|
Scheme S1.
|
|
Scheme S2. (SCHEME III)
|
|
Scheme S4.
|
|
Figure 1. Qualitative effects of S(−)-CPB on the ClC-0 mutant C212S. (A–C) Family of voltage clamp traces recorded from one inside-out patch expressing the point mutant C212S of ClC-0. Currents were elicited using the pulse protocol shown in the inset. The concentration of intracellularly applied S(−)-CPB is indicated. (D) The apparent open probability (symbols) obtained from the initial current at the −100 mV tail pulse (see materials and methods) together with fits of to the data (solid lines).
|
|
Figure 2. Inhibition of single protopores. A single C212S channel present in an inside-out patch was measured at a holding potential of −80 mV (top traces) and −100 mV (bottom traces) in the absence (left traces) and presence (right traces) of 1 mM S(−)-CPB. Only short stretches (4.3 s) of longer recording periods (>60 s for the recordings in the presence of CPB) are show for each condition. Next to each trace is shown (on the same current scale as the current trace) an amplitude histogram of the complete recording together with a fit of the sum of three Gaussian components. The peaks of the Gaussian fits are indicated as dashed lines on the current traces. The respective area of each Gaussian component was used to calculate the relative open probability of each conductance state (see Saviane et al. 1999) plotted in Fig. 3. Solutions are given in materials and methods. Traces were filtered at 1 kHz for the analysis.
|
|
Figure 3. Binomial analysis of the single pore block. The occupancy of each of the three conductance levels (0, 1 “pore” open, 2 “pores” open) obtained from the Gaussian fits of Fig. 2 are plotted as bars for each of the four conditions of Fig. 2 ([A] V =−80mV, no CPB; [B] V =−80 mV, 1 mM S(−)-CPB; [C] −100 mV, no CPB; [D] −100 mV, 1 mM S(−)-CPB). The squares in each graph indicate the best fit assuming an independent gating of two identical protopores according to ([A] p = 0.37; [B] p = 0.15; [C] p = 0.21; [D] p = 0.06). Mean values (n ≥ 3, ±SD) were as follows: for −80 mV, 0 CPB, p = 0.41 ± 0.03; for −80 mV, 1 mM CPB, p = 0.19 ± 0.03; for −100 mV, 0 CPB: p = 0.25 ± 0.03; for −100 mV, 1 mM CPB, p = 0.09 ± 0.02; for −120 mV, 0 CPB, p = 0.18 ± 0.01; and for −120 mV, 1 mM CPB, p = 0.05 ± 0.01.
|
|
Figure 4. Kinetic analysis of a single channel at −80 mV. The long recordings shown in A (control) and B (with 1 mM CPB in the bath solution; from the same patch) were subjected to idealization after filtering at 500 Hz as described by Saviane et al. 1999. Traces were filtered at 50 Hz for display. Dwell times of the three conductance levels were fitted with a maximum likelihood procedure (Saviane et al. 1999). In the absence of CPB dwell times (C–E, circles) could be well fitted with single exponential functions (double open levels, [C] τ2 = 10.8 ms; single open levels, [D] τ1 = 14.6 ms; closed times, [E] τ0 = 18.8 ms; see solid lines). In the presence of CPB, the dwell distribution of the double open conductance level (C, squares) could be well fitted with a single exponential with the same time constant as in the absence of CPB. The dwell time distribution of the single conductance level in the presence of CPB (D, squares) was much better described by the sum of two exponential functions that were fixed at 14.6 and 23 ms, respectively, as described in Kinetic Analysis of Single Channels. The distribution of zero current epochs (E, squares) was fitted with the sum of three exponential functions (E, dashed line) where two of the time constants were fixed to 18.8 and 37.6 ms, respectively, as described in the text and a larger time constant was determined by the fit to 220 ms. Qualitatively similar results were obtained in two patches at −80, −100, and −120 mV.
|
|
Figure 5. Complete unblock at positive voltages. To test if mutant C212S is slightly blocked by S(−)-CPB at positive voltages, a test pulse of the form shown in A was applied every 3 s. The intracellular solution of the inside-out patch was repeatedly switched between CPB-free and 5 mM S(−)-CPB containing solution. (B) The current responses after consecutive solution changes. The dashed line indicates the steady-state current level at 60 mV under all conditions. Clearly, the relief from inhibition at this voltage is practically complete. The dotted line indicates the zero current level.
|
|
Figure 6. CPB-bound protopores do not contribute to steady-state conductances. The fast component of tail currents at −100 mV is used to identify the fraction of open pores that are drug-free at the end of a conditioning prepulse. Tail currents after prepulses of 700 ms duration to various voltages (traces are shown for −160, −60, and +40 mV) in the absence (A) and after addition of 1 mM CPB to the intracellular solution (B). The smooth lines in A are fits with single exponentials with the same time constant τf = 19.5 ms. The smooth lines in B are fits with double exponentials, comprising a fast component with the same time constant, τf = 19.5 ms, and a slower one, τs = 253 ms, equally independent of the prepulse. (C) The ratio ΔIf/I0 of the amplitude of the fast exponential relaxation, ΔIf, to that of the initial current, I0, is plotted as a function of the prepulse voltage, both for CPB-free conditions (squares) and for 1 mM CPB (circles). (mean values ± SD, n = 4).
|
|
Figure 7. CPB-bound protopores are not conductive during recovery from inhibition. C212S channels in a patch exposed to 5 mM CPB were first maximally inhibited by a step to −140 mV for 2 s, the brought to +60 mV for variable times, Tp, and finally brought to −100 mV. The records shown in A, obtained for Tp = 10, 30, 50, and 110 ms, show a progressive decrease of the initial inhibition. However, the tail relaxation is in all cases a double exponential (smooth lines) with the same fast and slow time constants. (B) Plot of the relative amplitude of the fast relaxation, ΔIf/I0, as a function of Tp; at all times during the removal of inhibition all conducting pores undergo fast relaxations as in the absence of CPB.
|
|
Figure 8. Study of the voltage dependence of CPB-inhibition. Two different pulse protocols are used to study the voltage dependence of on (A) and off (B) CPB inhibition. Recordings obtained from the same patch in 5 mM CPB with protocol A or with protocol B are shown in C and D, respectively. For clarity, only a few current traces are shown at the test voltages indicated. (C) After full removal by the prepulse to +60 mV, the on kinetics and steady-state level of CPB inhibition are measured for various steps to more negative voltages, Vp; the current relaxations at Vp are fitted by double exponential functions (smooth lines) with a fast time constant, τf, very close to that of normal deactivation, and a much larger time constant, τs, reflecting CPB-binding kinetics. At the onset of the following tail pulse to +60 mV, all the noninhibited channels open within ∼1 ms (Accardi and Pusch 2000), yielding an “instantaneous” current proportional to the unbound probability, pU(Vp). (D) After strong inhibition by a prepulse to −140 mV, the conductance increase for various steps to more positive voltages is fitted by a double exponential function for Vp < −60 mV and by a single exponential function for larger Vp's (smooth lines). (E) Shown the pU = punbound values of the same patch shown in C for the single experiment using the pulse data shown in C. pU was calculated as described in the text. The solid lines are fits of , and are shown only for clarity.
|
|
Figure 9. Steady-state binding of CPB to C212S protochannels. (A) Mean values of pU are plotted versus the concentration of S(−)-CPB for the voltages indicated in the figure (n ≥ 4 patches for each data point; error bars indicate SD). The solid lines are fits of with the apparent dissociation constant, Kd, as free parameter at each voltage. The resulting Kd values are plotted in B as a function of voltage. The dashed lines were obtained from the simulation of Fig. 4 with the parameters given in Table .
|
|
Figure 10. Kinetic analysis of macroscopic currents. Fast and slow time constants, τf and τs, were determined using the pulse protocols described in Fig. 8. Time constants obtained with the protocol of Fig. 8 A (i.e., with a positive prepulse) are indicated as “on” and plotted with filled symbols, whereas time constants obtained with the protocol of Fig. 8 B (i.e., with a negative prepulse) are indicated as “off” and plotted as open symbols. Each point is the mean of at least three determinations and the error bars indicate ±SD. The straight solid lines that fit the slow and the fast time constants in the positive voltage range have a steepness of a apparent gating valence of 0.43 for the slow time constants and a valence of 0.58 for the fast time constant, and illustrates the exponential voltage dependence of the time constants at positive voltages. The dashed lines were obtained by the simulation with Fig. 4 (see materials and methods). The solid line through the fast time constants is the least squares fit of , using and , obtained for β0 = 6.8 s−1, zβ = −0.36.
|
|
Figure 11. Kinetics of CPB association (on) and dissociation (off). The apparent rates of association, ron, and dissociation, roff, were calculated from the steady-state unbound probability, pU, and from the slow time constants of current relaxations, τs, using . (A) Plots of mean values of ron and roff at −140 mV as a function of the CPB concentration, c; the straight lines are fits with roff = koff = 0.88 s−1 and ron = c · kon = c · 1.84 mM−1 s−1. (B) The apparent second order association rate constant, kon, and first order dissociation rate constant, koff, estimated as shown in A are plotted as a function of voltage (open symbols); For positive voltages, only koff could be obtained as pU is close to unity; the solid lines in B were obtained by the simulation with Fig. 4 (see materials and methods).
|
|
Figure 12. Effect of [Cl]e on the block. [Cl]e was changed from 110 mM (standard solution) to 10 mM (Cl− was replaced by glutamate) on outside-out patches with no or with 5 mM S(−)-CPB present in the intracellular (pipette) solution. (A) The effect of this maneuver on the apparent open probability in the absence of CPB. The parameters of the fits () are as follows: for high [Cl]e, pmin = 0.13, V1/2 = −88.8 mV, and z = 0.95; for low [Cl], pmin = 0.08, V1/2 =−51.9 mV, and z = 0.79. Using the analysis described for normal [Cl] the apparent Kd (B) and the effective on (C) and off (D) rate constants were determined as a function of voltage. The dashed lines in B–D were obtained in the following way: an arbitrary function was fitted to the data in high [Cl]e and the same function was replotted, shifted by 38 mV. Solid lines in C and D were calculated from a simulation of Fig. 4. The parameters were as in Table , except that the values for konC(0) and koffC(0) were adjusted slightly (see Table legend).
|
|
Figure 13. Effect of [Cl]i on the block. [Cl]i was changed from 104 mM (standard solution) to 14 mM (Cl− was replaced by glutamate) on inside-out patches with no or with 1 mM S(−)-CPB present in the intracellular solution. (A) The effect of this maneuver on the apparent open probability in the absence of CPB. The parameters of the fits () are as follows: for high [Cl]i, pmin = 0.16, V1/2 = −90 mV, and z = 0.98; low [Cl]i: pmin = 0.04, V1/2 = −57.8 mV, and z = 0.94. Using the analysis described for normal [Cl] the apparent Kd (B) and the effective on (C) and off (D) rate constants were determined as a function of voltage. The dashed line in C was obtained in the following way: an arbitrary function was fitted to the data in high [Cl]i and the same function was replotted, multiplied by a factor of five. Solid lines in C and D were calculated from a simulation of Fig. 4. The parameters were as in Table .
|
|
Figure 14. Open channel block of mutant K519E. Example traces of an inside-out patch without CPB (A) and with 5 mM CPB (B). The pulse protocol is as in Fig. 8 C, with longer pulse durations and a constant tail pulse to +60 mV. (C–E) Experiments analogous to those of Fig. 5 in which 5 mM CPB was repeatedly perfused and washed away. Note the incomplete relief from inhibition even at +140 mV. Bars indicate 0.5 s and 20 pA, respectively.
|
|
Figure 15. Voltage dependence of the block of mutant K519E. (A) From the mean values of the ratio of the steady-state current in presence and absence of CPB, I(CPB)/I(0), for CPB concentrations of 1, 5, and 10 mM the apparent dissociation constant, Kd, was obtained by a fit with for voltages ranging from −140 to +140 mV. The resulting Kd values are plotted as a function of voltage. The straight line represents a fit of a simple exponential function of the points at 60, 100, and 140 mV. It has a slope corresponding to an apparent electrical distance of 0.15. (B) Apparent open probability of the fast gate of the mutant K519E in the presence of 5 mM CPB. After long saturating pulse to voltages between −140 and +140 mV, the currents recorded at the beginning of a pulse to −140 mV were normalized and plotted as a function of the prepulse voltage. The solid line is the Boltzmann fit obtained with using parameters pmin = 0.085; V1/2 = −43.0 mV, and z = 0.90.
|
|
|
