|
Figure 1. (A) Macroscopic current traces obtained from WT Cx32*Cx43E1 unapposed hemichannels after polarizations in 20-mV increments between −90 and 50 mV from a holding potential of 0 mV. The small increase in current level at large positive voltages results from the activation of an endogenous oocyte channel as assessed by recording uninjected oocytes. (B) Macroscopic current traces obtained from an uninjected oocyte illustrating typical endogenous currents present in oocytes obtained from Xenopus 1 frogs in Cs-MES bath solutions. There is variability in the form and magnitude of endogenous currents among oocytes, but in most cases endogenous currents contribute <300 nA.
|
|
Figure 2. G45C and V38C residues are modified by MTSEA–biotin-X. Segments of patch clamp records of G45C and V38C after the addition of MTSEA–biotin-X to the bath solution at a final concentration of 1 or 0.5 mM. Arrows at the bottom of each panel mark step changes in conductance that are consistent with the reaction of a cysteine residue with MTSEA–biotin-X. Changes in current levels were determined with the all-point histograms shown on the right side of each panel. (A) Inside-out recording of two G45C channels. Nine conductance changes are evident, consistent with modification of 9 of 12 available G45C residues. The first modification event occurred 20 s after the bath application of 1 mM MTSEA–biotin-X. (B) Outside-out recording of a single G45C channel illustrating five conductance changes, consistent with modification of five of six available cysteine residues. The first modification event occurred 6 s after the bath application of 1 mM MTSEA–biotin-X. (C) Inside-out recording of a single V38C channel illustrating four conductance changes consistent with modification of four of six available subunits. The first modification event occurred 18 s after the bath application of 1 mM MTSEA–biotin-X. (D) Outside-out recording of a single V38C channel illustrating six conductance changes. It is not clear if the final conductance change, marked by an asterisk, resulted from a reaction, gating event, or channel loss; thus, at least five of six cysteine residues are modified. The first modification event occurred 64 s after the bath application of 0.5 mM MTSEA–biotin-X. The longer time to the first modification event correlates with the lower concentration of MTSEA–biotin-X used in this record (0.5 mM vs. 1.0 mM) than was used in the previous records. (D1) This is an enlargement of the boxed region of D. The event marked by the single asterisk is a Vj-gating transition that occurred after modification of two cysteine residues. The event marked by two asterisks is a loop-gating transition that occurred after three cysteine subunits were modified by MTSEA–biotin-X. Note that the loop- or slow-gating event involves a series of small amplitude transitions giving the appearance of a complex gating event with a measurable time constant, whereas the Vj-gating event occurs as a single fast step between the open state and a subconductance state. Zero current level indicated by the dashed line is the leak-subtracted current.
|
|
Figure 3. DTT and cadmium alter levels of A43C currents. Macroscopic currents elicited from an oocyte expressing A43C unapposed hemichannels with the voltage paradigm shown at the top of the panel. The central bar indicates the time and duration at which the bath solution containing 100 mM Cs-MES, 1.8 mM CaCl2, and 10 mM HEPES, pH 7.6, was exchanged with the same bath solution except containing either 20 µM DTT or 10 µM CdCl2. Currents increased to a steady-state level approximately fivefold greater than the initial currents after exposure to 20 µM DTT and decreased to ∼80% of maximum after treatment with 10 µM CdCl2. The reduction in current was not changed substantially by washing with the bath solution but was reversed to pre-cadmium levels after wash with 20 µM DTT.
|
|
Figure 4. Western blots of WT (Cx32*Cx43E1) and mutant (A43C) membrane-inserted hemichannels. + and − symbols at the bottom of each panel denote treatment of the sample with or without either 50 mM DTT or 10 mM TPEN before SDS-electrophoresis through 5–40% gradient polyacrylamide gels. The position of pre-stained molecular weight standards (Thermo Fisher Scientific) are presented as bars on the right side of the figure. The band with molecular weight ∼50 kD corresponds to a connexin dimer, whereas the monomer has an electrophoretic mobility comparable to a molecular weight standard of ∼27 kD. Only treatment with DTT reduces the dimer to the monomeric connexin form.
|
|
Figure 5. Cadmium locks A43C unapposed hemichannels in a closed state. (A) Macroscopic currents elicited from an oocyte expressing A43C unapposed hemichannels with the voltage paradigm, steps from −90 to 0 mV, shown at the top of the panel. The central bar indicates the time and duration for which the bath solution containing 100 mM Cs-MES, 1.8 mM CaCl2, and 10 mM HEPES, pH 7.6, was exchanged with the same bath solution except containing either 20 µM DTT or 10 µM CdCl2. Currents were decreased to ∼85% of maximum levels after treatment with 10 µM CdCl2. The reduction in current could only be reversed after a second exposure to 20 µM DTT. (B) 10 µM CdCl2 was applied to the channel after a long-duration hyperpolarizing step that would favor closure of both loop and Vj gates. After wash with Cs-MES bath solution, the extent of current reduction was assessed by a series of polarizing steps between −90 and 50 mV. Currents were reduced by ∼80% by Cd2+ treatment when the channels resided in a closed state. Currents could only be recovered fully after exposure to 20 µM DTT. (C) Macroscopic currents elicited from an oocyte expressing A43C unapposed hemichannels with the voltage paradigm, steps between 0 and 30 mV, which strongly favors population of the open-channel state. Application of 10 µM CdCl2 had no effect on the level of A43C current, indicating that A43C residues do not coordinate Cd2+ when the channel resides in the open state.
|
|
Figure 6. Modification of A43C residues by bBBr occurs at voltages that favor channel closure. (A) Macroscopic currents attributable to A43C channels are irreversibly decreased when exposed to bBBr during a voltage paradigm, steps between 0 and −90 mV, which elicits gating transitions between the open (0 mV) and either loop- and/or Vj-gating closed states (−90 mV). (B) Application of bBBr during a voltage paradigm, steps from 0 to 30 mV, which favors population of the open-channel state, has no effect on A43C macroscopic currents. The central bar in both panels indicates the time and duration for which the bath solution containing 500 µM TCEP was exchanged with the same bath solution containing 1 mM bBBr and subsequently washed with TCEP containing bath solution.
|
|
Figure 7. Effect of 10 µM CdCl2 on heteromeric channels formed by coexpressing 1:1 mixtures of WT and A43C RNA. (A) Heteromeric channels were treated with 20 µM DTT and washed with Cs-MES, 1.8 mM Ca2+ bath solution before the application of Cd2+ shown in the trace segment. After the reduction of macroscopic currents by treatment with 10 µM CdCl2, currents were restored to 75% of pre-cadmium levels by washing with cadmium-free bath solution. The result is interpreted to indicate that 25% of channels form a high affinity cadmium site that “locks” the channel in a closed conformation (see Results). (B) Comparison of the kinetics of channel activation upon depolarization to 50 mV before (black trace), during (green trace), and after (red trace) the application of Cd2+ in the trace shown in A. The two current traces before Cd2+ were averaged, as were the final two current traces after wash. The green trace is current trace obtained just before wash with Cd2+-free solution. The similarity among normalized traces indicates that the differences in current levels are not a consequence of differences in the kinetics of activation, but most likely reflects the proportion of channels that can be activated by the voltage step.
|
|
Figure 8. Unapposed N2E+A43C hemichannels are locked by cadmium and accessible to bBBr modification when the channel is closed by loop gating, but not Vj gating. (A) Macroscopic current traces obtained from N2E+A43C unapposed hemichannels after polarizations to voltages shown in the inset of 0 mV. Current relaxations at hyperpolarizing and depolarizing potentials result from the closure of loop and Vj gates, respectively. (B) Macroscopic currents were recorded from oocytes expressing the double mutation, N2E+A43C. The N2E mutation reverses the polarity of Vj gating from closure at hyperpolarizing potentials to closure at depolarizing potentials but does not change the polarity of loop gating. The application of 10 µM CdCl2 during a series of voltage steps between 0 and −90 mV causes a large decrease in membrane currents that can only be reversed by treatment with 20 µM DTT. Polarizations to −90 mV strongly favor the closure of loop gates, whereas residency in the open state is favored at a holding potential of 0 mV. (C) Macroscopic currents recorded from N2E+A43C oocyte using a similar protocol to that used in Fig. 5 B. 10 µM CdCl2 was applied to the channel after a long-duration depolarizing step to 50 mV that strongly favors closure of only the Vj gates. After wash with Cs-MES bath solution, the effect of cadmium on Vj gate closed channels was assessed by the application of a series of polarizing steps between 0 and 50 mV. This polarization elicits gating transitions between the open and Vj-gating closed states. Currents were similar to those obtained before cadmium application, demonstrating that A43C residues do not coordinate cadmium when the channel is closed by Vj gating. (D) Macroscopic current traces elicited by repeated polarizations between 0 and −90 mV, a voltage paradigm that elicits transitions between open and loop gate closed channels. At the position indicated, channels were exposed to 1 mM bBBr in the presence of 500 µM TCEP and subsequently washed with 500 µM TCEP. The observed reduction in current is consistent with a reaction between the A43C residue and bBBr. (E) Macroscopic current traces elicited by repeated polarizations between 0 and 50 mV, a voltage paradigm that results in the opening and closing of Vj gates. At the position indicated, channels were exposed to 1 mM bBBr in the presence of 500 µM TCEP and subsequently washed with 500 µM TCEP. The absence of any reduction in current is interpreted to indicate the inaccessibility of A43C residues to bBBr modification when the N2E+A43C channel resides in the open state and the Vj-gated closed state.
|
|
Figure 9. Cadmium and bBBr reduce A40C currents. (A) Reduction of A40C unapposed hemichannel currents by CdCl2 is reversed by washing. Macroscopic currents attributable to A40C residues were examined by applying a series of voltage polarizations between −90 and 50 mV. The application of 10 µM CdCl2 during this voltage paradigm causes a ∼90% reduction in current in the case shown. However, unlike A43C channels, currents are fully restored by wash with cadmium-free bath solution. These results are interpreted to indicate that A40C residues do not form a high affinity cadmium binding site when the channel is closed by either loop or Vj gating. (B) A40C unapposed hemichannels are accessible to bBBr modification. BBBr was applied to the channels at the time indicated in the bar line as the channel was repeatedly stepped from −90 to 50 mV. Currents were reduced by ∼50% at positive potentials and were not reversed by wash.
|
|
Figure 10. Helical wheel representation of reactive residues at the TM1/E1 border of Cx46 and Cx32*Cx43E1. (A) Helical wheel representation of residues R33 through G46 of Cx46. The residues span the TM1/E1 border, with residue E42 believed to be located on the membrane border. Residues accessible to MTS reagents applied to both the intercellular and extracellular face of the unapposed hemichannel are shown in red. These residues reside on one side of the helix, subtend an arc of ∼80°, and line the pore of the open Cx46 unapposed hemichannel. Residue R33 is topmost in the figure, with the residue number increasing counterclockwise. Data are taken from Kronengold et al. (2003). (B) Helical wheel representation of homologous residues R32 through G45 of Cx32*Cx43E1 that line the pore of the open unapposed Cx32*Cx43E1 channel. Residues examined in this study are marked with asterisks. Two residues, V38C and G45C, react with MTSEA–biotin-X when the reagent is applied from either the intracellular or extracellular face of the channel and consequently are assigned as pore lining. Expression levels of S42C and E41C were insufficient to allow studies of accessibility to thiol modifying reagents, whereas W44C did not express membrane current. These residues are marked by two asterisks. (C) Helical wheel representation of the loop gate closed Cx32*Cx43E1 channel. The helical wheel pictured in B was rotated 140° clockwise to position residues A40C and A43C. This position accounts for the observed sensitivity of residues A40C and A43C to Cd2+ and their modification by bBBr when the channel is closed by loop gating (see Discussion).
|
