Molecular mechanism of BK channel activation by the smooth muscle relaxant NS11021

Abstract

Large-conductance Ca2+-activated K+ channels (BK channels) are activated by cytosolic calcium and depolarized membrane potential under physiological conditions. Thus, these channels control electrical excitability in neurons and smooth muscle by gating K+ efflux and hyperpolarizing the membrane in response to Ca2+ signaling. Altered BK channel function has been linked to epilepsy, dyskinesia, and other neurological deficits in humans, making these channels a key target for drug therapies. To gain insight into mechanisms underlying pharmacological modulation of BK channel gating, here we studied mechanisms underlying activation of BK channels by the biarylthiourea derivative, NS11021, which acts as a smooth muscle relaxant. We observe that increasing NS11021 shifts the half-maximal activation voltage for BK channels toward more hyperpolarized voltages, in both the presence and nominal absence of Ca2+, suggesting that NS11021 facilitates BK channel activation primarily by a mechanism that is distinct from Ca2+ activation. 30 µM NS11021 slows the time course of BK channel deactivation at -200 mV by ∼10-fold compared with 0 µM NS11021, while having little effect on the time course of activation. This action is most pronounced at negative voltages, at which the BK channel voltage sensors are at rest. Single-channel kinetic analysis further shows that 30 µM NS11021 increases open probability by 62-fold and increases mean open time from 0.15 to 0.52 ms in the nominal absence of Ca2+ at voltages less than -60 mV, conditions in which BK voltage sensors are largely in the resting state. We could therefore account for the major activating effects of NS11021 by a scheme in which the drug primarily shifts the pore-gate equilibrium toward the open state.

Figure S1.

Representative fits of BK channel activation and deactivation time course with a single exponential for currents with 10 µM Ca²⁺ at the cytosolic side of the patch. (A) Representative macroscopic currents (black lines), activated by voltage step from −100 mV to between +40 mV and +200 mV (20-mV increments). Single exponential fits (Eq. 2) are shown as blue dashed lines. (B) Time constants corresponding to the fits in A, plotted as a function of voltage. (C) Representative macroscopic currents (black lines), driven by voltage steps from +100 mV to between −20 mV and −240 mV (20-mV increments). Single exponential fits are shown as blue dashed lines. (D) Time constants corresponding to the fits in C plotted as a function of voltage.

(Scheme 1)

(Scheme 2)

Figure S2.

G–V-Ca²⁺ relations with predictions from Scheme 1 using parameters from Fit A, with adjustable parameter J₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit A in Table 1. Lines represent the predictions with J₀ = 0.08. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with J₀ = 0.09 (0.1 µM), 0.12 (1 µM), 0.21 (10 µM), and 0.32 (30 µM). (F) Fitted values for J₀ versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in J₀ with increasing [NS11021] (but see more detailed explanation in Results).

Figure S3.

G–V-Ca²⁺ relations with predictions from Scheme 1 using parameters from Fit A, with adjustable parameter D. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit A in Table 1. Lines represent the predictions with D = 29. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with D = 33 (0.1 µM), 40 (1 µM), 57 (10 µM), and 74 (30 µM). (F) Fitted values for D versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in D with increasing [NS11021] (but see more detailed explanation in Results).

Figure S4.

G–V-Ca²⁺ relations with predictions from Scheme 1 using parameters from Fit B, with adjustable parameter J₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit B in Table 1. Lines represent the predictions with J₀ = 0.06. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with J₀ = 0.08 (0.1 µM), 0.11 (1 µM), 0.22 (10 µM), and 0.36 (30 µM). (F) Fitted values for J₀ versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in J₀ with increasing [NS11021] (but see more detailed explanation in Results).

Figure S5.

G–V-Ca²⁺ relations with predictions from Scheme 1 using parameters from Fit B, with adjustable parameter D. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit B in Table 1. Lines represent the predictions with D = 16. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with D = 18 (0.1 µM), 23 (1 µM), 42 (10 µM), and 61 (30 µM). (F) Fitted values for D versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in D with increasing [NS11021] (but see more detailed explanation in Results).

Figure S6.

G–V-Ca²⁺ relations from Scheme 1 using parameters from Fit B to describe changing parameter J₀ in addition to L₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit B in Table 1. Lines represent the predictions with L₀ = 1.0 × 10⁻⁵, J₀ = 0.06. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with L₀ = 1.6 × 10⁻⁵, J₀ = 0.06 for 0.1 µM NS11021; L₀ = 3.2 × 10⁻⁵, J₀ = 0.06 for 1 µM NS11021; L₀ = 9.6 × 10⁻⁵, J₀ = 0.07 for 10 µM NS11021; L₀ = 2.1 × 10⁻⁴, J₀ = 0.09 for 30 µM NS11021. (F) Fitted values for L₀ (black) and J₀ (red) versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in L₀ in combination with an increase in J₀, with increasing [NS11021].

Figure S7.

G–V-Ca²⁺ relations from Scheme 1 using parameters from Fit B to describe changing parameter D in addition to L₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit B in Table 1. Lines represent the predictions with L₀ = 1.0 × 10⁻⁵, D = 15. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with L₀ = 1.6 × 10⁻⁵, D = 16 for 0.1 µM NS11021; L₀ = 3.2 × 10⁻⁵, D = 16 for 1 µM NS11021; L₀ = 9.6 × 10⁻⁵, D = 19 for 10 µM NS11021; L₀ = 2.1 × 10⁻⁴, D = 21 for 30 µM NS11021. (F) Fitted values for L₀ (black) and D (red) versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in L₀ in combination with an increase in D, with increasing [NS11021].

Figure S8.

Description of BK channel activity at nominally 0 Ca²⁺ with 0 or 30 µM NS11021 using Scheme 2. τ versus voltage from patches with nominally 0 µM Ca²⁺, with 0 (filled circles) or 30 µM NS11021 (open circles). Lines represent fits with Scheme 2 using parameters in Table 3 and Table S3: 0 µM NS11021, red; 30 µM NS11021, adjusted for γ δ, solid blue; adjusted for α, β, γ, and δ, dotted blue; adjusted for α, β, γ, δ, and D, dashed blue; adjusted for γ, δ, and D, dashed purple. These suggest that the major effects of NS11021 on voltage-dependent gating kinetics may be explained in part by adjusting γ and δ in Scheme 2 in combination with adjustments in other voltage-dependent rate constants.

Figure S9.

Effect of NS11021 on V_1/2 over a range of [Ca²⁺]. V_1/2 plotted as a function of [NS11021] for nominally 0 (orange), 1 (green), 10 (blue), and 100 µM Ca²⁺ (black). Experimental data (mean ± SEM, circles) are plotted along with the predicted V_1/2 for Fit A (solid line) and Fit B (dotted line), with adjustable parameter L₀.

Figure S10.

G–V-Ca²⁺ relations from Scheme 1 using parameters from Fit A to describe changing parameter C in addition to L₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit A in Table 1. Lines represent the predictions with L₀ = 2.8 × 10⁻⁶, C = 2.5. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with L₀ = 4.3 × 10⁻⁶, C = 2.4 for 0.1 µM NS11021; L₀ = 9.0 × 10⁻⁶, C = 2.3 for 1 µM NS11021; L₀ = 3.4 × 10⁻⁵, C = 2.2 for 10 µM NS11021; L₀ = 8.7 × 10⁻⁵, C = 2.1 for 30 µM NS11021. (F) Fitted values for L₀ (black) and C (red) versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in L₀ in combination with a decrease in C, with increasing [NS11021].

Figure S11.

G–V-Ca²⁺ relations from Scheme 1 using parameters from Fit B to describe changing parameter C in addition to L₀. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed under Fit B in Table 1. Lines represent the predictions with L₀ = 1.0 × 10⁻⁵, C = 18. (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with L₀ = 1.6 × 10⁻⁵, C = 17 for 0.1 µM NS11021; L₀ = 3.2 × 10⁻⁵, C = 16 for 1 µM NS11021; L₀ = 9.6 × 10⁻⁵, C = 15 for 10 µM NS11021; L₀ = 2.1 × 10⁻⁴, C = 15 for 30 µM NS11021. (F) Fitted values for L₀ (black) and C (red) versus [NS11021]. These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in L₀ in combination with a decrease in C, with increasing [NS11021].

Figure 1.

Effects of NS11021 on BK channel gating. (A) Structure of NS11021. (B) Representative voltage protocols for channel activation (left) and deactivation (right). Numbers correspond to voltage (in mV) and duration of activating pulse (in ms). (C) Representative BK currents from an inside-out patch with 10 µM Ca²⁺ and the indicated [NS11021] at the cytosolic side of the membrane. “0 µM NS11021” experiments contained 0.1% DMSO (vehicle) with no added NS11021. (D) G–V relations from patches with 10 µM Ca²⁺ and the indicated [NS11021]. Solid lines represent fits with a Boltzmann equation. Values for Boltzmann equation parameters are provided in Table S1. (E) Time constant of relaxation (τ) versus voltage (10 µM Ca²⁺ and the indicated [NS11021]). Time constants were estimated by fitting activation and deactivation traces with a single exponential (Eq. 2 and Fig. S1). Increasing [NS11021] primarily slows channel deactivation.

Figure 2.

NS11021 activates BK channels over a range of [Ca²⁺]. (A) G–V relations from patches with nominally 0 µM Ca²⁺. (B) Time constant (τ) versus voltage from patches with nominally 0 µM Ca²⁺. Time constant at −200 mV (τ₋₂₀₀) for each [NS11021]: 0 µM, 0.15 ± 0.01 ms (n = 33); 0.1 µM, 0.16 ± 0.02 ms (n = 11); 1 µM = 0.36 ± 0.05 ms (n = 7); 10 µM, 0.65 ± 0.08 ms (n = 5); 30 µM, 0.79 ± 0.12 ms (n = 14). (C) G–V relations from patches with 1 µM Ca²⁺. (D) τ versus voltage from patches with 1 µM Ca²⁺. τ₋₂₀₀ for each [NS11021]: 0 µM, 0.21 ± 0.03 ms (n = 21); 0.1 µM, 0.20 ± 0.06 ms (n = 8); 1 µM, 0.34 ± 0.09 ms (n = 10); 10 µM, 0.78 ± 0.12 ms (n = 7); 30 µM, 1.41 ± 0.21 ms (n = 11). (E) G–V relations from patches with 100 µM Ca²⁺. (F) τ versus voltage from patches with 100 µM Ca²⁺. τ₋₂₀₀ for each [NS11021]: 0 µM, 0.53 ± 0.04 ms (n = 11); 0.1 µM, 0.56 ± 0.12 ms (n = 8); 1 µM, 0.99 ± 0.11 ms (n = 6); 10 µM, 2.01 ± 0.22 ms (n = 6); 30 µM = 3.22 ± 0.25 ms (n = 9). In A, C, and E, solid lines represent fits with a Boltzmann equation; parameters can be found in Table S1.

Figure 3.

Effects of NS11021 on truncated BK channels. (A) Topology of the truncated BK channel construct Slo1c-Kv-MinT, showing two of the four domains side by side. Arrow illustrates the permeation pathway. The VSD (S0–S4, white), PGD (S5–S6, purple), and 11 residues from the Kv1.4 tail (blue) are sufficient for trafficking of these voltage-gated channels to the plasma membrane. (B) Representative Slo1c-Kv-MinT currents (nominally 0 Ca²⁺). Patches were held at 0 mV and stepped to voltages ranging from +50 mV to +250 mV, followed by a step to +160 mV for tail current measurement. (C) G–V relations from Slo1c-Kv-MinT at nominally 0 Ca²⁺, with 0 (red) or 30 µM NS11021 (blue). V_1/2 shifted from 238 ± 8.1 mV (n = 13) to 161 ± 2.6 mV (n = 7) with addition of 30 µM NS11021. (D) G–V relations from Slo1c-Kv-MinT at 100 µM Ca²⁺, with 0 (red) or 30 µM NS11021 (blue). For these data, V_1/2 shifted from 214 ± 3.4 mV (n = 24) to 158 ± 2.0 mV (n = 5) with addition of 30 µM NS11021.

Figure 4.

NS11021 increases P_o at negative voltages in nominally 0 Ca²⁺. (A) Representative BK currents in nominally 0 Ca²⁺ at −80 mV, with 0 µM NS11021. NP_o = 2.5 × 10⁻⁴. (B) BK currents from the same patch as in A, following addition of 30 µM NS11021. NP_o = 1.5 × 10⁻². Patch was estimated to have 127 active channels, determined by dividing the maximum macroscopic current amplitude by the single-channel conductance. (C and D) Representative channel openings from the traces in A and B, respectively (from positions indicated by *), at an expanded time scale. (E) P_o versus voltage for 0 µM (red circles) and 30 µM NS11021 (blue triangles) activity. Data points represent mean P_o ± SEM from three to eight experiments. Lines represent best fit with P_o = L₀exp(−z_LV/k_BT). Parameters were 0 µM NS11021 (z_L = 0.57 e₀, L₀ = 2.3 × 10⁻⁵); 30 µM NS11021 (z_L = 0.55 e₀, L₀ = 1.1 × 10⁻³). Estimated numbers of channels in these multichannel patches ranged from 22 to 199.

Figure 5.

Effects of NS11021 on open and closed dwell time distributions. (A) Distribution of open times with 0 µM NS11021 (red circles). Line represents fit with a single exponential, τ = 0.13 ms. Distribution contains 396 events; this and subsequent distributions were normalized to contain 100,000 events. (B) Distribution of closed times with 0 µM NS11021 (red circles). Line represents fit with sum of two exponentials, with τ and percentage area of each component as indicated. Number of events = 395. (C) Distribution of open times with 30 µM NS11021 (blue circles) added to the same patch. Line represents fit with sum of three exponentials, with τ and percentage of area as indicated. Number of events = 11,115. (D) Distribution of closed times with 30 µM NS11021 (blue circles). Line represents fit with sum of four exponentials, with τ and percentage area as indicated. Number of events = 11,145. Distributions in A–D show activity in nominally 0 Ca²⁺, −80 mV.

Figure 6.

Estimate of apparent stoichiometry of NS11021 action. (A) Normalized G (G/G_max) versus [NS11021] in nominally 0 Ca²⁺. Data were grouped by voltage and fitted with a Hill equation (Eq. 3; black lines) to estimate Hill coefficient (n_H) and EC₅₀. Parameters at each voltage were 70 mV (circle; n_H = 1.2; EC₅₀ = 28 µM), 80 mV (square; n_H = 1.1; EC₅₀ = 20 µM), 90 mV (triangle; n_H = 1.0; EC₅₀ = 8.8 µM), 100 mV (diamond; n_H = 1.1; EC₅₀ = 6.4 µM), 110 mV (hexagon; n_H = 0.8; EC₅₀ = 3.4 µM), and 120 mV (tilted square; n_H = 1.2; EC₅₀ = 2.4 µM). (B) G/G_max versus [NS11021] with 1 µM Ca²⁺. 50 mV (circle; n_H = 1.0; EC₅₀ = 24 µM), 60 mV (square; n_H = 0.9; EC₅₀ = 13 µM), 70 mV (triangle; n_H = 0.7; EC₅₀ = 7.5 µM), 80 mV (diamond; n_H = 0.8; EC₅₀ = 4.0 µM), 90 mV (hexagon; n_H = 0.8; EC₅₀ = 2.7 µM), and 100 mV (tilted square; n_H = 0.8; EC₅₀ = 1.5 µM). (C) G/G_max versus [NS11021] with 10 µM Ca²⁺. −30 mV (circle; n_H = 1.6; EC₅₀ = 26 µM), −20 mV (square; n_H = 1.4; EC₅₀ = 21 µM), −10 mV (triangle; n_H = 1.0; EC₅₀ = 14 µM), 10 mV (hexagon; n_H = 0.7; EC₅₀ = 5.3 µM), and 20 mV (tilted square; n_H = 0.7; EC₅₀ = 4.9 µM). (D) Hill coefficients at each [Ca²⁺]; parameters at each voltage are shown as circles; mean at a given [Ca²⁺] is indicated by horizontal line, and error bars show the SEM. Means for each [Ca²⁺] are 0 Ca²⁺ (1.06 ± 0.06; n = 6), 1 µM Ca²⁺ (0.86 ± 0.04; n = 6), and 10 µM Ca²⁺ (1.08 ± 0.19; n = 5).

Figure 7.

G–V-Ca²⁺ relations with predictions from Scheme 1 using parameters from Fits A and B. (A) G–V-Ca²⁺ relations in the absence of NS11021 bathed in 0 µM (orange triangle), 1 µM (green circle), 10 µM (blue square), and 100 µM Ca²⁺ (black diamond). Parameters are listed in Table 1. Lines represent the predictions with L₀ = 2.8 × 10⁻⁶ (Fit A, solid lines) and L₀ = 1.0 × 10⁻⁵ (Fit B, dashed lines). (B–E) G–V-Ca²⁺ relations in 0.1, 1, 10, and 30 µM NS11021, respectively. Lines represent predictions with L₀ = 4.3 × 10⁻⁶ (0.1 µM), 9.0 × 10⁻⁶ (1 µM), 3.4 × 10⁻⁵ (10 µM), and 8.7 × 10⁻⁵ (30 µM) for Fit A, and L₀ = 1.6 × 10⁻⁵ (0.1 µM), 3.2 × 10⁻⁵ (1 µM), 9.6 × 10⁻⁵ (10 µM), and 2.1 × 10⁻⁴ (30 µM) for Fit B. (F) Fitted values for L₀ versus [NS11021] (Fit A as black/filled circles, Fit B as white/open circles). These suggest that the major effects of NS11021 on G–V-Ca²⁺ relations may be explained in part by an increase in L₀ with increasing [NS11021].

Figure 8.

Description of BK channel activity at nominally 0 Ca²⁺ using Schemes 1 and 2. (A) P_o versus voltage with 0 μM NS11021 (red circles) and 30 µM NS11021 (blue triangles). Solid line represents P_o predicted from Scheme 1 using parameters in Table 1, Fit A. Predicted P_o with 30 µM NS11021 was generated by using Fit A substituting L₀ = 8.7 × 10⁻⁵ (solid thin line), J₀ = 0.32 (dashed line), or D = 74 (dotted line). Using parameters from Fit A in Table 1 resulted in a χ² value of 3.85. χ² values obtained by changing only one parameter to account for P_o at 30 µM NS11021: for L₀, 0.41; for J₀, 0.77; for D, 3.15; thus, substitution of L₀ resulted in the lowest χ² value (best fit) for these data. (B) P_o versus voltage as in A, except with solid line showing P_o predicted using Fit B. P_o with 30 µM NS11021 was generated with Fit B by substituting L₀ = 2.1 × 10⁻⁴ (solid thin line), J₀ = 0.36 (dashed line), or D = 61 (dotted line). Using parameters from Fit B resulted in a χ² value of 1.18. χ² values obtained by changing only one parameter to account for P_o at 30 µM NS11021: for L₀, 1.91; for J₀, 3.13; for D, 4.73. Again, substitution of L₀ resulted in the lowest χ² value (best fit) for these data. All χ² values for A and B based on five total data points. (C) τ versus voltage from patches with nominally 0 µM Ca²⁺, with 0 (filled circles) or 30 µM NS11021 (open circles). Lines represent fits with Scheme 2 using parameters in Table 3: red line for 0 µM NS11021; solid blue line for 30 µM NS11021 adjusted for γ δ; dashed blue line for 30 µM NS11021 adjusted for α, β, γ, δ, and D. (D) τ versus voltage as in C with solid lines showing fits with Scheme 2 using parameters in Table 3: red line for 0 µM NS11021; green line for 30 µM adjusted for α β; purple line for 30 µM adjusted for D. Adjusting α and β or D can only describe the gating kinetics with 30 µM NS11021 at negative voltages (less than −100 mV) when in combination with adjustment of γ and δ.

(Scheme 3)