|
Figure 1. . Location of HERG cysteine mutants. (A) Putative transmembrane topology of a single subunit of the HERG potassium channel. Mutations are located at the end of the S3-S4 linker nearest to the S4 segment, shown by the cartoon dye molecule at the top. (B) Amino acid sequence of the S4 region for HERG and the Shaker potassium channel. Positively charged residues are shown in bold. The underlined residues in the HERG sequence were individually mutated to cysteines and labeled with a fluorescent probe for further study. The underlined residues in the Shaker sequence have been studied previously using this fluorescence technique (Mannuzzu et al., 1996; Cha and Bezanilla, 1997, 1998; Loots and Isacoff, 1998).
|
|
Figure 2. . Fluorescence signals and current simultaneously recorded in response to a single voltage step from oocytes expressing E518C, E519C, or L520C and treated with TMRM. In each recording, the membrane was held at −90 mV, depolarized to 30 mV for 1 s, and then stepped to −110 mV.
|
|
Figure 3. . L520C-TMRM fluorescence signals during channel activation and deactivation. (top left) Currents resulting from an envelope of tails protocol (voltage records shown below); oocytes were depolarized from −90 to +30 mV for varying durations, and then hyperpolarized to −110 mV. The time course of the peak inward tails reveals the time course of activation during the depolarizing step. (top right) Deactivating tail current measured at −110 mV after a 2-s depolarization to +30 mV. (bottom) The kinetics of the fluorescence are compared with the kinetics of activation and deactivation. From a holding potential of −90 mV, the membrane was depolarized to +30 mV for 2 s, and then stepped to −110 mV. The fluorescence trace is unaveraged. The solid gray line at left plots the kinetics of activation as measured by currents resulting from the envelope of tails protocol (top left); the peak tail current upon hyperpolarization to −110 mV is plotted as a function of time (τ = 84 ms). The thin solid line at right shows the deactivating tail current from the top right panel, scaled to the magnitude of the fluorescence change. This tail current was best fit with two exponentials; the faster component (τ = 83 ms) comprises 83% of the total magnitude, and the slower component (τ = 334 ms) makes up the remaining 17%. The solid gray line at right plots only the faster of the two components, scaled to match the fluorescence. For comparison, τF on depolarization = 67 ms and τF on hyperpolarization = 88 ms.
|
|
Figure 5. . An NH2-terminal deletion produces similar effects on deactivation and the fluorescence signal of L520C-TMRM. Current traces (top) and fluorescence signals (middle) recorded from oocytes expressing L520C-TMRM and L520C:Δ2–137-TMRM. Channel deactivation is monitored after a prolonged step to +30 mV (2 s for L520C-TMRM; 1 s for L520C:Δ2–137-TMRM).
|
|
Figure 6. . Kinetics of the fast fluorescence changes for E519C-TMRM and E518C-TMRM parallel the kinetics of channel gating during recovery from and re-onset of inactivation. (top) Fluorescence changes and current recorded from oocytes expressing 519C-TMRM (left) or 518C-TMRM (right) during a three-step voltage protocol. The 519C-TMRM trace is the average of 10 recordings acquired at a sampling rate of 1 kHz; the 518C-TMRM trace is the average of 20 recordings acquired at a sampling rate of 10 kHz. (bottom) The fluorescence traces shown on an expanded time scale. Solid black lines show the current trace scaled to correspond to the magnitude of the fluorescence. Each segment of the pulse protocol was scaled independently to allow for a comparison of the kinetics.
|
|
Figure 7. . The voltage dependence of steady-state inactivation compared with the voltage dependence of the fast fluorescence change measured for 519C-TMRM and 518C-TMRM. (A) Protocol used to determine the voltage dependence of inactivation. Oocytes injected with wild-type HERG were depolarized to 30 mV for 1 s, briefly hyperpolarized to voltages ranging from −130 to 10 mV for 30 ms and stepped to 0 mV for 300 ms. (B) Open circles represent the current measured immediately after the second depolarizing step to 0 mV (A, arrow) as a function of voltage. Currents decrease at negative potentials due to deactivation during the 30 ms pulse. The current after correction for deactivation (see results) is shown by the closed circles. (C) Fluorescence signals reported by E519C-TMRM in response to a tail current protocol; the membrane was depolarized to 30 mV for 1 s from a holding potential of −90 mV, then stepped to potentials varying from −130 to −10 mV in 40-mV steps. Each trace is the average of 10 recordings acquired at a sampling rate of 1 kHz. (D) Voltage dependence for the fast fluorescence change (ΔF) on hyperpolarization, normalized to the maximum observed ΔF (closed circles). Fluorescence signals measured during the second voltage step were fit with either a single or double exponential to obtain the magnitude of the initial fast change in fluorescence upon hyperpolarization. The average steady-state inactivation-voltage curve for E519C-TMRM is also shown (open circles, n = 3) normalized to a maximum value of 1. Steady-state inactivation-voltage curves were constructed as described for B. (E) Fluorescence signals reported by E518C-TMRM in response to a tail current protocol; the membrane was depolarized to 90 mV for 200 ms from a holding potential of −90 mV, and then stepped to potentials varying from −130 to −10 mV in 40-mV steps. Each trace is the average of 20 recordings acquired at a sampling rate of 10 kHz. (F) Voltage dependence for the fast fluorescence signal of E518C-TMRM, together with the steady-state inactivation curve (both determined as for D). Throughout, the error bars indicate SEM.
|
|
Figure 8. . Modulation of inactivation does not affect fast fluorescence changes. (A) External TEA slows inactivation but has no effect on the fluorescence signal of E518C-TMRM. After a short hyperpolarizing step to produce recovery, a second depolarizing step elicits rapid reinactivation; both ionic current and fluorescence are monitored before and after application of 50 mM external TEA. For comparison, the fluorescence traces for each condition were shifted to the same level at the end of the first depolarizing pulse; the current traces are simply overlaid. (B) Fluorescence and current traces measured from oocytes expressing E519C:G628C:G631C-TMRM in response to a three-step protocol. Inactivation is eliminated, but the fluorescence signal is similar to E519C-TMRM in Fig. 6 A. The external solution contained 2 mM KCl and 98 mM NaCl.
|
|
Figure 9. . Properties of fast fluorescence change after depolarization recorded from E519C-TMRM. (A) Fluorescence and current traces recorded from E519C-TMRM during a 20 ms voltage pulse to 30 mV. The dashed line indicates approximately the end of the capacity transient observed in the current trace upon depolarization. (B) Voltage dependence of the fast fluorescence change observed upon depolarization for E519C-TMRM. Fluorescence changes recorded during a series of voltage steps 150 ms in duration ranging from −130 mV to 110 mV. (C) Plot of the fluorescence 5 ms after depolarization (B, arrow); each point is an average of 10 data points from the fluorescence trace at each voltage. The points were fit with a Boltzmann function (Vmid = 15.6 mV, and zδ = 0.75 e0).
|
|
Figure 10. . Fluorescence traces recorded from oocytes expressing 519C labeled with TMRM or Oregon green (OG). Fluorescence changes were measured in response to a 1-s voltage step to 30 mV followed by a hyperpolarizing step to −90 mV.
|
|
Figure 11. . Two four-state models to describe fluorescence changes observed in HERG. Model A has two voltage dependent transitions, one fast (q → Q) and one slow (r → R). The shaded boxes show relative magnitudes of the fluorescence coefficients used to describe fluorescence changes observed for each mutant. Model B has three sequential voltage dependent transitions.
|
|
Figure 12. . Simulations of fluorescence and current data using Model A and Model B. (A, top) Simulated fluorescence and current traces (black lines) from Model A are overlaid on fluorescence and current traces recorded during a three-step protocol (gray lines) from oocytes expressing E518C, E519C, or L520C and labeled with TMRM. (bottom) Simulation of the occupancy of each state in Model A as a function of time during a three-step protocol. Simulations of fluorescence changes, current and state occupancies used the parameters shown in Table II. The kinetic parameters from fitting E519C-TMRM data were used for all simulations; only the fluorescence coefficients were varied to simulate E518C-TMRM and L520C-TMRM fluorescence data (Table II). (B) Description of Model A showing a likely relationship between states of the voltage sensor and gating states of the pore (i.e., closed, open, and inactivated). (C, top) Simulated current trace (black line) from Model B is overlaid on the recorded current trace (gray line). (bottom) Simulation of the occupancy of each state in Model B as a function of time during a three-step protocol. Simulations of current and state occupancies used the parameters shown in Table III. Simulations of fluorescence data using Model B (Table III) are nearly identical to those shown in A for Model A.
|
|
Figure 4. . Voltage dependence of the current and the slow fluorescence change reported by L520C-TMRM. (A) Fluorescence changes measured during 2-s voltage steps ranging from −90 to 70 mV. The holding potential is −90 mV; the membrane is hyperpolarized to −110 mV after depolarization. (B) Voltage dependence of the fluorescence at the end of a 2-s voltage pulse (closed circles, indicated by the arrow in A). The F-V curve shown is an average of six normalized F-V curves. Each point in an individual F-V curve represents the average of the last 20 data points during depolarization. The average, normalized g-V relationship measured during the same experiments is plotted for comparison (open circles, n = 6). Each g-V relationship was constructed as described for Table I. Error bars denote the standard error. The curves shown are Boltzmann fits of each dataset (for F-V, Vmid = −30 mV, and zδ = 2.3 e0; for g-V, Vmid = −28 mV, and zδ = 2.2 e0).
|
