|
Figure 1. Titrateable voltage sensor models. (A) Gating currents were modeled as transitions between H and D connected by voltage-dependent rate constants α and β. The charge carried by the titrateable voltage sensor in each transition between H and D is either z or z + z1, as determined by the equilibrium dissociation constant, pK (Ki or Ko), of the titrateable group and the surrounding pH (Hi+ or Ho+ if the titrateable group gets exposed to the internal or external solution, respectively). Since the S4 segment, which contains most of the voltage sensing residues, is positively charged, depolarization of the membrane (gray box) moves the voltage sensor from the inside of the membrane towards the outside, thereby moving charge z or z + z1, in the transition. (B) The titrateable voltage sensor model shown in A was extended to incorporate the formation of a one-ion proton pore at depolarized potentials. In the D states, the titrateable residue has access to both the internal and external solutions simultaneously, and this creates a proton pathway through the membrane. Proton binding in the D states from the external solution depends on the dissociation constant, pKo (Ko = γo/ɛo) and the external pH (Ho); binding from the internal solution depends on pKiP (KiP = γip/ɛip) and pHi.
|
|
Figure 2. Histidine scanning mutagenesis. The various experimental outcomes of histidine scanning mutagenesis (summarized in text on the right) depend on the accessibility of the histidine (H, unprotonated; H+, protonated) to solution protons during gating. Histidine accessibility during gating is monitored as pH-dependent changes in the gating charge displacement (Q) evoked by membrane potential pulses (shown at top). A depolarizing pulse (ΔV) moves the positively charged voltage sensor (represented as a cylinder) from the inside of the membrane (gray box) towards the outside, thereby displacing gating charge Qon. Repolarization (−ΔV) returns the gating charge Qoff. If the histidine is part of the voltage sensor, modulation of Qon by the internal pH (pHi) indicates internal exposure and modulation of Qoff by the external pH (pHo) indicates external exposure. The relationship between Qon and Qoff is shown schematically in pHo/pHi 9.2/9.2 (left) and in pHo/pHi 5/9.2 (right) for four different histidine exposure possibilities.
|
|
Figure 3. Gating charge displacement in the R377H channel is unaffected by internal protons. (A) Gating current records from the R377H channel measured from an inside-out macropatch in symmetric NMDG-MS pH 7.4 solutions. The superimposed currents are in response to various test pulses from a holding potential of −90 mV, also shown superimposed at the top. Onset of the test pulse stimulates an ON-gating current (Qon), and repolarization of the membrane causes the returning OFF-gating current (Qoff). (B) The same sequence of superimposed gating current records from the same macropatch after exchange of the pH 7.4 internal solution (pHi 7.4) with a pHi 5 internal solution. (C) ON- and OFF- gating currents were each integrated over time and plotted as a function of pulse potential to obtain the voltage dependence of steady-state charge displacement (Q-V curves). Q-V curves are shown for the ON- (left panel, open symbols) and OFF-gating currents (right panel, closed symbols) displayed in A (pHi 7.4, triangles) and B (pHi 5, squares). The ON and OFF Q-V curves for the gating currents measured in symmetric pH 7.4 solutions are plotted on the same graph in the inset of the right panel. Each Q-V curve was fit to a sum of two Boltzmann distributions (lines). (Experiment D03238a)
|
|
Figure 4. Gating charge displacement in the R377H channel is unaffected by external protons. (A) Gating current records from the R377H channel measured with the cut-open oocyte voltage clamp in symmetric NMDG-MS solutions, pH 7.4 in the inside and pH 9.2 in the outside (pHo). The superimposed currents are in response to various test pulses from a holding potential of −90 mV, also shown superimposed at the top. (B) The same sequence of superimposed gating current records from the same oocyte in pHo 5 external solution. (C) The voltage dependence of gating charge displacement (Q-V curves) for the ON- (left panel, open symbols) and OFF-gating currents (right panel, closed symbols) displayed in A (pHo 9.2, circles) and B (pHo 5, squares). (Experiment D11017h)
|
|
Figure 5. Gating charge displacement in the K374H channel is unaffected by internal protons. (A) Gating current records from the K374H channel measured from an excised inside-out macropatch in symmetric NMDG-MS solutions, pH 9.2 in the outside and pH 7.4 in the inside (pHi). The superimposed currents are in response to various test pulses from a holding potential of −90 mV, shown superimposed at the top. (B) Gating current in response to a 0 mV test pulse from the same macropatch during perfusion of pHi 5 internal solution. (C) Q-V curves for the ON- (open symbols) and OFF-gating currents (closed symbols) displayed in A (pHi 7.4, triangles). Each Q-V curve was fit to a sum of two Boltzmann distributions (lines). The internal side was perfused with pHi 5 solution, and test pulses to 0 mV were intermittently applied 20 s later to measure the saturating charge displacement. A test pulse to 20 mV was also applied once to confirm saturation of charge displacement. The average of the charge displaced in the ON- and OFF- gating currents from seven of these 0-mV test pulses and the 20 mV pulse during pHi 5 perfusion are shown (squares). (D) Qon (open symbols) and Qoff (closed symbols) in response to 0 mV test pulses measured in pHi 7.4 (triangles) and during internal pHi 5 perfusion (squares). The lines through the data are linear regressions. (Experiment D11077b)
|
|
Figure 6. Gating charge displacement in the K374H channel is unaffected by external protons. (A) Gating current records from the K374H channel measured in the cut-open oocyte configuration in symmetric NMDG-MS pH 9.2 solutions (pHi 9.2 and pHo 9.2). The superimposed currents are in response to various test pulses from a holding potential of −90 mV, also shown superimposed at the top. (B) The same sequence of superimposed gating current records from the same oocyte in pHo 7.4 external solution. (C) The same sequence of superimposed gating current records from the same oocyte in pHo 5 external solution. (D) Q-V curves for the ON- (left panel, open symbols) and OFF-gating currents (right panel, closed symbols) displayed in A (pHo 9.2, circles), B (pHo 7.4, triangles), and C (pHo 5, squares). (Experiment D06088a)
|
|
Figure 7. Gating currents of the R368H channel displace titrateable charge and transport protons. Using the cut-open oocyte voltage clamp, R368H channel gating currents were recorded in internal HB solution, pH 7.4, and various external HB solutions (A, pHo 5.2; B, pHo 7.4; C, pHo 9.2). The intracellular pH (pHi) was measured with a H+-sensitive electrode. All gating currents were recorded from the same membrane area. The gating currents in each pHo group were elicited by a family of test pulses (in millivolts) from a pre- and postpotential pulse of −110 mV (represented at the top of each group). The test pulse value corresponding to each current is shown on the left. The normalized charge displacement in each gating current (Rel. Q, shown to the right of each trace) was obtained by integrating the OFF-gating currents (Fig. 9 B) and normalizing to the maximum value of each pHo group. Linear leak current was subtracted from each current trace off-line as described in materials and methods (Data Analysis of I-V Curves). (Experiment D11200a)
|
|
Figure 8. Titration of gating charge displacement in the R368H channel. The gating currents in response to a 50-mV test pulse, were recorded in three pH gradients from the same oocyte described in Fig. 7. The current records are shown here superimposed along with the corresponding gating charge displacement obtained by integrating the gating currents over the duration of the ON and OFF potential pulses. (Experiment D11200a)
|
|
Figure 9. Voltage dependence of proton transport and gating charge displacement in the R368H channel. A family of pulse-evoked R368H channel gating currents and pHi were simultaneously measured in five different pH gradients across the membrane, as described and shown for three of the gradients in Fig. 7. pH gradients were established using HB recording solutions by leaving the internal solution constant and varying pHo: pHo/pHi 5.2/6.6 (▪), pHo/pHi 6.3/6.9 (▾), pHo/pHi 7.4/7.0 (▴), pHo/pHi 8.3/7.1 (♦), and pHo/pHi 9.2/7.1 (•). (A) Voltage dependence of steady-state proton current amplitudes in five pH gradients (proton transport I-V curves). The Nernst equilibrium potential established by each pH gradient, EH+, is also displayed. All five proton current I-V curves were simultaneously fit to an expression for proton current values predicted from a titrateable voltage sensor model (see theory, second term of ) with voltage-dependent pKas (). The best fit values for pKo and pKi are shown in two forms: first, pK(V) = pK(0) ± δ(FV/2.3RT) to highlight δ, the fraction of the electric field sensed, followed by pK(V) at room temperature to highlight the slope of the voltage dependence, where V is in units of millivolts. Other parameters of the fit are: N = 1.82e10, z = 3.09, z1= 0.061, δα1 = 0.012, δα2 = 0.376, α1(0) = 8464/s, α2(0) = 67215/s, and β1(0) = 4775/s. (B) Corresponding Q-V curves for the OFF-gating currents in five pH gradients. Total gating charge per channel (Q/N) was estimated by fitting Qmax versus pHo to the Henderson-Hasselbach equation (Q/N = Qmin + {(Qmax − Qmin)/[1 + exp[2.3(pHo − pKo)]]}). The fit predicted a pKo of 5.8 and saturated at a maximum value of 8.37 nC. Assuming that the fully protonated voltage sensor displaces 12.5 e0 per channel as the wild-type, N = 8.37 nC/12.5 e0 = 4.2e9. (Experiment D11200a)
|
|
Figure 10. Gating currents of the R371H channel. Using the cut-open oocyte voltage clamp with a H+ sensitive electrode, R371H channel gating currents and pHi were simultaneously measured in three pH gradients: pHo/pHi 8.3/6.6 (A), pHo/pHi 6.3/6.2 (B), and pHo/pHi 5.2/5.8 (C). The gradients were established with HB recording solutions; the external solution was varied while the internal pH 5 solution was left unchanged. All gating currents were recorded from the same membrane area. The gating currents in each pHo/pHi group (shown superimposed in A–C) were elicited by the pulse protocol series shown at the top. The left panel of A–C is an enlargement of the ON-gating currents shown fully in the right panel. Linear leak current was subtracted from each current trace off-line as described in materials and methods (Data Analysis of I-V Curves). (D) Voltage dependence of steady-state proton current amplitudes (proton current I-V curves) measured in three pH gradients: pHo/pHi 8.3/6.6 (♦), pHo/pHi 6.3/6.2 (▾), and pHo/pHi 5.2/5.8 (▪). IH was determined by taking the isochronal, steady-state amplitudes of the ON-gating currents shown in A–C. The linear leak component has already been subtracted out as described in materials and methods. The Qoff-V curve associated with each I-V curve was used to generate the voltage dependence of the gating capacitance, Cg. Each Cg-V curve (lines) was scaled to its associated I-V curve. (Experiment D11210b)
|
|
Figure 11. Proton transport and conduction by the R371H channel voltage sensor. I-V plots of the steady-state proton currents, IH, measured in five different pH gradients: pHo/pHi 9.2/7.8 (•), pHo/pHi 8.3/6.6 (♦), pHo/pHi 7.4/6.34 (▴), pHo/pHi 6.3/6.2 (▾), and pHo/pHi 5.2/5.8 (▪). The Nernst equilibrium potential established by each pH gradient, EH+, is also displayed. Gating currents and pHi were measured simultaneously as described and shown for three of the gradients in Fig. 10. All five proton current I-V curves were simultaneously fit to an expression for proton current values (see theory, ) predicted from a model of a titrateable voltage sensor that forms a proton pore upon depolarization. The proton current predicted by the model is the sum of a proton transport current, IT, and a proton pore current, IP. The parameters of the best fit to the model (shown as lines) are: pKo(0) = 6.67, pKi(0) = 8.07, δo = 0.460, δi = 0.373, δip = 0.934, γo = 93.5/s, ɛo(0) = 4.33e8/(M s), γip = 138/s, N = 3.08e10, z = 1.14, z1 = 0.844, δα1 = 0.139, δα2 = 0.879, α1(0) = 7247/s, α2(0) = 818/s, and β1(0) = 2,369/s. (Experiment D11210b)
|
|
Figure 12. Gating currents of the E283Q/R371H channel. (A) Using the cut-open oocyte voltage clamp, E283Q/R371H channel gating currents were measured in symmetric NMDG-MS solutions with internal pH 9.2, and either pHo 9.2 (top currents) or pHo 5 (bottom currents). All gating currents were recorded from the same membrane area. The gating currents in each pHo group (shown superimposed) were elicited by the pulse protocol series shown at the top. (B) Q-V curves for the ON- (open symbols) and OFF-gating currents (closed symbols) displayed in A (pHo 9.2, circles; pHo 5, squares). (C) The voltage dependence of steady-state ON-gating currents shown in A (circles, pHo 9.2; squares, pHo 5). The linear leak component has already been subtracted out, as described in materials and methods (Data Analysis of I-V Curves). (Experiment D11037a)
|
|
Figure 13. A model of voltage sensor operation. A model illustrating the movement of the Shaker K+ channel S4 segment from a hyperpolarized position (left) to a depolarized position (right). Only five transmembrane segments (S1–S5) of one subunit are shown for clarity. The hydrophobic region surrounding the internal crevice is represented as a hatched cloud shaped by the membrane and the sides of S1 and S5 facing each other. Depolarization causes a rotation of the S4 segment and movement of the first four S4 charges from the internal crevice to an externally facing crevice. The external crevice cannot be seen; it faces the back between the S2 and S3 segments.
|
