEM.
	Figure 2. Cx43 mutations do not alter gap junction voltage gating. Oocyte pairs were subjected to hyperpolarizing and depolarizing transjunctional potentials (Vj ) while recording junctional currents (Ij ). (a) Wild-type Cx43 gap junction channels had Ij s that decreased symmetrically at higher values of Vj . Ij s between cell pairs expressing Cx43-G8V (b), Cx43-A44V (c), or Cx43-E227D (d) behaved in a similar fashion. (e) Steady-state voltage gating of wild-type Cx43 (flled squares) showed an approximately symmetric decline in steady state conductance at increasing values of Vj . Data from cell pairs expressing Cx43-G8V (open squares), Cx43-A44V (open circles), or Cx43-E227D (open triangles) were similar to wild-type Cx43.
	Figure 3. Cx43 mutations induce large hemichannel currents in Xenopus oocytes. Single cells were clamped at a holding potential of −40 mV and subjected to voltage pulses ranging from −30 to +60mV in 10mV steps. H2O (a) and wild-type Cx43 (b) injected cells displayed negligible membrane currents. Cx43-G8V (c), Cx43- A44V (d), and Cx43-E227D (e) expressing oocytes displayed much larger hemichannel currents than wild-type Cx43. (f) Steady-state currents from each pulse were plotted as a function of membrane voltage. Steady state currents in wild-type Cx43 (flled squares) or H2O injected control cells (flled circles) were negligible at all tested membrane voltages. Cx43-G8V (open squares) expressing cells exhibited signifcantly increased steadystate currents at all voltages compared to either control or wild-type Cx43 oocytes. Cx43-A44V (open circles), or Cx43-E227D (open triangles) currents were similar to those observed in control cells at negative voltages, but increased at positive potentials. Data are the mean±SEM.
	Figure 4. Calcium and gadolinium ions block mutant Cx43 hemichannel activity. Oocytes expressing Cx43 mutations were sequentially stepped from −70 mV to +20mV (a) while being perfused with medium containing 2mM Ca2+, 0.2mM Ca2+, or 5 µM Gd3+ (b). In the example shown for a Cx43-E227D expressing cell (c), hemichannel currents were initially suppressed by 2mM Ca2+, and increased markedly when the extracellular calcium was reduced to 0.2mM. Hemichannel currents were also suppressed by perfusion with 5 µM Gd3+, and recovered rapidly when Gd3+ was washed out with 0.2mM Ca2+. Currents were suppressed again by increasing the extracellular calcium to 2.0mM.
	Figure 5. Representative examples of patch clamp recordings from cell-attached patches containing single Cx43-G8V (a), Cx43-A44V (b), and Cx43-E227D (c) hemichannels in symmetric 140mM KCl solutions. Single hemichannel currents were recorded in response to 8-s voltage ramps between −70 and +70mV. Current-voltage relations for Cx43-A44V and Cx43-E227D were linear, whereas those of Cx43-G8V showed slight inward rectifcation. All three mutations show closing transitions to subconductance states. Occasional transitions to the fully closed state are also seen (b). (d) Cell-attached patches from wild-type Cx43 injected oocytes failed to show single channel activity. (e) Mean slope conductances measured at Vm =0 are similar for the three mutant hemichannels. Values represent the mean±SEM from 5 to 7 recordings.
	Figure 6. Gating of single Cx43-E227D hemichannels at positive and negative voltages. Representative examples of recordings from a Cx43-E227D expressing oocyte in a cell-attached patch confguration containing one to two open channels at voltages ranging from −110 to +70mV. Cx43-E227D hemichannels were predominantly open at inside positive voltages. Occasional closures to subconductance states were observed at high positive voltages. In contrast, Cx43-E227D hemichannels showed voltage-dependent closures to negative voltages. While channels are predominantly open at −30 mV, dwell times in the open state decreased with hyperpolarization to −50, −70 and −90 mV. Residence in subconductance states or the fully closed state was favored at the more hyperpolarized voltages. Te zero current level, i.e. the fully closed state (C), the current levels of one (O1) or two fully open hemichannels (O2) and that of the subconductance state (S) are depicted by dotted lines. Te zero current level was determined as described in Methods.
	Figure 7. Expression of mutant Cx43 in transfected HeLa cells results in targeting to gap junction plaques and formation of functional intercellular channels. Cx43-G8V (a,b), Cx43-A44V (c,d), and Cx43-E227D (e,f) transfected cells (blue DAPI stain) displayed a strong Cx43 (red) labeling that concentrated at cell-tocell interfaces (white arrowheads) and correlated with GFP (green) fuorescence. (g) Measurement of gap junctional coupling in transfected cell pairs showed that all three Cx43 mutations induced similar high levels of conductance. (h) A single gap junction channel recording for Cx43-A44V shows transitions between the fully open state and closed state with a unitary conductance of 113 pS. Data are the mean±SEM.

References [+] :

Altman, Inflammatory linear verrucose epidermal nevus. 1971, Pubmed