Voltage-dependent structural interactions in the Shaker K(+) channel - PubMed (original) (raw)

Voltage-dependent structural interactions in the Shaker K(+) channel

S K Tiwari-Woodruff et al. J Gen Physiol. 2000 Feb.

Abstract

Using a strategy related to intragenic suppression, we previously obtained evidence for structural interactions in the voltage sensor of Shaker K(+) channels between residues E283 in S2 and R368 and R371 in S4 (Tiwari-Woodruff, S.K., C.T. Schulteis, A.F. Mock, and D. M. Papazian. 1997. Biophys. J. 72:1489-1500). Because R368 and R371 are involved in the conformational changes that accompany voltage-dependent activation, we tested the hypothesis that these S4 residues interact with E283 in S2 in a subset of the conformational states that make up the activation pathway in Shaker channels. First, the location of residue 283 at hyperpolarized and depolarized potentials was inferred by substituting a cysteine at that position and determining its reactivity with hydrophilic, sulfhydryl-specific probes. The results indicate that position 283 reacts with extracellularly applied sulfhydryl reagents with similar rates at both hyperpolarized and depolarized potentials. We conclude that E283 is located near the extracellular surface of the protein in both resting and activated conformations. Second, we studied the functional phenotypes of double charge reversal mutations between positions 283 and 368 and between 283 and 371 to gain insight into the conformations in which these positions approach each other most closely. We found that combining charge reversal mutations at positions 283 and 371 stabilized an activated conformation of the channel, and dramatically slowed transitions into and out of this state. In contrast, charge reversal mutations at positions 283 and 368 stabilized a closed conformation, which by virtue of the inferred position of 368 corresponds to a partially activated (intermediate) closed conformation. From these results, we propose a preliminary model for the rearrangement of structural interactions of the voltage sensor during activation of Shaker K(+) channels.

PubMed Disclaimer

Figures

Figure 2

Modification of C283 by extracellular MTS reagents. (A) Currents were recorded from _Shaker_-IR, R371Q-IR, and E283C+R371Q-IR channels before (#) and after (*) treatment with 5 mM MTSET applied in the 50-mM RbCl bath solution during continuous perfusion. From a holding potential of −100 mV, the membrane was pulsed to +40 mV. (B) Currents were recorded from _Shaker_-IR, R371Q-IR, and E283C+R371Q-IR channels before (#) and after (*) treatment with 5 mM MTSES applied in the 50-mM RbCl bath solution during continuous perfusion. Pulse protocol was the same as in A.

Figure 1

Functional expression of E283C+R371Q-IR channels in Xenopus oocytes. (A) _Shaker_-IR, R371Q-IR, and E283C+R371Q-IR were expressed individually in Xenopus oocytes. 24 h after injection of RNA, currents were recorded using a two-electrode voltage clamp. From a holding potential of −100 mV, the membrane was pulsed to voltages between −100 and +80 mV in 20-mV increments. Tail currents were evoked by return to the holding potential, −100 mV. The bath contained 50 mM RbCl in a modified Barth's solution. (B) P o-V curves for _Shaker_-IR (•), R371Q-IR (▴), and E283C+ R371Q-IR (▪) channels were derived from isochronal measurements of tail current amplitude. Tail current amplitudes at the indicated potentials were normalized to the maximal amplitude obtained in the experiment. To derive midpoint voltages (Vmid) and slope factors (a), the P o-V curves were fitted with a Boltzmann function: IR, Vmid = −23.3 ± 4.8, a = 9.9 ± 2.5 (n = 8); R371Q-IR, Vmid = −45.4 ± 3.0, a = 5.0 ± 1.0 (n = 7); and E283C+R731Q-IR, Vmid = −13.6 ± 3.3, a = 8.9 ± 1.8 (n = 14). Values are shown in millivolt units as mean ± SD.

Figure 3

Rate of modification of deactivation kinetics in E283C+R371Q-IR channels treated with MTS reagents. (A) The pulse protocol for the experiment is shown. At 5.6-s intervals, the membrane was pulsed from the holding potential of −100 to +40 mV for 200 ms. The first current trace was recorded in the control bath solution containing 50 mM RbCl, whereas for subsequent pulses, the bath solution was supplemented with 5 mM MTSET or 1 mM MTSES (arrow). Oocytes were subjected to continuous perfusion with the indicated solutions. (B) Representative tail currents recorded from single cells after every fifth pulse to +40 mV are shown superimposed. Currents recorded in control solution after the first pulse (#) and in MTS-containing solution after the last pulse (*) are labeled as indicated. Tail currents were fitted with single exponential functions, shown as solid curves superimposed on current traces, to derive deactivation time constants (τdeact.). (C) Fitted τdeact. values were plotted as a function of perfusion time. A time constant for modification was derived by fitting a single exponential function to the data (solid curve). For the representative experiment shown, τ for 5 mM MTSET was 20.0 s and τ for 1 mM MTSES was 52.0 s. (D) Apparent second-order rate constants, calculated as 1/(time constant of modification × concentration), are shown in a box plot. Means (M−1 s−1) ± SD were 11.8 ± 2.0 (n = 14), and 32.2 ± 15.5 (n = 8) for MTSET and MTSES, respectively. The box plot depicts the statistical distribution of the data: × indicates the 95th (top) and 5th (bottom) percentile points, error bars represent the 90th (top) and 10th (bottom) percentiles, the upper and lower margins of the box correspond to the 75th and 25th percentiles, solid lines within the box mark the median, while □ indicates the mean.

Figure 4

Onset and reversal of Cd2+ acceleration of E283C+R371Q-IR deactivation kinetics. (A) Using the pulse protocol shown in Fig. 3 A, the first pulse was recorded in control bath solution, whereas subsequent pulses were recorded in bath solution supplemented with 300 μM CdCl2. Representative tail currents recorded from a single cell after successive pulses are shown superimposed. Currents recorded in control solution after the first pulse (#) and in Cd2+-containing solution after the last pulse (*) are labeled as indicated. Tail currents were fitted with single exponential functions, shown as solid curves superimposed on current traces. (B) The pulse protocol was repeated in the same cell after saturation of the Cd2+ effect. The first pulse was recorded in the Cd2+-containing solution, and subsequent pulses were recorded in control bath solution supplemented with 1 mM EGTA. Representative tail currents recorded from the same cell as in A are shown. Currents recorded in Cd2+-containing solution after the first pulse (*) and in EGTA-containing solution after the last pulse (+) are labeled as indicated. Tail currents were fitted with single exponential functions, shown as solid curves superimposed on current traces. (C, left) Deactivation time constants (τdeact.) obtained from fits shown in A and B were plotted as a function of perfusion time. Values of τdeact. obtained in control (▴), Cd2+-containing (○), and EGTA-containing (□) bath solutions are shown. Time constants for the onset (7.5 s) and reversal (45.3 s) of the Cd2+ effect were derived as described in Fig. 3 C. (Right) Box plots show apparent second-order rate constant for onset of the Cd2+ effect, calculated as 1/(time constant of modification × concentration), and apparent first-order rate constant for reversal in the presence of EGTA, calculated as 1/time constant of modification. Means ± SD for modification by Cd2+ and reversal in the presence of EGTA were 430 ± 97 M−1 s−1 (n = 8), and 0.018 ± 0.003 s−1 (n = 8), respectively. (D) P o-V curves for E283C+R371Q-IR channels were derived from isochronal measurements of tail current amplitude before (▴) and after (○) treatment with Cd2+, and after reversal with EGTA (□). (E) Currents were recorded from _Shaker_-IR channels in the presence (*) and absence (#) of 300 μM CdCl2 by pulsing from a holding potential of −100 to +40 mV. The bath solution was modified Barth's saline containing 50 mM KCl [50 mM KCl, 39 mM NaCl, 2.4 mM NaHCO3, 0.82 mM MgSO4, 0.33 mM Ca (NO3)2, 0.41 mM CaCl2, 10 mM HEPES, pH 7.5].

Figure 5

Modification of C283 by extracellular MTS reagents applied at −100 or +40 mV. (A) The pulse protocols for the experiments are shown. While clamping the membrane at either −100 mV (“resting” protocol) or +40 mV (“activated” protocol), 5 mM MTSET or 1 mM MTSES (MTS) was applied for 2 min, followed by 2 mM DTT (DTT) for 30 s. MTS reagents or DTT were added to the control (50 mM RbCl) bath solution and applied by continuous perfusion. Before (1) and after (2) treatment, current traces were recorded while perfusing with the control solution by pulsing from the holding potential of −100 to +40 mV for 200 ms. (B) Representative tail currents obtained before (1) and after (2) treatment with 5 mM MTSET at −100 mV (left) or +40 mV (right) are shown. Traces were fitted with a single exponential function, which is shown (solid curve) superimposed on the tail currents. (C) Box plot of deactivation time constants (τdeact.) derived from single component fits is shown for E283C+R371Q-IR channels before and after treatment with MTS reagents at −100 or +40 mV, as indicated. Mean (ms) ± SD were: untreated control, 90.9 ± 18.5 (n = 15); MTSET at −100 mV, 35.2 ± 6.8 (n = 13); MTSET at +40 mV, 43.6 ± 9.7 (n = 12); MTSES at −100 mV, 23.4 ± 5.4 (n = 6); and MTSES at +40 mV, 34.7 ± 9.1 (n = 7). Time constants obtained after treatment at −100 or +40 mV do not differ significantly, as assessed by a one way ANOVA (P > 0.01).

Figure 6

Rate of modification of deactivation kinetics in E283C+R371Q-IR channels treated with MTS reagents at −100 or +40 mV. The pulse protocol was the same as in Fig. 5 A, except that the time of application of the MTS reagents was varied between 5 and 60 s. Box plot of apparent second-order rate constants for modification by MTS reagents at either −100 or +40 mV is shown. Mean (M−1 s−1) ± SD were: MTSET at −100 mV, 17.9 ± 6.7 (n = 5); MTSET at +40 mV, 5.2 ± 2.1 (n = 5); MTSES at −100 mV, 70.2 ± 23.9 (n = 5); MTSES at +40 mV, 19.9 ± 5.6 (n = 10). Difference in reaction rates with both MTS reagents at −100 and +40 mV are statistically significant, as assessed by a one way ANOVA (P < 0.01).

Figure 8

Activation at hyperpolarized potentials in E283R+R371E-IR channels. (A) E283R+R371E-IR channels were expressed in oocytes, and currents were recorded without leak subtraction. The membrane potential was stepped from the holding potential of −100 mV to voltages between −100 and +80 mV in +20-mV increments. The oocyte was continuously perfused with bath solution containing 1 mM KCl in a modified Barth's saline. (B) Unsubtracted currents were recorded from an uninjected oocyte of the same batch using the same protocol as in A. (C) Currents were recorded from E283E+R371E-IR channels in the presence of 118 mM TEA using the same protocol as in A. These records were obtained from the same oocyte used in A. This concentration of TEA did not completely block the channels; in some oocytes, a larger fraction of the current was blocked (data not shown). (D) Unsubtracted currents were recorded from an uninjected oocyte of the same batch in the presence of 118 mM TEA using the same protocol as in A. These records were obtained from the same oocyte used in B.

Figure 7

Onset and reversal of Cd2+ effect at −100 and +40 mV. (A) The pulse protocols for the experiments are shown. While clamping the membrane at either −100 mV (“resting” protocol) or +40 mV (“activated” protocol), 300 μM CdCl2 (indicated by Cd2+) or 1 mM EGTA (indicated by EGTA) was applied for 2 min. Currents were recorded in the control bath solution (50 mM RbCl lacking EGTA) during 200-ms test pulses from −100 to +40 mV before (1), between (2), and after (3) Cd2+ and EGTA treatment. (B) Box plot of deactivation time constants obtained after application of Cd2+ and EGTA at −100 or +40 mV is shown. Time constants (τdeact.), obtained by fitting a single exponential function to tail current traces, were: untreated control, 81.4 ± 16.9 (n = 7); Cd2+ at −100 mV, 26.9 ± 4.8 (n = 7); EGTA at −100 mV, 81.5 ± 36.8 (n = 4); Cd2+ at +40 mV, 30.5 ± 11.9 (n = 6); and EGTA at +40 mV, 85.0 ± 32.9 (n = 4). Values are given as mean (ms) ± SD. Deactivation time constants for Cd2+ applied at −100 and +40 mV were not statistically different (one way ANOVA, P > 0.05). EGTA completely reversed the Cd2+ effect at −100 and +40 mV (P > 0.05, compared with untreated control).

Figure 9

Current kinetics depend on pulse protocol in E283R+R371E-IR channels. (A) E283R+R371E-IR channels were expressed in oocytes, and currents were recorded without leak subtraction. The membrane potential was stepped from the holding potential of −100 mV to voltages between +80 and −100 mV in −20-mV increments. These records were obtained from the same oocyte used in Fig. 8 A, but the voltage pulses were applied in the opposite order compared with Fig. 8. The oocyte was continuously perfused with bath solution containing 1 mM KCl. (B) Unsubtracted currents were recorded from an uninjected oocyte of the same batch using the same protocol as in A. These records were obtained from the same oocyte used in Fig. 8 B. (C) Currents were recorded from E283E+R371E-IR channels in the presence of 118 mM TEA chloride using the same protocol as in A. These records were obtained from the same oocyte used in A and Fig. 8 A. (D) Unsubtracted currents were recorded from an uninjected oocyte of the same batch in the presence of 118 mM TEA using the same protocol as in A. These records were obtained from the same oocyte used in B and Fig. 8 B.

Figure 10

Slow deactivation kinetics do not account for activation at hyperpolarized potentials in E283R+R371E-IR channels. (A) The pulse protocol for the experiment is shown. Three test pulses from the holding potential of −100 to +60 mV were applied. The interepisode interval (Δt) at −100 mV was varied. The oocyte was continuously perfused with bath solution containing 1 mM KCl. (B) Δt = 5 s. (C) Δt = 15 s. (D) Δt = 30 s. (E) Δt = 45 s. (F) Δt = 60 s. Capacitative transients have been truncated in B–F.

Figure 11

Activation shifted to depolarized potentials by the presence of E283R+R368E subunits in heterotetrameric channels. (A) RNA encoding _Shaker_-IR and inactivation-tagged E283R+R386E subunits was coinjected into oocytes at ratios of 1:10 (top) or 1:1 (middle), or _Shaker_-IR RNA was injected alone (bottom). Currents were elicited by pulsing from a holding potential of −100 to +70 mV. A constant amount of _Shaker_-IR RNA was injected in each case. Note the change in current amplitude as a function of injection ratio. (B) G-V curves are shown for _Shaker_-IR (▵) and Shaker wild-type (•) channels and for the steady state (□, noninactivating) and peak (▪, ▴, inactivating) components of currents obtained after coexpression of _Shaker_-IR and E283R+R368E subunits at RNA ratios of 1:1 (▴, □) and 1:10 (▪). Similar results were obtained for an injection ratio of 1:20 (data not shown). Currents were evoked by 100-ms pulses from the holding potential of −100 mV to potentials ranging from −90 to +90 mV in 20-mV increments. Conductances were calculated using the chord conductance equation from inactivating (peak minus steady state, filled symbols) or noninactivating (steady state, open symbols) current amplitudes, normalized to the maximum conductance obtained in the experiment, and plotted as a function of pulse potential. Values are shown as mean ± SEM obtained from three experiments with four to eight oocytes tested per experiment. Midpoint voltages in units of millivolts ± SD, obtained by fitting the conductance data with a single Boltzmann function, were: _Shaker_-IR, −19 ± 2; Shaker wild type, −10 ± 2; inactivating component of _Shaker_-IR and E283R+R368E coinjected at a 1:1 ratio, 30 ± 2; inactivating component of _Shaker_-IR and E283R+R368E coinjected at a 1:10 ratio, 28 ± 3; noninactivating component of _Shaker_-IR and E283R+R368E coinjected at a 1:1 ratio, −21 ± 4. (C) E283R+R368E subunits exert a dominant negative effect on the current amplitude. RNA encoding _Shaker_-IR and E283R+ R368E subunits was injected at molar ratios of 1:1, 1:10, or 1:20, or _Shaker_-IR RNA was injected alone. A constant amount of _Shaker_-IR RNA was injected in each case. Currents were evoked by pulsing from −100 to +70 mV. Maximum (peak) current amplitudes are shown in a histogram as mean ± SD, n = 5–15 for each injection ratio. Statistical significance was evaluated using Student's t test (***P < 0.001).

Figure 12

Tentative model of structural rearrangements that accompany voltage-dependent activation in Shaker K+ channels. In the resting state, E283 is near the extracellular surface of the protein (this study), whereas R368 and R371 are near the intracellular surface (Larsson et al. 1996; Starace et al. 1997). In an intermediate closed conformation, R368 is in proximity with E283. R371 is shown in parentheses because our data do not suggest its position in this conformation. In the activated conformation, E283 and R371 are in proximity.

Cited by

Fifty years of gating currents and channel gating.
Catacuzzeno L, Conti F, Franciolini F. Catacuzzeno L, et al. J Gen Physiol. 2023 Aug 7;155(8):e202313380. doi: 10.1085/jgp.202313380. Epub 2023 Jul 6. J Gen Physiol. 2023. PMID: 37410612 Free PMC article. Review.
Roles for Countercharge in the Voltage Sensor Domain of Ion Channels.
Groome JR, Bayless-Edwards L. Groome JR, et al. Front Pharmacol. 2020 Feb 28;11:160. doi: 10.3389/fphar.2020.00160. eCollection 2020. Front Pharmacol. 2020. PMID: 32180723 Free PMC article. Review.
Voltage-dependent gating in K channels: experimental results and quantitative models.
Catacuzzeno L, Sforna L, Franciolini F. Catacuzzeno L, et al. Pflugers Arch. 2020 Jan;472(1):27-47. doi: 10.1007/s00424-019-02336-6. Epub 2019 Dec 20. Pflugers Arch. 2020. PMID: 31863286 Review.
Voltage vs. Ligand II: Structural insights of the intrinsic flexibility in cyclic nucleotide-gated channels.
Romero-Romero S, Martínez-Delgado G, Balleza D. Romero-Romero S, et al. Channels (Austin). 2019 Dec;13(1):382-399. doi: 10.1080/19336950.2019.1666456. Channels (Austin). 2019. PMID: 31552786 Free PMC article.
Analogues of ERβ ligand chloroindazole exert immunomodulatory and remyelinating effects in a mouse model of multiple sclerosis.
Karim H, Kim SH, Lauderdale K, Lapato AS, Atkinson K, Yasui N, Yamate-Morgan H, Sekyi M, Katzenellenbogen JA, Tiwari-Woodruff SK. Karim H, et al. Sci Rep. 2019 Jan 24;9(1):503. doi: 10.1038/s41598-018-37420-x. Sci Rep. 2019. PMID: 30679747 Free PMC article.

References

1. Aggarwal S.K., MacKinnon R. Contribution of the S4 segment to gating charge in the Shaker K+ channel. Neuron. 1996;16:1169–1177. - PubMed
1. Baker O.S., Larsson H.P., Mannuzzu L.M., Isacoff E.Y. Three transmembrane conformations and sequence-dependent displacement of the S4 domain in Shaker K+ channel gating. Neuron. 1998;20:1283–1294. - PubMed
1. Bezanilla F., Armstrong C.M. Inactivation of the sodium channel. I. Sodium current experiments. J. Gen. Physiol. 1977;70:549–566. - PMC - PubMed
1. Bezanilla F., Perozo E., Stefani E. Gating of Shaker K+ channelsII. The components of gating currents and a model of channel activation. Biophys. J. 1994;66:1011–1021. - PMC - PubMed
1. Cha A., Bezanilla F. Characterizing voltage-dependent conformational changes in the Shaker K+ channel with fluorescence. Neuron. 1997;19:1127–1140. - PubMed

Voltage-dependent structural interactions in the Shaker K(+) channel - PubMed (original) (raw)