Electrophysiological characterization of the state-dependent inhibition of Kv7.1 and IKs by UCL2077

Figures

• 

  Fig. 1. UCL2077 is an extremely potent inhibitor of Kv7.1 channels. (A) Chemical structure of UCL2077. (B) Representative Kv7.1 current traces were recorded using the whole-cell patch clamp technique on tsA201 cells with increasing concentrations of UCL2077 (shades of red: 10 pM to 3 nM) and 300 nM HMR1556 (in gray), as indicated. Recordings were obtained by applying a 4-second depolarizing pulse to 60 mV from a holding potential of –80 mV followed by 2 seconds to –40 mV (protocol shown below). (C) The average concentration-response curve (mean ± SEM, n = 3–7) was measured at the peak currents (filled circles) and at the tail currents (empty squares) after subtraction of the HMR1556-insensitive currents and fitted to the Hill equation (eq. 1). IC50s were 450 and 70 pM at the peak and at the tail currents, respectively (Table 2).

• 

  Fig. 4. UCL2077 modifies the GV relationships of Kv7.1. (A) Kv7.1 currents elicited by a series of 4-second depolarizing pulses from –80 to +60 mV in 5 mV steps, followed by 2 seconds at –40 mV to elicit tail currents (protocol below) in the absence (black) and in the presence (red) of 100 pM UCL2077. For clarity, only traces every 20 mV are depicted. (B) Conductance-voltage (G-V) relationships were measured from tail currents (see symbols in panel A) and fit to a Boltzmann function (eq. 6). Summary mean ± SEM normalized G-V curves in the absence (black circles) and in the presence (red squares) of 100 pM UCL2077, n = 5. Filled squares are normalized to their own Gmax, whereas empty squares are normalized to their respective control’s Gmax. (C) Before-after relationship of the Boltzmann slope (k, top) and V50 (bottom), n = 5. Empty symbols correspond to the representative current traces shown in panel A. (D) The voltage dependence of the normalized current of Kv7.1 was analyzed at the peak currents (black circles), at the tail currents (empty squares), and at the end of the 4-second depolarizing pulse (gray diamonds), n = 5. Dashed line: control G-V relationship. Datapoints above –10 mV…

• 

  Fig. 2. UCL077 induces biphasic current decay in Kv7.1. (A) Representative current traces normalized to their maximum peak current before (black) and during (red) application of 300 pM UCL2077. Current decay in the absence and in the presence of UCL2077 was measured as a proportion of the peak current to the end of the depolarization: (Ipeak – Iend)/Ipeak. (B) Summarized before-after values from individual experiments (n = 6 throughout the figure). Empty circles correspond to the shown representative trace currents in panel A. (C) The UCL2077-dependent current decay enhances both the fast and the slow components of the current decay, measured before (left) and during (right) application of 300 pM UCL2077. (D) The fast component was fitted to a single exponential function (eq. 2) from which the time constant and amplitude (relative to the maximum peak current) were obtained. The difference between the extrapolation of the exponential fit and the real current level at the end of the depolarization was the amplitude of the slow component (see panel C), expressed as a proportion of the maximum peak current. Statistics and exact P values for all the comparisons can be found in Table 3. ∗P <…

• 

  Fig. 3. Kinetic analysis and onset of the UCL2077-mediated inhibition of Kv7.1. (A) Fractional inhibition traces were obtained by calculating the difference between control and UCL2077-exposed current traces pulsed to +60 mV and normalizing it sample by sample to the control trace, ie, (Icontrol – IUCL2077)/Icontrol for each datapoint. The traces were better fit to a biexponential equation with a fast (τfast) and a slow (τslow) component (eq. 4). (B) The concentration dependence of 1/τfast was fitted to a linear regression (n = 4–7), from which the slope approximates the rate of UCL2077 binding (kON), and the Y-intercept unbinding (kOFF). The ratio kOFF/kON estimates an IC50 of 580 pM (eq. 5). The slow component was not well estimated during 4-second pulses, so we did not analyze its concentration dependence. (C) The first 200 ms from selected fractional inhibition traces are shown at the indicated concentrations of UCL2077. Arrows indicate intersection at the x-axis (zero inhibition).

• 

  Fig. 5. Characterization of UCL2077-mediated Kv7.1 block in the presence of KCNE1 (IKs). (A) Representative IKs current traces before and during application of a series of increasing UCL2077 concentrations (100 pM to 10 nM) and 300 nM HMR1556. HMR1556-insensitive currents were negligible and were not accounted for in the dose-response curve. (B) Concentration-response curve for IKs measured at the peak and tail currents, fit to the Hill equation (eq. 1) to obtain IC50 values of 520 pM and 590 pM, respectively (Table 2). Kv7.1 dose-response curves from Fig. 1 are overlain in green. Symbols represent mean ± SEM of 4–7 independent experiments. (C) IKs currents elicited by a series of 4-second depolarizing pulses from –80 to +80 mV in 10 mV steps, followed by 2 s to –40 mV (protocol below) in the absence (black) and in the presence (red) of 100 pM UCL2077. (D) Summarized mean ± SEM normalized G-V relationships in the absence (black circles) and in the presence (red squares) of 100 pM UCL2077, n = 5. Filled squares are normalized to their own Gmax, whereas empty squares are normalized to their respective control’s Gmax. (E) Before-after analysis of the parameters obtained by fitting data in…

• 

  Fig. 6. UCL2077 inhibition is state dependent. (A) Six-state model scheme for Kv7.1 and IKs channels: RC, resting closed; IC, intermediate closed; AC, activated closed; RO, resting open; IO, intermediate open; AO, activated open. (B) Left: cartoon of the S2 and S4 domains of WT Kv7.1, highlighting the gating and charge transfer center residues E160 (E1), F167 (F0), E170 (E2), R228 (R1), R231 (R2), Q234 (Q3), R237 (R4), and H240 (H5). Middle: cartoon of the E1R/R2E mutation, which functionally locks the channel in the IO state. Right: cartoon of the E1R/R4E mutation, which functionally locks the channel in the AO state. (C) Representative traces of WT, E1R/R2E and E1R/R4E currents in the absence (top) and in the presence (bottom) of KCNE1, before (control, gray) and in the presence of 300 pM UCL2077 (colored in black, blue, or orange). (D) Summary concentration-response curves in the absence (top) and in the presence (bottom) of KCNE1 fitted to the Hill equation (eq. 1). IC50 values were 70 pM (WT; black circles, n = 4–7), 0.61 pM (E1R/R2E; blue squares, n = 3–9), 21 nM (E1R/R4E; orange triangles, n = 3–6), 590 pM (WT + KCNE1; empty black circles, n = 4–7), 240 pM (E1R/R2E + KCNE1; empty…

• 

  Fig. 7. Kinetic model for UCL2077 inhibition. (A) Gating scheme for Kv7.1 channels. The possible states were RC, resting closed; IC, intermediate closed; AC, activated closed; RO, resting open; IO, intermediate open; AO, activated open; and AN, activated-nonconducting flicker state; IB: intermediate UCL2077-bound; AB: activated UCL2077-bound. L, M, and N are allosteric factors, C is the concentration of UCL2077, and α is the ratio between the dissociation constant (KD = kOFF/kON) at the IO state and at the AO state. Parameters are in Table 1. (B) Gating scheme for IKs channels. (C, D) Experimental and simulated current traces for Kv7.1 and IKs at the indicated concentrations of UCL2077. (E) Simulated concentration-response curves (“sim”, as lines) overlain with the experimental data (“exp”, symbols are mean ± SEM) from Fig. 6 in the absence (left) and in the presence of KCNE1 (right). For WT Kv7.1, experimental data are after subtraction of HMR1556-insensitive currents. For E1R/R2E and E1R/R4E (both in the absence and in the presence of KCNE1), data were normalized to set the minimum response (Ymin) to 0 while preserving the maximum response (Ymax) to account for endogenous currents. (F)…