|
Figure 1. Increases in pH result in increased Slo3 current activation. (A) An inside-out patch expressing Slo3 channels was bathed with 0 Ca2+ solutions with pH set to the indicated nominal values. Patches were activated by the indicated voltage protocol. Minimal current activation is observed with pH 7.0 even at +240 mV. Increases in pH result in faster and more robust current activation. (B) Tail currents from the panels in A are shown on a faster time base on the left and with larger amplification on the right. Note that even at the highest pH, the tail currents do not exhibit saturation in activation of conductance.
|
|
Figure 2. pH dependence of Slo3 macroscopic conductance. In A, steady-state currents were measured from records as in Fig. 1 and converted to conductances. For results from any individual patch, G/V curves were normalized to the maximal conductance observed with pH 9.0. The full set of pH was not tested on all patches, but each point is the mean of 4â18 patches. In B and C, macroscopic conductance was determined both from steady-state currents (B) or tail currents (C) for a set of four patches at pH of 7.4, 7.8, 8.0, and 9.0. For tail current G/Vs a single current value at 100 μs after the nominal onset of the voltage step was used.
|
|
Figure 3. Empirical properties of Slo3 G/V curves. In A, the voltage of half activation (Vh) is plotted as a function of proton concentration. Vh estimates were based on Boltzmann fits (Eq. 1) in which Gmax was not constrained. In B, Gmax obtained from fits of Eq. 1 to G/V curves at each pH are plotted as a function of [H+]. The solid line is a fit of Hill equation (G([H+]) = Gmax/{1 + ([H+]/Kd)n}), yielding an effective Kd of 23.3 ± 2.9 nM and with n = 1.9 ± 0.4. In C, the normalized conductance is plotted as a function of [H+] for a range of command potentials. Solid lines show the best fit of a similar Hill function to conductance estimates at each voltage. The range of estimates for Kd was 15â21 nM corresponding to pH of 7.7â7.8, with minimal voltage dependence in the estimates. Estimates of n ranged from 1.9 to 2.2.
|
|
Figure 4. There is appreciable pH-dependent Slo3 current at negative potentials. In A, families of Slo3 currents activated by the indicated voltage protocol are shown for pH 7.0 and 8.5. Test steps to potentials from â300 to +300 mV were preceded by a conditioning step to â100 mV. Each trace corresponds to a single sweep. In B, high gain records for traces in A at negative potentials are displayed on a faster time base showing the absence of channel activity at pH 7.0. In each case, a brief period at 0 mV preceded the step to â100 mV, before the final test step. In C, similar traces at pH 8.5 show increased variance due to channel activity. In D, current traces for the same range of voltages at pH 7.0 are shown for an inside-out patch from an oocyte injected with DEPC. In E, traces are from the same patch as in D, but at pH 8.5, showing that pH increases leak current without any obvious increase in current variance.
|
|
Figure 5. Evaluation of the average Slo3 current magnitude at negative potentials. In A, segments of currents obtained at either pH 7.0 or pH 8.5 are compared over a range of voltages for the patch shown in Fig. 4. Dotted line corresponds to the zero current level, and the dashed line corresponds to a baseline that reflects a leak conductance that is voltage independent over this range of voltages. The baseline level was chosen so that the range of current levels positive to this baseline was similar to the variance observed around the baseline for currents at pH 7.0. In B, total amplitude histograms for traces at pH 7.0 and pH 8.5 are shown, with vertical lines corresponding to the baseline used in the two cases. Note the asymmetry in the traces at pH 8.5 and symmetry at pH 7.0. The Slo3 conductance at negative potentials was estimated based on the integral of current values in the distributions at pH 8.5, after baseline subtraction.
|
|
Figure 6. Voltage dependence of Slo3 conductance at negative potentials. In A, the voltage dependence of leak conductance in inside-out patches from DEPC-injected oocytes is displayed at each of three pH, 7.0, 7.6, and 8.5. The increase in leak conductance with increases in pH is not associated with any visually observable increase in current variance. In B, the voltage dependence of Slo3 conductance over the range of â300 to +300 mV is shown for two different methods of correction for leak current. Solid circles correspond to conductance estimates in which the leak conductance observed at pH 7.0 was subtracted on a patch by patch basis from the conductance observed at pH 8.5. This represents the maximum possible estimate of the Slo3 conductance at negative potentials. Open symbols correspond to conductance estimates derived from defining a linear leak conductance at pH 8.5 based on the most positive current over voltages from â100 through â240 mV (as illustrated in Fig. 6). This represents the minimal possible estimate of Slo3 conductance at negative potentials. In C, Slo3 conductance estimates at pH 8.5 over the range of â300 through â220 mV were fit by G(V) = L(0)*exp(zL*V/kT), yielding an estimate of zL of 0.04 ± 0.08. After conversion of normalized conductance to absolute Po (Zhang et al., 2006), L(0) was 1.25 ± 0.10 * 10â3. In D, the log(G) vs. V relationships for pH 8.5, 7.6, and 7.0 are shown.
|
|
Figure 7. Slo3 steady-state conductance can be described by Scheme 1. In A, the log(Po) vs. voltage relationships at pH 7.0, 7.6, and 8.5 were fit (red line) with Scheme 1, assuming zL = 0.04, and constraining C and E. Macroscopic conductances were converted to Po based on the estimates of unitary current Po in the accompanying paper (Zhang et al., 2006). In B, qa, the mean activation charge displacement, was determined from the slope of the log(Po) vs. V relationship. Each qa estimate was determined from the slope of the log(Po) vs. V values for each sequential set of four voltages, covering a 60-mV range. Error bars correspond to the confidence limit on the fit to the slopes. The red line corresponds to the expected qa/V relationship based on the fit shown in A. In C, the data in A were replotted on a linear scale to show the adequacy of the fit at higher Po values. In D, the predictions of the allosteric derived from the fit shown in A are compared with the Po vs. V relationship shown in Fig. 2 for a different set of patches in which currents were measured over a wider range of pH. Red lines correspond to pH 7.0, 7.6, and pH 8.5, respectively, with solid symbols reflecting those concentrations. In E, estimates of the voltage dependence of single channel Po provided in the accompanying paper (Zhang et al., 2006) show good agreement with the expectations derived from fitting A.
|
|
Figure 8. The primary differences between the voltage dependence of Slo3 and Slo1 conductance arise from differences in zL and D. In A, predicted log(Po /V) relationships for Slo3 based in parameters in Table I for pH 7.0, 7.6, and 8.5 are shown in red, along with predictions for Slo1 based on values from Horrigan and Aldrich (2002; and Table I) for [Ca2+] of 0.01, 1, and 10 μM. In B, the same relationships are plotted on a linear Po scale. In C, the value of D for Slo3 was changed from 4 to 25 to approximate that for Slo1, resulting in a similar maximal saturating Po at each pH. In D, the value of D for Slo1 was changed from 25 to 4. In E, zL for Slo3 was changed from 0.04 e to 0.3 e to approximate that for Slo1, changing both the limiting slope at negative potentials and the tendency toward saturation at positive potentials. In F, the value of zL for Slo1 was changed to 0.04 e. In G, values for zL and D for Slo3 were simultaneously changed to those describing Slo1, while in H, values for zL and D for Slo1 were changed to those characteristic of Slo3. In both cases, simultaneously changing both zL and D in large measure accounts for the characteristic differences between Slo1 and Slo3.
|
|
Figure 9. Slo3 activation time course is defined by two exponential components. In A, the Slo3 activation time course is illustrated for pH from 9.0 to 7.4 (top to bottom) at voltages from +40 to +240 mV. Blue lines indicate single exponential fits (I(t) = Aexp(ât/Ï) + C)) to the activation time course, which at pH 7.6, 7.8, and 8.0 fails to describe the activation time course. In B, the same traces are shown on a logarithmic time axes along with either a single exponential (left traces) or a two exponential (I(t) = Asexp(ât/Ïs) + Afexp(ât/Ïf) + C; right traces) fit to the activation time course. The activation time course is best described by two exponential components.
|
|
Figure 10. Changes in Slo3 activation time course is more strongly influenced by pH than voltage. Traces from Fig. 9 were normalized and grouped to illustrate the dependence of activation time course either on voltage (A) or pH (B). In A, traces on the top show the activation time course at pH 9.0 for +160, +200, and +240 mV on either a linear (left) or logarithmic time base. Blue lines correspond to a fit of two component exponential function. Traces in the bottom panels of A show the corresponding activation time course for pH 7.6 for +160, +200, and +240 mV. In B, the activation time course for normalized currents is compared as in A but for pH 7.4, 7.6, 7.8, 8.0, and 9.0 at +240 mV (top sets of traces) and at +160 mV (bottom sets of traces). Blue lines are two component exponential fits.
|
|
Figure 11. Properties of two components of slo3 activation. In A, fast (red) and slow (blue) time constants of activation at the indicated pH are plotted as a function of activation potential. Error bars are SEM of seven to eight estimates. The single exponential described activation at pH 9.0 closely approximates the fast time constant of activation, while the single exponential describing activation at pH 7.4 approximates the slow time constant. The fast time constant appears to become somewhat faster with depolarization. In B, the percent of the slow activation component is plotted as a function of voltage for three different pH. The relative contribution of the slow component does not change with command potential, but decreases with increases in pH. In C, the percent of the slow component is plotted as a function of pH for two different voltages. The percent of the slow component increases as pH is decreased. At pH 7.4, slow component contribution approaches 100%, while at pH 9.0, it is less than 5%. In D, the absolute contribution of the slow component to normalized conductance is plotted as a function of voltage for different pH. In E, the contribution of the fast component to conductance is plotted as a function of voltage for different pH. All conductance values in D and E were normalized within patches to the maximum conductance observed at pH 9.0. In F, the voltage dependence of the sum of the fast and slow component of conductance is illustrated, approximating the macroscopic steady-state G/Vs.
|
|
Figure 12. Slo3 deactivation contains two components. In A, the indicated voltage step protocol was used to examine tail current properties at potentials from â200 to +180 at various pH as indicated. Peak tail current at â200 mV is markedly nonohmic in comparison to current at +200 mV. 50 kHz sampling rate; 10 kHz filter. In B, the tail currents in A at potentials from +40 to â200 mV shown on a faster time base to emphasize the fast and slow components of the decay process. Traces on the left are shown on linear time scale, while those on the right are plotted on a logarithmic time scale.
|
|
Figure 13. Dependence of deactivation on voltage and pH. In A, tail currents at pH 8.5 are shown on a linear time scale for â40, â120, and â200 mV. In B, traces from A are shown on a logarithmic time scale along with a two component exponential fit (red lines) to the decay process. In C, tail currents from B were normalized to their maximum amplitude to illustrate the decrease in the relative amplitude of the slow exponential component at more negative potentials. In D, tail currents at â200 mV are compared for pH 7.6, 8.0, and 8.5. In E, the traces in D are plotted on a logarithmic time scale. For the example at pH 8.5, the best fit of a single exponential function (blue line) and a two component exponential function (red line) are compared. Time constant for the single exponential fit was 0.28 ms, while for the two component fit, Ïf = 38 μs and Ïs = 0.89 ms. In F, normalized tail currents at â200 mV, but at different pH, show a similar time course. Fits are as in E. In G, slow (Ïs) and fast (Ïf) time constants from tail currents are plotted as a function of potential for pH 7.4, 8.0, and 9.0. For Ïf, the line corresponds to Ïf(V) = Ïf(0)exp(zV/kT), where Ïf(0) = 0.071 ± 0.004 μs and z = 0.147 ± 0.01 e. In H, the voltage dependence of the contribution (percent) of the slow component to the tail current is shown for pH 7.4, 8.0, and 9.0.
|
|
Figure 14. Residues implicated in voltage sensing in mSlo1 are conserved in mSlo3. Alignment compares mSlo1 and mSlo3 sequence from S1âS4. Boxed regions indicate S1âS4 defined by standard hydropathy plots, while black horizontal lines highlight putative α-helical segments defined from structural analysis (Jiang et al., 2003). Charged residues highlighted in red are identical between Slo1 and Slo3, while other identical residues are shaded in gray. Charged residues that are not conserved are highlighted in blue. All residues shown to influence both zJ and zL in Slo1 (Ma et al., 2006) are conserved between Slo1 and Slo3 (â¾). A number of charged Slo1 residues shown to have no effect on either zJ or zL (â¡) are not strictly conserved in Slo3, while mutation of R207 has novel effects on zL.
|
