Differentially altered Ca2+ regulation and Ca2+ permeability in Cx26 hemichannels formed by the A40V and G45E mutations that cause keratitis ichthyosis deafness syndrome

Abstract

Mutations in GJB2, which encodes Cx26, are one of the most common causes of inherited deafness in humans. More than 100 mutations have been identified scattered throughout the Cx26 protein, most of which cause nonsyndromic sensorineural deafness. In a subset of mutations, deafness is accompanied by hyperkeratotic skin disorders, which are typically severe and sometimes fatal. Many of these syndromic deafness mutations localize to the amino-terminal and first extracellular loop (E1) domains. Here, we examined two such mutations, A40V and G45E, which are positioned near the TM1/E1 boundary and are associated with keratitis ichthyosis deafness (KID) syndrome. Both of these mutants have been reported to form hemichannels that open aberrantly, leading to "leaky" cell membranes. Here, we quantified the Ca(2+) sensitivities and examined the biophysical properties of these mutants at macroscopic and single-channel levels. We find that A40V hemichannels show significantly impaired regulation by extracellular Ca(2+), increasing the likelihood of aberrant hemichannel opening as previously suggested. However, G45E hemichannels show only modest impairment in regulation by Ca(2+) and instead exhibit a substantial increase in permeability to Ca(2+). Using cysteine substitution and examination of accessibility to thiol-modifying reagents, we demonstrate that G45, but not A40, is a pore-lining residue. Both mutants function as cell-cell channels. The data suggest that G45E and A40V are hemichannel gain-of-function mutants that produce similar phenotypes, but by different underlying mechanisms. A40V produces leaky hemichannels, whereas G45E provides a route for excessive entry of Ca(2+). These aberrant properties, alone or in combination, can severely compromise cell integrity and lead to increased cell death.

Figure 1.

Regulation of WT Cx26 and A40V and G45E mutant hemichannels by voltage and extracellular Ca²⁺. Representative conductance-voltage (G-V) relationships are shown expressed in Xenopus oocytes at different external Ca²⁺ concentrations ranging from nominal (0 added) to 2 mM obtained by applying slow (3 min) voltage ramps from 50 to −100 mV to oocytes expressing WT Cx26 (A), G45E (B), or A40V (C). The holding potential before and after each ramp was maintained at −20 mV. All conductances were normalized to the maximum value measured in 0 Ca²⁺. In WT and both mutant hemichannels, conductance declined strongly with hyperpolarization except in low extracellular Ca²⁺. Increasing Ca²⁺ shifted activation in the depolarizing direction in all cases. However, both mutant hemichannels appeared less sensitive to the effects of Ca²⁺, particularly at low Ca²⁺ concentrations. This reduced sensitivity to Ca²⁺ was strongest in A40V. In addition, G45E hemichannels showed stronger voltage-dependent closure at positive voltages (up to 50 mV).

Figure 2.

Steady-state Ca²⁺ response curves show that A40V hemichannels are less sensitive to external Ca²⁺ than WT Cx26 and G45E hemichannels. (A) Examples of recordings of WT Cx26 (top) and A40V (bottom) hemichannel currents and effects of increasing concentrations of extracellular Ca²⁺ (0, 0.2, 0.3, 0.5, 1.0, 2.0, and 5.0 mM). Oocytes were voltage clamped and maintained at −40 mV throughout. (B) Summary of data for WT Cx26 (■), A40V (∇), and G45E (○) compiled from several experiments such as that shown in A. Each point is the mean ± SEM of ∼20 experiments. (C) Plot showing resting membrane potentials measured in oocytes 24–48 h after RNA injection for WT Cx26 (■), A40V (∇), and G45E (○). Each point represents a measurement from an individual oocyte. Horizontal lines represent means of −25.1 ± 9.4 mV, −13.1 ± 5.5 mV, and −22.3 ± 8.0 mV for WT Cx26, A40V, and G45E, respectively. (D) The left panel shows the mean holding currents for oocytes from C when initially clamped to −40 mV. A40V-expressing oocytes, which showed the lowest resting potentials, showed the largest holding currents. The same oocytes exposed to 0 added Ca²⁺ (right) showed similar maximum current levels, indicating similar levels of expression.

Figure 3.

G45E hemichannels exhibit increased permeability to Ca²⁺. (A) Representative currents elicited by a series of voltage steps applied to oocytes expressing WT Cx26, A40V, and G45E. Oocytes were voltage clamped to −20 mV. The applied voltage step protocol consisted of 10-s steps from 60 to −100 mV in intervals of 10 mV followed by a 5-s step to −110 mV. Each series was repeated in extracellular solutions containing 0.2 mM Ca²⁺, 1.8 mM Ca²⁺, and 1.8 mM Ba²⁺. Currents from oocytes expressing any one of the three hemichannel types were similar in low Ca²⁺ (0.2 mM). Upon raising Ca²⁺ to 1.8 mM, a large transient inward current developed in oocytes expressing G45E when the voltage was stepped to −110 mV after depolarizing steps that activated these hemichannels. Equimolar substitution of Ca²⁺ with Ba²⁺ abolished the large transient inward current. Also, the hemichannel currents in Ba²⁺ resembled those in low Ca²⁺. (B and C) The transient current observed in oocytes expressing G45E represents a Ca²⁺-activated chloride current. (B) A representative recording of the reversal potential of the transient current for an oocyte bathed in 100 mM NaCl and 0.2 mM Ca²⁺. At various voltages (as indicated), oocytes were briefly exposed to 1.8 mM Ca²⁺ (indicated by the black boxes). The transient current that developed (filled in gray) reversed between −20 and −30 mV and was followed by a decay in the hemichannel current that reversed upon returning to 0.2 mM Ca²⁺. This protocol was repeated in an external solution in which 100 mM NaCl was replaced with a 50:50 NaCl/NaAsp. (C) Plot of peak transient current after exposure to 1.8 mM Ca²⁺ in 100 mM NaCl and 50:50 NaCl/NaAsp. Reversal potential shifted in accordance with the change in the extracellular Cl concentration.

Figure 4.

WT Cx26 hemichannels are also permeable to Ca²⁺. (A and B) Currents from oocytes expressing G45E (A) and WT Cx26 (B) recorded 24 and 48 h after injection of RNA. At 48 h, recordings are shown in 1.8 and 5.0 mM extracellular Ca²⁺. Voltage step protocols are the same as in Fig. 3 A. At 48 h, oocytes expressing G45E show larger hemichannel currents in 1.8 Ca²⁺ (compared with 24 h) along with larger Ca²⁺-activated Cl currents. Upon raising Ca²⁺ to 5 mM, G45E hemichannels continued to activate at positive voltages and to induce Ca²⁺-activated Cl currents. With increased expression of WT Cx26 hemichannels, Ca²⁺-activated Cl currents were also induced, albeit more weakly in comparison to G45E. (C) Peak chloride current is plotted as a function of the membrane conductance. Measurements were obtained from the same voltage protocol as shown in A and B). The magnitude of the chloride current was measured at the peak that developed at −110 mV after the activating voltages steps. Baseline was taken as the instantaneous current upon stepping to −110 mV that preceded the development of the large inward transient. The membrane conductance was measured at the end of the activating voltage step, the bulk of which was caused by hemichannel activation. Open symbols are data from WT Cx26, and closed symbols are from G45E. Experiments from individual oocytes are plotted as different symbol shapes. Lines represent fits (by eye) to the data for illustration purposes. n = 8 oocytes for G45E, and n = 9 oocytes for WT Cx26.

Figure 5.

G45E hemichannels exhibit a larger unitary conductance. Representative examples of patch clamp recordings from inside-out patches containing single WT Cx26 (A), A40V (B), and G45E (C) hemichannels. The single hemichannel currents shown are in response to 8-s voltage ramps between ±70 mV and are leak subtracted (see Materials and methods). In contrast to other hemichannels that generally show inward current rectification in symmetric 140 mM KCl, WT Cx26 shows slight outward rectification. A recording of a single Cx50 hemichannel in the same solutions is included for comparison in A.

Figure 6.

Cysteine substitutions at A40 and G45 produce functional hemichannels, but G45C exhibits impaired function as a result of coordination of metal ions. (A–C) Current traces from oocytes are shown expressing WT Cx26 (A), A40C (B), and G45C (C). Oocytes were held at −40 mV throughout and were exposed to a low extracellular Ca²⁺ (0.2 mM) solution with and without 10 µM of the heavy metal chelator TPEN. Oocytes expressing G45C show little current in 0.2 mM Ca²⁺ in the absence of TPEN and a large potentiation >10-fold after addition of TPEN. A40C hemichannel currents are robust in the absence of TPEN, and a small increase occurred after addition. TPEN had no effect on WT Cx26 hemichannels, which were robust upon lowering Ca²⁺ to 0.2 mM. (D) Summary of the effects of TPEN. Each bar represents the fold change (mean ± SEM) in current after TPEN in 0.2 mM Ca²⁺ (n = 9 for WT Cx26, n = 17 for A40C, and n = 51 for G45C).

Figure 7.

SCAM experiments using different MTS reagents show robust and differential effects on G45C. (A and B) The effects of MTS reagents on hemichannel currents are shown from G45C (A) and A40C (B) recorded at −40 mV. Before the recordings shown, oocytes were exposed to a solution containing 3 mM EGTA (0 Ca²⁺) to open hemichannels and then were exposed to the same solution containing 0.2 mM MTSET (positively charged, left) or 2.0 mM MTSES (negatively charged, right). G45C currents showed rapid and robust response to both MTS reagents that were opposite in sign, decreasing for MTSET and increasing for MTSES. Both MTS reagents decreased A40C currents, but the effects were more modest and slow in time course. (C) Summary of the percent change in current (mean ± SEM) upon application of the MTS reagent. n = 5 for WT Cx26, n = 7 for A40C, and n = 24 for G45C for MTSET. n = 7 for WT Cx26, n = 7 for A40C, and n = 23 for G45C for MTSES.

Figure 8.

Single-channel SCAM indicates that G45 is a pore-lining residue. (A) Recordings of single G45C hemichannels in excised patches. The top trace is a recording from an inside-out patch held at −40 mV. Exposure to MTSET produced a stepwise reduction in current (discernable levels indicated by dashed lines). The bottom recording is from an outside-out patch held at 40 mV. Exposure to MTSES produced a stepwise increase in current (discernable levels indicated by dashed lines). C and O depict fully closed and open states in both traces. (B) Examples of G45C hemichannel currents before and after (3 min) application of MTS reagents. The left panel shows a single G45C hemichannel recording in the absence of added reagent. The solid line is a fit to the open hemichannel current. The middle and right panels show effects of MTS reagents. Solid black lines represent fits to the open hemichannel I-V relationships before MTS application. Data and fits (gray lines) are after MTS application. (C) Same as in B for single A40C hemichannels. All currents were elicited by 8-s voltage ramps between ±70 mV. Currents were leak subtracted as described in Materials and methods.

Figure 9.

A40V and G45E mutants make functional cell–cell channels. (A) Bar graph showing junctional conductances, g_j (mean ± SEM), measured in oocyte pairs injected with H₂O (control, n = 10), WT Cx26 (n = 12), A40V (n = 8), and G45E (n = 7). Coupling was robust for WT Cx26 and both mutants compared with the control. Representative examples of junctional currents (left) along with steady-state G_j-V_j relationships (right) obtained from WT Cx26 (n = 4), A40V (n = 3), and G45E (n = 4) cell pairs. G_j is g_j normalized to the maximum about V_j = 0. Records were taken from oocytes that were paired for shorter times to avoid series access resistance errors that accompany strong coupling. V_j steps between ±120 mV were applied in 10-mV increments. (B) Bar graph showing junctional conductances (mean ± SEM) measured in N2A cell pairs transfected with A40V and G45E (n = 42 for A40V, and n = 53 for G45E). Examples of junctional currents are shown below. V_j steps between ±120 mV were applied in 10-mV increments.