Oh S et al. (2000), Stoichiometry of transjunctional voltage-gating...

XB-ART-10763

J Gen Physiol 2000 Jul 01;1161:13-31.

Show Gene links Show Anatomy links

Stoichiometry of transjunctional voltage-gating polarity reversal by a negative charge substitution in the amino terminus of a connexin32 chimera.

Oh S , Abrams CK , Verselis VK , Bargiello TA .

???displayArticle.abstract???
Gap junctions are intercellular channels formed by the serial, head to head arrangement of two hemichannels. Each hemichannel is an oligomer of six protein subunits, which in vertebrates are encoded by the connexin gene family. All intercellular channels formed by connexins are sensitive to the relative difference in the membrane potential between coupled cells, the transjunctional voltage (Vj), and gate by the separate action of their component hemichannels (Harris, A.L., D.C. Spray, and M.V. Bennett. 1981. J. Gen. Physiol. 77:95-117). We reported previously that the polarity of Vj dependence is opposite for hemichannels formed by two closely related connexins, Cx32 and Cx26, when they are paired to form intercellular channels (Verselis, V.K., C.S. Ginter, and T.A. Bargiello. 1994. Nature. 368:348-351). The opposite gating polarity is due to a difference in the charge of the second amino acid. Negative charge substitutions of the neutral asparagine residue present in wild-type Cx32 (Cx32N2E or Cx32N2D) reverse the gating polarity of Cx32 hemichannels from closure at negative Vj to closure at positive Vj. In this paper, we further examine the mechanism of polarity reversal by determining the gating polarity of a chimeric connexin, in which the first extracellular loop (E1) of Cx32 is replaced with that of Cx43 (Cx43E1). The resulting chimera, Cx32*Cx43E1, forms conductive hemichannels when expressed in single Xenopus oocytes and intercellular channels in pairs of oocytes (Pfahnl, A., X.W. Zhou, R. Werner, and G. Dahl. 1997. Pflügers Arch. 433:733-779). We demonstrate that the polarity of Vj dependence of Cx32*Cx43E1 hemichannels in intercellular pairings is the same as that of wild-type Cx32 hemichannels and is reversed by the N2E substitution. In records of single intercellular channels, Vj dependence is characterized by gating transitions between fully open and subconductance levels. Comparable transitions are observed in Cx32*Cx43E1 conductive hemichannels at negative membrane potentials and the polarity of these transitions is reversed by the N2E substitution. We conclude that the mechanism of Vj dependence of intercellular channels is conserved in conductive hemichannels and term the process Vj gating. Heteromeric conductive hemichannels comprised of Cx32*Cx43E1 and Cx32N2E*Cx43E1 subunits display bipolar Vj gating, closing to substates at both positive and negative membrane potentials. The number of bipolar hemichannels observed in cells expressing mixtures of the two connexin subunits coincides with the number of hemichannels that are expected to contain a single oppositely charged subunit. We conclude that the movement of the voltage sensor in a single connexin subunit is sufficient to initiate Vj gating. We further suggest that Vj gating results from conformational changes in individual connexin subunits rather than by a concerted change in the conformation of all six subunits.

???displayArticle.pubmedLink??? 10871637
???displayArticle.pmcLink??? PMC2229612

???displayArticle.grants??? [+]

Species referenced: Xenopus
Genes referenced: dtl gja1 gjb1 gjb2

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

	Figure 1. Representative conductance–voltage relations and macroscopic current traces of intercellular channels formed by the designated wild-type and chimeric connexins expressed in pairs of Xenopus oocytes. In all cases, junctional conductance is normalized to Vj = 0. Positive Vj is relative to the cytoplasmic entrance of the hemichannel appearing on the right side of the channel designation. Initial conductance is depicted by ▾, steady state conductance by ▿.
	Figure 2. Macroscopic recordings of Xenopus oocytes expressing Cx32Cx43E1 and Cx32N2ECx43E1 conductive hemichannels. (A) Outward currents in oocytes expressing Cx32Cx43E1 were elicited from a holding potential at −90 mV, with 30-s voltage steps ranging from +20 to +60 mV in 10-mV increments. Inward tail currents resulted from the repolarization of the membrane to −90 mV. In 5 mM Ca2+-containing bath solution, tail currents show an initial increase followed by a slow return to base line with a multiexponential time course. (B) The time course of the tail current relaxations is prolonged by reductions in the Ca2+ concentration of the bath solution from 5 to 2 mM. (C) Tail currents elicited by repolarization of the membrane to −90 mV in bath solutions containing 2 mM Ba2+ and no added Ca2+. The replacement of Ca2+ with Ba2+ alters the kinetics of tail current relaxations and removes initial rise in tail currents. It is likely that the ionic substitution reduces the contribution of Ca2+-activated Cl− currents. (D) Outward currents in Cx32N2ECx43E1 oocytes were elicited from a holding potential at −90 mV, with 30-s voltage steps ranging from 0 to +40 mV in 10-mV increments. Inward tail currents resulted from the repolarization of the membrane to −90 mV. The bath solutions contained 5 mM Ca2+. At voltages exceeding +30 mV, currents plateau consistent with the reversal of gating polarity by the N2E substitution. The time constants of tail-current relaxations are substantially faster than those observed in Cx32*Cx43E1 conductive hemichannels. (Inset) Current traces elicited by positive voltage steps are expanded.
	Figure 3. Voltage dependence of rapid gating transitions in single Cx32Cx43E1 conductive hemichannels. (A) Cell-attached patch recordings of a single Cx32Cx43E1 conductive hemichannel at different voltages, showing that the frequency of rapid transitions increases at larger hyperpolarizing potentials. The holding potential was flipped at the dotted vertical line from the hyperpolarizing to the depolarizing potential denoted. Leak currents were not subtracted. (B) An event list histogram of closed times at −70 mV. The histogram is best fit by a single exponential with a time constant, τ = 1.5 ms.
	Figure 4. Voltage-dependent gating of a Cx32Cx43E1 conductive hemichannel. (A) A cell-attached recording of a single Cx32Cx43E1 conductive hemichannel at −80 mV. In addition to the rapid flickers (Fig. 3), the channel resides at longer-lived subconductance levels, with durations that range from tens of milliseconds (shown on expanded trace) to tens of seconds. (B) An event list histogram of closures with duration longer than 50 ms at −80 mV. The histogram is best fit by a double exponential with time constants, τ1 = 56.8 ms and τ2 = 593 ms. (Inset) The portion of the histogram from 0 to 2 s is shown.
	Figure 5. Single-channel recordings of Cx32Cx43E1 conductive hemichannel illustrating three different gating mechanisms. (A) Vj gating is illustrated by the outside-out patch recording at −100 mV. The channel can enter one of three subconductance states denoted as S1, S2, and S3. An amplitude histogram of this trace is shown on the right side of record. Single channels in excised membrane patches are less stable than those of cell-attached patches. The basis for this difference is unknown. (B) A cell-attached patch recording of a single Cx32Cx43E1 conductive hemichannel obtained at a holding potential of −70 mV. The channel undergoes a series of gating transitions () that may result in full closure, termed loop gating. (C) A cell-attached patch recording of a single Cx3243E1 conductive hemichannel illustrating gating transitions to a subconductance level (arrows) at a depolarizing membrane potential of +80 mV (the section of the record between numerals 1 and 2). At positions 1 and 2, the membrane potential was reversed from −80 to + 80 mV and from +80 to −80 mV, respectively. Vj gating events are evident in the portions of the current trace elicited by the hyperpolarization of the membrane to −80 mV. Leak currents were not subtracted.
	Figure 6. Current–voltage relations of Cx32Cx43E1 conductive hemichannels obtained by the application of ±100-mV ramps in a cell-attached patch configuration. Arrows indicate the direction of the time axis. In A and C, voltage was ramped from −100 to +100 mV, in B and D from +100 to −100 mV. In the 3-s-duration voltage ramps shown in A and B, current rectifies inwardly and the channel closes to subconductance levels only at hyperpolarizing potentials. The overall form of the I-V relation elicited by the positive going ramp shown in A is similar to that elicited by the negative going ramp shown in B. (C and D) In 8-s voltage ramps, an additional gating transition () is observed at depolarizing potentials (see Fig. 5 C). The baseline currents were adjusted to zero in these records. The records were digitally filtered at 200 Hz for presentation.
	Figure 7. Single-channel recordings from Xenopus oocytes expressing Cx32N2ECx43E1 conductive hemichannels. (A) Cell-attached recording of a single Cx32N2ECx43E1 conductive hemichannel shows that the channel remains in a fully open state (−50 mV) and gates only upon depolarization (+50 mV). Leak currents were not subtracted. (B) The steepness of the voltage dependence of Cx32N2ECx43E1 conductive hemichannels is illustrated. The open probability of the channel decreases considerably over the 20-mV increment (+20 to +40 mV) shown (left). (Right) The open probability is intermediate at +30 mV. (C) A portion of a cell-attached record showing a loop gating event at a membrane potential of −60 mV. (D) The current–voltage relation of a single Cx32N2ECx43E1 conductive hemichannel obtained with 3-s voltage ramps from −100 to +100 mV in a cell-attached patch recording configuration. Four sequential current traces are superimposed. Vj-gating transitions are observed only at depolarizing potentials. Currents passing through the fully open state rectify outward slightly in contrast to the inward rectifying currents observed for Cx32Cx43E1 conductive hemichannels. The leak current of the patch. Baseline currents were adjusted to 0 pA. Current traces were digitally filtered at 200 Hz for presentation.
	Figure 8. Vj gating of a heteromeric conductive hemichannels containing both Cx32Cx43E1 and Cx32N2ECx43E1 subunits. (A) The cell-attached records illustrate that Vj gating is bipolar. The heteromeric hemichannel containing a single Cx32Cx43E1 subunit makes transitions between fully open and subconductance levels at both positive and negative potentials (+50 and −80 mV, top). The heteromeric hemichannel containing a single Cx32N2ECx43E1 subunit makes transitions between fully open and subconductance levels at both positive and negative holding potentials (−120 and +80 mV, middle). (Bottom) The Vj-gating transitions to the two subconductance levels (S1 and S2) at depolarizing potential (+80 mV) are attributable to the presence of a single Cx32N2ECx43E1 subunit. The smaller amplitude transition (, bottom) is attributable to a voltage-dependent process also observed in homomeric Cx32Cx43E1 conductive hemichannels at depolarizing membrane potentials exceeding +70 mV (Fig. 5 C). Leak currents were not subtracted. (B) The open probability–voltage relation of homomeric Cx32N2ECx43E1 conductive hemichannel (•) and heteromeric conductive hemichannel containing a single Cx32N2ECx43E1 subunit (▴) are shown. The open probabilities of each channel are plotted as a function of holding potential. Open probabilities were calculated from the ratio of the areas under the peaks of all-point histograms. The numbers in parentheses are the numbers of different records that were concatenated and used to generate the all-point histograms at each voltage. The thin lines are fits of the data points to a Boltzmann relation with Microcal Origin 4.0 software. (C) Current–voltage relation of single heteromeric conductive hemichannel (containing one Cx32N2ECx43E1 and five Cx32*Cx43E1 subunits) obtained in a cell-attached configuration with 3-s voltage ramps from −150 to +150 mV. Four current traces from the same single channel record are superimposed. The heteromeric hemichannel gates between fully open and subconductance levels at both positive and negative potentials are shown. The dotted line is an exponential fit to the subconductance levels at negative potentials. The failure of the fitted curve to intersect the I-V relation of the substates at positive potentials illustrates that the channel substates observed at negative potentials do not correspond to the substates observed at positive potentials. The baseline current was adjusted to 0 pA and current traces were digitally filtered at 200 Hz for presentation.
	Figure 9. Cell-attached patch recordings of a nonconnexin channel endogenous to Xenopus oocytes. (A) A record of a single channel at a holding potential of +30 and −50 mV. The channel has a unitary conductance of 20–25 pS at +30 mV and 30 pS at −50 mV. (B) Currents elicited by alternating ±90-mV voltage steps. The channels are slowly activated by depolarization (+90 mV). At hyperpolarizing membrane potentials (−90 mV), the activated channels close and appear to run down. (C) The current–voltage relation of the endogenous Xenopus oocyte channel resulting from a 3-s voltage ramp from +100 to −100 mV. Baseline currents were not adjusted.
	Figure 10. A comparison of concerted (A) and individual (B) subunit gating models for a heteromeric hemichannel containing a single N2E connexin subunit. In the concerted model shown in A, the movement of a voltage sensor in one subunit causes a similar conformational change in all six connexin subunits regardless of the orientation of their voltage sensors in the electric field. Substates arise by a concerted change in the conformation of all six subunits. The state depicted Closed “1st” (+/−) indicates that the channel enters the same substate in response to either membrane depolarization (+) or hyperpolarization (−). In the individual model shown in B, each subunit can respond autonomously to changes in the polarity of the electric field. Substates can result from changes in the conformation of one subunit, as illustrated (B, top) (in response to membrane depolarization, +) or by the changes in the conformation of multiple subunits as illustrated (bottom) (in response to membrane hyperpolarization, −). See text for details. Gray circle, positively charged subunit in an open conformation; solid circle, negatively charged subunit in an open conformation; gray triangle, positively charged subunit in a closed conformation; solid triangle, negatively charged subunit in a closed conformation; solid square, negatively charged subunit in another closed conformation.

References [+] :

Barrio, Gap junctions formed by connexins 26 and 32 alone and in combination are differently affected by applied voltage. 1991, Pubmed, Xenbase

	Figure 1. Representative conductance–voltage relations and macroscopic current traces of intercellular channels formed by the designated wild-type and chimeric connexins expressed in pairs of Xenopus oocytes. In all cases, junctional conductance is normalized to Vj = 0. Positive Vj is relative to the cytoplasmic entrance of the hemichannel appearing on the right side of the channel designation. Initial conductance is depicted by ▾, steady state conductance by ▿.
	Figure 2. Macroscopic recordings of Xenopus oocytes expressing Cx32Cx43E1 and Cx32N2ECx43E1 conductive hemichannels. (A) Outward currents in oocytes expressing Cx32Cx43E1 were elicited from a holding potential at −90 mV, with 30-s voltage steps ranging from +20 to +60 mV in 10-mV increments. Inward tail currents resulted from the repolarization of the membrane to −90 mV. In 5 mM Ca2+-containing bath solution, tail currents show an initial increase followed by a slow return to base line with a multiexponential time course. (B) The time course of the tail current relaxations is prolonged by reductions in the Ca2+ concentration of the bath solution from 5 to 2 mM. (C) Tail currents elicited by repolarization of the membrane to −90 mV in bath solutions containing 2 mM Ba2+ and no added Ca2+. The replacement of Ca2+ with Ba2+ alters the kinetics of tail current relaxations and removes initial rise in tail currents. It is likely that the ionic substitution reduces the contribution of Ca2+-activated Cl− currents. (D) Outward currents in Cx32N2ECx43E1 oocytes were elicited from a holding potential at −90 mV, with 30-s voltage steps ranging from 0 to +40 mV in 10-mV increments. Inward tail currents resulted from the repolarization of the membrane to −90 mV. The bath solutions contained 5 mM Ca2+. At voltages exceeding +30 mV, currents plateau consistent with the reversal of gating polarity by the N2E substitution. The time constants of tail-current relaxations are substantially faster than those observed in Cx32*Cx43E1 conductive hemichannels. (Inset) Current traces elicited by positive voltage steps are expanded.
	Figure 3. Voltage dependence of rapid gating transitions in single Cx32Cx43E1 conductive hemichannels. (A) Cell-attached patch recordings of a single Cx32Cx43E1 conductive hemichannel at different voltages, showing that the frequency of rapid transitions increases at larger hyperpolarizing potentials. The holding potential was flipped at the dotted vertical line from the hyperpolarizing to the depolarizing potential denoted. Leak currents were not subtracted. (B) An event list histogram of closed times at −70 mV. The histogram is best fit by a single exponential with a time constant, τ = 1.5 ms.
	Figure 4. Voltage-dependent gating of a Cx32Cx43E1 conductive hemichannel. (A) A cell-attached recording of a single Cx32Cx43E1 conductive hemichannel at −80 mV. In addition to the rapid flickers (Fig. 3), the channel resides at longer-lived subconductance levels, with durations that range from tens of milliseconds (shown on expanded trace) to tens of seconds. (B) An event list histogram of closures with duration longer than 50 ms at −80 mV. The histogram is best fit by a double exponential with time constants, τ1 = 56.8 ms and τ2 = 593 ms. (Inset) The portion of the histogram from 0 to 2 s is shown.
	Figure 5. Single-channel recordings of Cx32Cx43E1 conductive hemichannel illustrating three different gating mechanisms. (A) Vj gating is illustrated by the outside-out patch recording at −100 mV. The channel can enter one of three subconductance states denoted as S1, S2, and S3. An amplitude histogram of this trace is shown on the right side of record. Single channels in excised membrane patches are less stable than those of cell-attached patches. The basis for this difference is unknown. (B) A cell-attached patch recording of a single Cx32Cx43E1 conductive hemichannel obtained at a holding potential of −70 mV. The channel undergoes a series of gating transitions () that may result in full closure, termed loop gating. (C) A cell-attached patch recording of a single Cx3243E1 conductive hemichannel illustrating gating transitions to a subconductance level (arrows) at a depolarizing membrane potential of +80 mV (the section of the record between numerals 1 and 2). At positions 1 and 2, the membrane potential was reversed from −80 to + 80 mV and from +80 to −80 mV, respectively. Vj gating events are evident in the portions of the current trace elicited by the hyperpolarization of the membrane to −80 mV. Leak currents were not subtracted.
	Figure 6. Current–voltage relations of Cx32Cx43E1 conductive hemichannels obtained by the application of ±100-mV ramps in a cell-attached patch configuration. Arrows indicate the direction of the time axis. In A and C, voltage was ramped from −100 to +100 mV, in B and D from +100 to −100 mV. In the 3-s-duration voltage ramps shown in A and B, current rectifies inwardly and the channel closes to subconductance levels only at hyperpolarizing potentials. The overall form of the I-V relation elicited by the positive going ramp shown in A is similar to that elicited by the negative going ramp shown in B. (C and D) In 8-s voltage ramps, an additional gating transition () is observed at depolarizing potentials (see Fig. 5 C). The baseline currents were adjusted to zero in these records. The records were digitally filtered at 200 Hz for presentation.
	Figure 7. Single-channel recordings from Xenopus oocytes expressing Cx32N2ECx43E1 conductive hemichannels. (A) Cell-attached recording of a single Cx32N2ECx43E1 conductive hemichannel shows that the channel remains in a fully open state (−50 mV) and gates only upon depolarization (+50 mV). Leak currents were not subtracted. (B) The steepness of the voltage dependence of Cx32N2ECx43E1 conductive hemichannels is illustrated. The open probability of the channel decreases considerably over the 20-mV increment (+20 to +40 mV) shown (left). (Right) The open probability is intermediate at +30 mV. (C) A portion of a cell-attached record showing a loop gating event at a membrane potential of −60 mV. (D) The current–voltage relation of a single Cx32N2ECx43E1 conductive hemichannel obtained with 3-s voltage ramps from −100 to +100 mV in a cell-attached patch recording configuration. Four sequential current traces are superimposed. Vj-gating transitions are observed only at depolarizing potentials. Currents passing through the fully open state rectify outward slightly in contrast to the inward rectifying currents observed for Cx32Cx43E1 conductive hemichannels. The leak current of the patch. Baseline currents were adjusted to 0 pA. Current traces were digitally filtered at 200 Hz for presentation.
	Figure 8. Vj gating of a heteromeric conductive hemichannels containing both Cx32Cx43E1 and Cx32N2ECx43E1 subunits. (A) The cell-attached records illustrate that Vj gating is bipolar. The heteromeric hemichannel containing a single Cx32Cx43E1 subunit makes transitions between fully open and subconductance levels at both positive and negative potentials (+50 and −80 mV, top). The heteromeric hemichannel containing a single Cx32N2ECx43E1 subunit makes transitions between fully open and subconductance levels at both positive and negative holding potentials (−120 and +80 mV, middle). (Bottom) The Vj-gating transitions to the two subconductance levels (S1 and S2) at depolarizing potential (+80 mV) are attributable to the presence of a single Cx32N2ECx43E1 subunit. The smaller amplitude transition (, bottom) is attributable to a voltage-dependent process also observed in homomeric Cx32Cx43E1 conductive hemichannels at depolarizing membrane potentials exceeding +70 mV (Fig. 5 C). Leak currents were not subtracted. (B) The open probability–voltage relation of homomeric Cx32N2ECx43E1 conductive hemichannel (•) and heteromeric conductive hemichannel containing a single Cx32N2ECx43E1 subunit (▴) are shown. The open probabilities of each channel are plotted as a function of holding potential. Open probabilities were calculated from the ratio of the areas under the peaks of all-point histograms. The numbers in parentheses are the numbers of different records that were concatenated and used to generate the all-point histograms at each voltage. The thin lines are fits of the data points to a Boltzmann relation with Microcal Origin 4.0 software. (C) Current–voltage relation of single heteromeric conductive hemichannel (containing one Cx32N2ECx43E1 and five Cx32*Cx43E1 subunits) obtained in a cell-attached configuration with 3-s voltage ramps from −150 to +150 mV. Four current traces from the same single channel record are superimposed. The heteromeric hemichannel gates between fully open and subconductance levels at both positive and negative potentials are shown. The dotted line is an exponential fit to the subconductance levels at negative potentials. The failure of the fitted curve to intersect the I-V relation of the substates at positive potentials illustrates that the channel substates observed at negative potentials do not correspond to the substates observed at positive potentials. The baseline current was adjusted to 0 pA and current traces were digitally filtered at 200 Hz for presentation.
	Figure 9. Cell-attached patch recordings of a nonconnexin channel endogenous to Xenopus oocytes. (A) A record of a single channel at a holding potential of +30 and −50 mV. The channel has a unitary conductance of 20–25 pS at +30 mV and 30 pS at −50 mV. (B) Currents elicited by alternating ±90-mV voltage steps. The channels are slowly activated by depolarization (+90 mV). At hyperpolarizing membrane potentials (−90 mV), the activated channels close and appear to run down. (C) The current–voltage relation of the endogenous Xenopus oocyte channel resulting from a 3-s voltage ramp from +100 to −100 mV. Baseline currents were not adjusted.
	Figure 10. A comparison of concerted (A) and individual (B) subunit gating models for a heteromeric hemichannel containing a single N2E connexin subunit. In the concerted model shown in A, the movement of a voltage sensor in one subunit causes a similar conformational change in all six connexin subunits regardless of the orientation of their voltage sensors in the electric field. Substates arise by a concerted change in the conformation of all six subunits. The state depicted Closed “1st” (+/−) indicates that the channel enters the same substate in response to either membrane depolarization (+) or hyperpolarization (−). In the individual model shown in B, each subunit can respond autonomously to changes in the polarity of the electric field. Substates can result from changes in the conformation of one subunit, as illustrated (B, top) (in response to membrane depolarization, +) or by the changes in the conformation of multiple subunits as illustrated (bottom) (in response to membrane hyperpolarization, −). See text for details. Gray circle, positively charged subunit in an open conformation; solid circle, negatively charged subunit in an open conformation; gray triangle, positively charged subunit in a closed conformation; solid triangle, negatively charged subunit in a closed conformation; solid square, negatively charged subunit in another closed conformation.