|
Figure 1. Connexin-selective inhibition of hemichannel currents by BQ+. (A) Membrane currents in a Cx50-expressing oocyte voltage clamped at −70 mV. Reduction of Ca2+ from 1.8 to 0.2 mM in a KCl solution promoted robust opening of Cx50 hemichannels, evident by the development of a large inward current. The application of 500 µM and 1 mM BQ+ in the continued presence of 0.2 mM Ca2+ caused a rapid and concentration-dependent reduction in current at −70 mV, an effect that was reversible upon washout. (B and C) Cx46 hemichannels were only moderately affected by 500 µM and 1 mM BQ+. Bar graph shows a comparison of the decrease in Cx46 (black bars) and Cx50 (gray bars) hemichannel currents caused by 500 µM and 1 mM BQ+. Because Cx46 activates only at positive voltages, tail current amplitudes were measured at −70 mV after 5-s depolarizing voltage steps to +50 mV. Each bar represents the mean ± SEM of six oocytes. (C) Selective inhibition by BQ+ of GJ channels formed by Cx50 but not Cx46. Bars represent the mean ± SEM of the decrease in junctional currents caused by the addition of BQ+ to patch pipettes (n ranged from 4 to 15). The decrease caused by BQ+ was measured after 12 min of achieving whole cell configuration.
|
|
Figure 2. BQ+ reduces the Po of Cx50 hemichannels in excised patches. Shown on top is a recording of a single Cx50 hemichannel in an outside-out configuration at a membrane potential of −50 mV. The application of 500 µM BQ+ to the extracellular face of the hemichannel caused a robust reduction in the Po of the hemichannel, evident by transitions of the hemichannel to the fully closed state (dotted line). The reduction in Po was reversible upon washout of BQ+. Amplitude histograms shown in the bottom panels indicate that the amplitude of the fully open state was not affected in the presence of 500 µM BQ+. Histograms were obtained from the current recording during the application of BQ+ (left) and after washout of BQ+ (right).
|
|
Figure 3. The decrease in Po of Cx50 hemichannels caused by BQ+ is concentration dependent and is a result of an increase in loop gating. Recordings of single Cx50 hemichannels before and after the addition of 300 µM, 500 µM, and 1 mM BQ+ applied to the extracellular face. 15-s segments of current traces recorded at a membrane potential of −70 mV along with corresponding all-points histograms obtained from longer (30-s) recordings (right of each trace) are shown. In the absence of BQ+, Cx50 hemichannels are predominantly open (O) at −70 mV, with occasional brief transitions to the fully closed state (C). The application of BQ+ caused a reduction in the Po, evident as an increase in the frequency of transitions to the fully closed state. Views of the current traces recorded with or without BQ+ at an expanded time scale. The boxed regions indicate the segments of the traces that were expanded. In the presence of BQ+, the transitions between open and closed states exhibit slow kinetics, indistinguishable from intrinsic gating events (see control trace) attributed to loop gating. Asterisks show more stable closing events in 1 mM BQ+.
|
|
Figure 4. BQ+ alters both open- and closed-time distributions of Cx50 hemichannels. (A) Representative open- and closed-time distributions at −70 mV in the absence (control) and the presence of 500 µM BQ+ plotted on a logarithmic abscissa. Lines are single-exponential fits to the histograms. (B) Concentration dependence of mean open (left) and mean closed (right) times.
|
|
Figure 5. Voltage-dependent inhibition of Cx50 hemichannels by extracellular BQ+ suggests a binding site in the pore. Recordings of single-hemichannel currents in excised patches in control (left) and in 500 µM BQ+ (right) at membrane potentials of +20, −30, −50, −70, and −90 mV are shown. BQ+ was applied to the extracellular side of the hemichannel. Dashed lines in the current traces indicate the closed state. In the absence of BQ+, Cx50 hemichannels predominantly reside in the open state at voltages between +20 and −70 mV. With further hyperpolarization, Cx50 hemichannels tend to gate to long-lived subconductance states (see trace at −90 mV). The application of 500 µM BQ+ to the extracellular side caused strong inhibition at large hyperpolarizing voltages, but little inhibition at smaller hyperpolarizing voltages (compare Po in BQ+ at −30 and −90 mV). An increase in inhibition with hyperpolarization, which tends to drive BQ+ into the hemichannel, is suggestive of a binding site within the aqueous pore.
|
|
Figure 6. Concentration and voltage dependence of the inhibition of single Cx50 hemichannels by BQ+ applied to the extracellular side. (A) Plots of fractional inhibition (FBQ) versus Vm at BQ+ concentrations of 300 µM, 500 µM, and 1 mM demonstrate the voltage dependence of inhibition. Solid lines are fits to a Boltzmann equation. The mean V1/2 values from Boltzmann analysis are −86, −53, and −30 mV, with slope factors of 26, 25, and 23 mV−1 for 300 µM, 500 µM, and 1 mM BQ+, respectively. (B) Plots of fractional inhibition (FBQ) versus BQ+ concentrations of −30, −50, −70, and −90 mV. The solid lines are fits to the Hill equation, with Kd values of 1.14, 0.556, 0.38, and 0.24 mM and Hill slopes of 1.6, 1.7, 1.9, and 1.7 at −30, −50, −70, and −90 mV, respectively.
|
|
Figure 7. Voltage dependence of inhibition by BQ+ applied to the cytoplasmic side of Cx50 hemichannels. Single-channel I-V relationships in control (top), in 500 µM BQ+ (middle), and in 1 mM BQ+ (bottom) applied either to the cytoplasmic side (A) or extracellular side (B) of Cx50 hemichannels. I-V relations in A and B were obtained in response to 8-s voltage ramps from −100 to +50 mV applied to excised inside-out patches and outside patches, respectively. Inhibition produced by cytoplasmic application of BQ+ (BQ+cyt) is promoted by depolarization, a voltage dependence that is opposite to that observed when BQ+ is applied to the extracellular side, indicating a binding site located in the pore. In addition, inhibition by intracellular BQ+, especially at a 1-mM concentration, occurs even at strong inside-negative voltages. Hyperpolarizing voltages are expected to prevent entry of BQ+ on the intracellular side into the pore, suggesting that the binding site is closer to the cytoplasmic end of the hemichannel. Solid lines superimposed on the current traces in BQ+ represent exponential fits to the open-state current in the absence of BQ+ to illustrate the decrease in conductance at high, but not lower, BQ+ concentrations (see Results).
|
|
Figure 8. Differences in BQ+ action when applied to the cytoplasmic and the extracellular sides. Recordings of a single Cx50 hemichannel with BQ+ added to the cytoplasmic side (A) and the extracellular side (B) obtained from separate patches are shown. Membrane potential was held constant at −30 mV in A and −50 mV in B. Regardless of the side of application, BQ+ increased loop-gating transitions. All transitions between the open and closed state are slow in kinetics, typical of loop gating, as shown in the expanded views of the current traces recorded at 500 µM BQ+ applied to the cytoplasmic and extracellular side (see boxed regions). The solid black line indicates the segments of the traces that were expanded. Notably, the noisy fluctuations from the closed state (dotted line) that characterized BQ+ action when applied to the extracellular side were much less prominent when BQ+ was applied to the cytoplasmic side. Currents were filtered at 1 kHz, and data were acquired at 5 kHz.
|
|
Figure 9. The NT contributes to the difference in the sensitivity of Cx46 and Cx50 hemichannels to BQ+. Concentration–response curves for the inhibition of macroscopic hemichannel currents in oocytes expressing Cx50 wt (closed squares), Cx46 wt (closed circles), Cx46*50NT (open squares), and Cx50*46NT (open circles). Oocytes were voltage clamped at −70 mV. Current magnitude in oocytes expressing Cx46*50NT, in which Cx46 NT sequence was replaced with that of Cx50, was substantially reduced by 0.5 mM to 3 mM BQ+. The reciprocal chimera, Cx50*46NT, was similar to Cx46 in being largely insensitive to BQ+ at all concentrations except 3 mM. Each point represents the mean ± SEM from four different oocytes
|
|
Figure 10. Illustration of BQ+ in the cytoplasmic vestibule of the connexin pore. The crystal structure of Cx26 by Maeda et al., (2009) with BQ+ placed in the wide vestibule constituted by NT domain, as suggested by the chimeric data, is shown. For clarity, only two of the six connexin subunits are illustrated. The NT and the TM2 domains of one of the subunits are shown in cyan and yellow, respectively. The pore funnel is formed by the NT TM1 and E1 domains, with the narrowest part ∼14.7 Å formed by amino-terminal ends of NT. Pore diameter at the cytoplasmic vestibule, constituted by the NT extending into TM2, widens to ∼40 Å. BQ+ (white) is shown in this wide vestibule. The placement and orientation of BQ+ is arbitrary. The image was prepared using PyMOL (http://www.pymol.org).
|
