Conductance and permeability of the residual state of connexin43 gap junction channels

Abstract

We used cell lines expressing wild-type connexin43 and connexin43 fused with the enhanced green fluorescent protein (Cx43-EGFP) to examine conductance and perm-selectivity of the residual state of Cx43 homotypic and Cx43/Cx43-EGFP heterotypic gap junction channels. Each hemichannel in Cx43 cell-cell channel possesses two gates: a fast gate that closes channels to the residual state and a slow gate that fully closes channels; the transjunctional voltage (V(j)) closes the fast gate in the hemichannel that is on the relatively negative side. Here, we demonstrate macroscopically and at the single-channel level that the I-V relationship of the residual state rectifies, exhibiting higher conductance at higher V(j)s that are negative on the side of gated hemichannel. The degree of rectification increases when Cl(-) is replaced by Asp(-) and decreases when K(+) is replaced by TEA(+). These data are consistent with an increased anionic selectivity of the residual state. The V(j)-gated channel is not permeable to monovalent positively and negatively charged dyes, which are readily permeable through the fully open channel. These data indicate that a narrowing of the channel pore accompanies gating to the residual state. We suggest that the fast gate operates through a conformational change that introduces positive charge at the cytoplasmic vestibule of the gated hemichannel, thereby producing current rectification, increased anionic selectivity, and a narrowing of channel pore that is largely responsible for reducing channel conductance and restricting dye transfer. Consequently, the fast V(j)-sensitive gating mechanism can serve as a selectivity filter, which allows electrical coupling but limits metabolic communication.

Figure 1.

V_j gating of heterotypic Cx43/Cx43-EGFP junctions formed between Novikoff and HeLaCx43-EGFP cells. (A) Phase-contrast (top) and fluorescent (bottom) images of a Novikoff/HeLaCx43-EGFP cell pair. The fluorescent image shows diffuse distribution of Cx43-EGFP in the plasma membrane of the HeLaCx43-EGFP cell and punctate staining representing junctional plaques (arrow) in the region of contact with the Novikoff cell. (B) G_j-V_j dependence of Cx43/Cx43-EGFP heterotypic junctions with data pooled from 19 cell pairs. G_j represent g_j normalized to its value at V_j = 0 mV. For each data point, g_j was measured at the end of a 30-s V_j step. Positive and negative V_js correspond to the Novikoff cell made relatively positive and negative, respectively. The solid lines for each polarity of V_j are fits of the data to the Boltzmann relation. The dashed line shows the Boltzmann G_j-V_j relation of homotypic Cx43 junctions (Bukauskas et al., 2001). (C) Selected examples illustrate differences in the kinetics of V_j dependence at positive and negative V_j steps applied to the Novikoff cell. Repeated ±14-mV pulses (inset) were applied between and on the top of V_j steps of −80, −100, and +100 mV to monitor junctional conductance. (D) I-V scatter plots obtained from the intervals I–IV indicated in C. The I-V plot obtained from time interval I corresponds to the open state and is linear. The I-V plots obtained from intervals II (red line) and III (green line) correspond to the residual state of the Cx43 hemichannel and show a small degree of rectification. The I-V (blue) plot obtained from interval IV (blue line) also corresponds to the open state, but shows variability among individual ramps as a result of continued decline in conductance with slow gating. (E) Corresponding g_jslope-V_j scatter plots from the intervals I–IV shown in C. Solid lines are linear regressions illustrating the lack of V_j dependence for channels in the open state and modest V_j dependence for channels in the residual state.

Figure 7.

Intercellular transfer of Alexa Fluor, a negatively charged dye, is restricted when channels are in the residual state. (A–C) Phase-contrast (A) and fluorescence (B and C) images showing a Novikoff cell pair in which dye transfer and coupling were examined in open and residual states. Locations of pipettes 1 and 2 used for dual voltage clamp recording and pipette 3 used for loading cell 1 with dye are indicated. RI-1, RI-2, and RI-3 are regions of interest from which fluorescence intensities were measured. Fluorescence intensities in the cells were calculated by subtracting the background fluorescence (RI-3) from the fluorescence measured in RI-1 and RI-2. (D) Plot of fluorescence intensity normalized to the maximum (NFI) in cell 1 and cell 2 over time. NFI in cell 1 rises upon opening the patch in pipette 3 (arrow) and reaches a plateau after ∼70 s. NFI in cell 2 shows little change during the time V_j is imposed when channels mainly reside in the residual state. Upon reopening channels by removal of V_j, NFI begins to rise immediately and reaches ∼14% of the maximum within 60 s (inset shows an expanded scale of NFI). Concomitant with an increase in NFI in cell 2, there is a decrease in cell 1, presumably due to rapid dye transfer to cell 2. Reimposition of V_j caused an immediate decline in NFI in cell 2 due to loss of transfer from cell 1 and dialysis with patch pipette 2. (E) Records of I_j and V₂ over time corresponding to fluorescence plot in D. V_j of +90 mV was applied to cell 2 which caused g_j to decline to a steady-state value of ∼10 nS. Between 90-mV V_j steps, small repeated ±10-mV V_j steps were applied to cell 2 to assess g_j, which remained constant at 43 nS. Conductance recovered rapidly upon removal of the +90-mV V_j step. Arrow indicates opening of the patch in pipette 3.

Figure 2.

I_j-V_j relationship of a single-channel recorded in Novikoff/HeLaCx43-EGFP cell pair. (A) The voltage protocol applied to the Novikoff cell consisted of repeated (1 s) ramps from −120 to +120 mV initiated and terminated by 50-ms epochs at −120 and +120 mV. The I_j record shows gating transitions between open and residual states only at negative V_j. The channel typically reopens when V_j approaches 0 mV and stays open at positive V_j. (B) I_j-V_j scatter plot of all data points from A. (C) Representative example of a V_j polarity reversal protocol used to examine the residual state current at positive V_js. Stepping to +85 mV shows the open channel current giving γ_open = 115 pS; no gating to the residual substate is observed. Reversal to V_j = −85 mV results in equal, but opposite, current giving the same γ_open before eventually gating to the residual state (γ_res = 28 pS, dashed line). Upon reversal back to V_j = +85 mV, the channel briefly remains in the residual state before opening, and shows a reduced γ_res of 15 pS (dotted line). (D) I-V_j scatter plot obtained from the record presented in C. The solid line is a fit of the data by using single exponential function, I_jres = I_o · (exp(b · V_j) − 1). The fitting parameters were as follows: I_o = 3.3 ± 0.4 pA, b = −0.007 ± 0.001 mV⁻¹ (n = 2,489); γ_res of Cx43/Cx43-EGFP channel rectifies decreasing nearly twofold when V_j changes from −100 to 100 mV.

Figure 3.

Summarized I_j-V_j plots for open and residual states obtained from voltage ramp and step protocols. (A) I_j-V_j scatter plot of the open state (35 ramps from four different cell pairs). The slope of a linear regression (solid line) gives γ_open= 115 pS (r² = 0.99; n = 9,200 data points). (B) I_j-V_j scatter plot for the residual state obtained from 25 records using the voltage ramp protocol and 12 records using the voltage step protocol. The solid line is an exponential fit of the data. Fitting parameters are as follows: I_o= 3.5 ± 0.2 pA, b = −0.0053 ± 0.0002 mV⁻¹ (n = 18,500 data points). (C) A γ_res/γ_open-V_j plot shows that the ratio γ_res/γ_open decreases nearly twofold over a ±100-mV V_j range.

Figure 4.

Residual conductance dependence on V_j in Cx43 homotypic channels. (A) I_j record obtained in the late stage of CO₂-induced uncoupling in a HelaCx43 cell pair when only a single channel was operating. The holding potential in cell 1 was −75 mV, and in cell 2 was −5 mV; repeated pulses of +105 mV were applied to cell 1. V_j of −70 mV was maintained throughout except for intermittent pulses. The channel predominantly resides in the residual state at V_j = −70 (dashed line). Upon V_j reversal to +35 mV, the channel briefly remains in γ_res before opening fully (lower solid line). At the end of this record, the channel closed fully (i.e., to a nonconducting state), as evidenced by a lack of a change in I_j in response to the last V_j step. Inset shows I_j during V_j reversal at an extended time scale. I_j at V_j = +35 mV gave a residual conductance of ∼18 pS (dotted line) compared with γ_res of ∼30 pS at V_j = −70 mV (dashed line). (B) I_j-V_j scatter plot of the residual state obtained from the data shown in A and three other similar experiments. The solid line shows the fit of the data to a single exponential function. The fitting parameters were as follows: I_o = 1.5 ± 0.5 pA, and b = −0.011 ± 0.003 mV⁻¹ (n = 894 data points).

Figure 5.

Rectification of the residual state in symmetric KAsp. (A) Single-channel currents in a Novikoff cell pair in response to repeated V_js ramps (from −85 to 85 mV) demonstrates gating transitions between open and residual states. (B) Summarized I_j-V_j scatter plot of data points from the open state. A linear regression (solid line) gives an open channel conductance of 46 pS (n = 9,100 data points; r² = 0.99). (C) Summarized I_j-V_j scatter plot of data points from the residual state (collected from five cell pairs). A fit to a single exponential function gave parameters as follows: I_o = 0.22 ± 0.03 pA, and b = 0.016 ± 0.002 mV⁻¹ (n = 2,920 data points; solid line). (E) γ_res/γ_open-V_j scatter plot calculated from the fitted curves shown in B and C. γ_res/γ_open declines from ∼0.18 to ∼0.07 when V_j changes from −100 to +6 mV.

Figure 6.

Rectification of the residual state in symmetric TEACl. (A) Single-channel currents in a Novikoff cell pair in response to repeated V_j ramps (from −85 to 85 mV) demonstrate gating transitions between open and residual states. (B) I_j-V_j scatter plot of data points corresponding to the open state collected from four cell pairs. The slope of the regression line (solid line) gives a conductance of 40 pS (n = 10,328 data points; r² = 0.99). (C) I_j-V_j scatter plot for the residual state collected from four cell pairs shows little rectification. The fitting parameters to a single exponential function were as follows: I_o = 0.22 ± 0.03 pA, and b = −0.016 ±0.002 mV⁻¹ (n = 3,584 data points; solid line). (D) γ_res/γ_open-V_j scatter plot calculated from the fitted curves shown in B and C. γ_res/γ_open declines from ∼0.25 to ∼0.21 when V_j changes from −100 to +100 mV.

Figure 8.

Intercellular transfer of ethidium bromide, a positively charged dye, is restricted when channels are in the residual state. (A) Schematic of a cell pair and the arrangement of pipettes used for recording and dye-loading. Fluorescence was measured in Novikoff cell pairs as described in Fig. 7. (B) Plot of NFI in cell 1 and cell 2 over time. NFI in the cell 1 increases upon opening the patch in pipette 3 (solid arrow) and approaches a plateau in ∼100 s. NFI in cell 2 shows no change during the time a V_j step of +90 mV was imposed and channels mainly reside in the residual state. Upon reopening channels by removal of V_j, NFI begins to rise and reaches ∼5% of the maximum within 150 s (inset shows an expanded scale of NFI; horizontal dotted line indicates zero fluorescence level). (C) Records of I_j and V₂ over time corresponding to fluorescence plot in B. V_j of +90 mV was applied to cell 2, which caused g_j to decline to a steady-state value of ∼9 nS. Small repeated ±12-mV V_j steps were used to assess g_j, which recovered rapidly up to 35 nS level upon a removal of long +90 mV V_j step.

Figure 9.

(A) Summarized γ_res/γ_open-V_j plots (from experimental data) for three different pipette solutions. Rectification is steeper in symmetric KAsp and reduced in symmetric TEACl. (B) Plot of single-channel conductance versus conductivity of the pipette solutions. Closed and open circles correspond to γ_open and γ_res, respectively. Closed and open triangles are data for γ_open and γ_res, respectively, in symmetric TEA⁺Asp⁻ taken from Valiunas et al. (1997). Solid and dashed lines are regression lines of the second order for γ_open and γ_res, respectively; dotted lines show confidential interval.

Figure 10.

Introduction of positive charge at the cytoplasmic end of a gated Cx43 hemichannel can account for the current rectification that arises upon gating of a Cx43 channel to the residual state. (A) Schematic presentation of a channel pore as a cylinder 100 Å in length and 10 Å in diameter. The location of the charge introduced with gating is illustrated as a square, and z indicates the number of elementary charges. (B) PNP generated I-V curve for the channel in symmetric 140 mM KCl. Voltages are positive and negative relative to the gated side. Diffusion coefficients were 1.96 × 10⁻⁵ cm²/s for K⁺ and 2.03 × 10⁻⁵ cm²/s for Cl⁻. The solid line corresponds to z = 0. Curves with open symbols were generated with z = +6, one charge per subunit, located at different positions along the length of the pore; the symbols correspond to those indicated in A. The direction of I_j rectification is consistent with the experimental data and increases as the charge is moved toward the cytoplasmic end of the channel. The inset shows the ratio between fluxes for Cl⁻ and K⁺ ions for the I-V curve indicated by open circles. The I-V curve with closed circles was generated with z = −6. (C) Charge profiles for a homotypic Cx43 channel. For the open channel (solid line), each hemichannel consists of positive charge located toward the cytoplasmic end and more centrally located negative charge (representing charges in at the M1/E1 border). Gating to the residual state is illustrated as an increase in positive charge in NT (dashed line). (D) PNP-generated I-V curves of open (solid line) and residual (dashed line) states for the charge profile shown in C. (E) PNP-generated plots of the ratio (γ_res/γ_open) calculated for the charge profile shown in C for a channel in symmetric KCl (solid line), KAsp (dashed line), and TEACl (dash-dot line). (F) The same as E, except that the charge profiles contained distributed positive or negative charge (z = ±0.2) corresponding to interaction of TEA and Asp, respectively, with the channel pore.