Fast and slow gating are inherent properties of the pore module of the K+ channel Kcv

Abstract

Kcv from the chlorella virus PBCV-1 is a viral protein that forms a tetrameric, functional K+ channel in heterologous systems. Kcv can serve as a model system to study and manipulate basic properties of the K+ channel pore because its minimalistic structure (94 amino acids) produces basic features of ion channels, such as selectivity, gating, and sensitivity to blockers. We present a characterization of Kcv properties at the single-channel level. In symmetric 100 mM K+, single-channel conductance is 114+/-11 pS. Two different voltage-dependent mechanisms are responsible for the gating of Kcv. "Fast" gating, analyzed by beta distributions, is responsible for the negative slope conductance in the single-channel current-voltage curve at extreme potentials, like in MaxiK potassium channels, and can be explained by depletion-aggravated instability of the filter region. The presence of a "slow" gating is revealed by the very low (in the order of 1-4%) mean open probability that is voltage dependent and underlies the time-dependent component of the macroscopic current.

Figure 1.

Kcv macro-currents recorded in Xenopus oocytes. (A) Currents recorded by two-electrode (TE) voltage clamp from a holding potential of −20 mV, followed by steps of 20 mV ranging between 100 and −200 mV, and returning at −80 mV. Bath solution: 50 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES, pH 7.5. Osmolarity was adjusted to 215 mOsm with mannitol. (B) I-V curve of the instantaneous current shown in A. (C) Currents recorded from macropatch in inside-out configuration. Holding potential of 0 mV, with steps +50 and −150 mV, returning at −80 mV. Pipette solution: 100 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 10 mM HEPES/pH 7.3. Bath solution: 100 mM KCl, 1 mM MgCl₂, 1 mM EGTA, and 10 mM HEPES/pH 7.3. (D) Putative structure of Kcv showing in orange and blue two of the four identical subunits of the tetrameric channel. The figure illustrates a snapshot of a molecular dynamics simulation of the Kcv model (Tayefeh et al. 2009).

Figure 2.

Single-channel recordings obtained from Kcv expressed in oocytes. (Left) Time series measured in the cell-attached mode at different potentials as indicated. Pipette solution contained (in mM): 100 KCl, 1.8 CaCl₂, 1 MgCl₂, and 10 HEPES, pH 7.3. (Right) Single-channel I-V curve of the experiment on the left. (Inset) Comparison of experiments performed in cell-attached (black symbols) and inside-out mode (red symbols) in the following bath solution (in mM): 100 KCl and 1 HEPES, pH 7.3. Mean values ± SD, unitary channel openings from two independent experiments; n ≥ 300 apart from the value at −140 mV in inside-out, in which n = 8.

Figure 3.

Voltage-dependent behavior of the apparent conductance in Kcv. Comparison of bursts obtained at ±100, ±120, and ±140 mV (top traces, positive potentials; bottom traces, negative potentials) from two different experiments. (Left) In this experiment, the decrease in conductance and the increase in noise are more marked at positive than at negative voltages. (Right) Opposite dependence on voltage of current reduction found in a different experiment. In this experiment, the decrease in apparent conductance and the increase in open-channel noise are more evident at negative than at positive potentials.

Figure 4.

Analysis performed by β distributions. (A) Point amplitude distributions from a time series recorded at +80 mV are shown (black line, close level; red line, open level). I_app = 5.17 pA (vertical line) is obtained from the maximum of the open-point amplitude histogram (red line). The green line is the open-point amplitude distribution obtained from the time series simulated by an O-C model and fitted to the measured distribution (red line). This procedure gives the best pairs of rate constants (k_OC = 58,400 s⁻¹ and k_CO = 42,200 s⁻¹ for the example shown). For each simulation, the current I was used as a fixed parameter and was changed stepwise in subsequent fits. The current value that gives the minimum value in the error sum (see B) is defined as I_true (vertical line). (B) Dependence of the error sum of the fit on the assumed value of I. I_true = 12.5 pA (vertical line) is the value of I that gives the minimum value in the error sum in this example.

Figure 5.

Voltage dependence of fast gating. (A) I-V curves of I_app (filled black circles) obtained from the time series directly by averaging over the bursts, and I_true (open red circles) evaluated from the β distribution fits. (B) Voltage dependence of k_OC (filled black squares) and k_CO (open red squares). (C) The gating factor R obtained from Kcv at 100 mM KCl and from MaxiK obtained at 50, 150, and 400 mM KCl. For Kcv, R_I and R_K are both displayed (open squares and filled squares, respectively); for MaxiK, only R_I is displayed. The smooth lines approximating the data points present the exponential fits by means of Eq. 3. Parameters for Kcv: R₀ = 1.2 ± 0.1 mV, V_G = 34 ± 3 mV, and R_K = 11 ± 3. Numbers of individual experiments are given for Kcv.

Figure 6.

Effect of increasing [K⁺]_in on Kcv outward flickering. (A) Open-channel I-V relationship of Kcv recorded from the same patch in cell-attached configuration, [K⁺]_in = ∼100 mM (black symbols and line), and in inside-out configuration, [K⁺]_in = 300 mM (red symbols and line). Pipette solution: 100 mM KCl, 1.8 mM CaCl₂, and 1 mM MgCl₂. Bath solution: 300 mM KCl, 1 mM EGTA, and 1 mM MgCl₂. (B) Voltage dependence of the calculated single-channel conductance (black, [K⁺]_in ∼100 mM; red, [K⁺]_in 300 mM). (C) Comparison of single-channel fluctuations recorded in [K⁺]_in ∼100 mM (left) and 300 mM (right) at about the same driving forces (∼+80 mV from current reversal voltage). (Bottom) Enlargements of top traces.

Figure 7.

Effect of increasing [K⁺]_out on Kcv inward flickering. (A) Open-channel I-V relationships of Kcv recorded in cell-attached configuration in [K⁺]_out = 100 mM (black symbols and line) and [K⁺]_out = 300 mM (red symbols and line). Pipette solution: 100 or 300 mM KCl, 1.8 mM CaCl₂, and 1 mM MgCl₂. Bath solution: 100 mM KCl, 1 mM EGTA, and 1 mM MgCl₂. (B) Voltage dependence of the calculated single-channel (black, [K⁺]_out ∼100 mM; red, [K⁺]_out 300 mM). (C) Comparison of single-channel fluctuations recorded in [K⁺]_out 100 mM (left) and 300 mM (right) at almost the same driving forces (∼−60 mV from current reversal voltage). (Bottom) Enlargements of top traces.

Figure 8.

Open probability in Kcv. (A) Time-dependent open probability in a 60-min-long multichannel record. 1-min extract is shown from three periods of different channel activity, cross-referenced by colored arrows in B. (B) NP_open calculated over 5-min-long periods increases with time (the nominal open probability is T_o/T_o+T_C, independently from the number of open levels). (C) Steady-state open probability: voltage dependence obtained from records of 1 min (n = 7 patches). (D) Consecutive traces at +80 and −100 mV and ensemble average of 100 similar traces (boxed). Activation kinetic is fitted (red line) by a single exponential (τ = 17 ms).