Two-pore domain potassium channels enable action potential generation in the absence of voltage-gated potassium channels

Abstract

In this study, we explored the possibility that two-pore domain potassium (K2P) channels are sufficient to support action potential (AP) generation in the absence of conventional voltage-gated potassium (KV) channels. Hodgkin-Huxley parameters were used to mimic the presence of voltage-gated sodium (NaV) channels in HEK-293 cells. Recombinant expression of either TREK-1 or TASK-3 channels was then used to generate a hyperpolarised resting membrane potential (RMP) leading to the characteristic non-linear current-voltage relationship expected of a K2P-mediated conductance. During conductance simulation experiments, both TASK-3 and TREK-1 channels were able to repolarise the membrane once AP threshold was reached, and at physiologically relevant current densities, this K2P-mediated conductance supported sustained AP firing. Moreover, the magnitude of the conductance correlated with the speed of the AP rise in a manner predicted from our computational studies. We discuss the physiological impact of axonal K2P channels and speculate on the possible clinical relevance of K2P channel modulation when considering the actions of general and local anaesthetics.

Fig. 1

Simulations to compare the influence of linear versus non-linear leak conductance relationships on AP repolarisation. a Properties of the voltage-gated sodium current (I _VGSC) used for simulation studies and dynamic-clamp experiments. The voltage and time-dependent conductance changes are plotted for simulations involving voltage steps from −40 to 0 mV in 5-mV increments. The voltage-dependence of ^τ _decay for I _VGSC is also plotted. b The current–voltage relationships generated by the linear (IΩ_-leak) and the non-linear potassium leak conductance (I _GHK-leak) that generated RMPs of −60 mV (red line), −70 mV (grey line) and −80 mV (black line) are shown. The value of the linear leak conductance was 3.2 nS for −60 mV, 4.9 nS for −70 mV and 8.8 nS for −80 mV. The values of the GHK leak conductance at the minimum membrane potential (i.e. close to the RMP) was 0.65 nS at −60 mV, 0.85 nS at −70 mV and 1.8 nS at −80 mV c Superimposed synaptic currents (I _EPSC) that were generated by double exponential functions with peak conductance values ranging from 0.5 to 10 nS. d The membrane voltage that resulted from summation of the currents generated by the conductance changes shown in a, b and c as well as the capacitive current generated by the voltage change. The superimposed voltage records were generated using the I _GHK-leak that resulted in an RMP of −70 mV. e Representative examples of voltage simulations comparing the linear (IΩ_-leak) and the non-linear (I _GHK-leak) leak conductance. APs are generated at all values of I _GHK-leak that fully repolarise back to the RMP. However, the lowest conductance value of IΩ_-leak that resulted in an RMP of −60 mV was insufficient to fully repolarise the RMP following stimulation by a 2-nS EPSC as shown by the trace indicated with a red circle. F, Plots of AP repolarisation for all simulations indicating the linear (IΩ_-leak) and non-linear (I _GHK-leak) leak that was capable of repolarisation the AP. The only condition where this was not possible was at an IΩ_-leak resulting in an RMP of −60 mV. In this scenario, AP repolarisation was incomplete after the first EPSC of 2 nS, and no subsequent APs could be generated. This was in stark contrast to the I _GHK-leak that resulted in an RMP of −60 mV that was always sufficient to repolarise the AP

Fig. 2

Removal of the endogenous voltage-gated potassium channels from the expression system. a Characterisation of conductances in HEK-293 cells illustrating the activation of an endogenous K _V near 0 mV. The insert shows a series of current records elicited by voltage steps from −100 to +40 mV. The current–voltage relationship of this K _V current has been plotted for recordings made in the presence (grey triangles) and absence (black symbols) of 1 mM TEA. b A voltage ramp from −20 to −150 mV was used to study the non-linear leak conductance present following TASK-3 and TREK-1 transfection. The current–voltage relationship of the resulting potassium leak conductance (G _K) recorded from a TREK-1-transfected cell is shown in the presence (grey trace) and absence (black trace) of 10 mM TEA. For comparison, the current–voltage relationship for another un-transfected cell is also plotted to illustrate the small endogenous linear leak that is also present in these cells. Note the lack of any endogenous K _V activation over the voltage ranges examined in the absence of TEA. c Inhibition curves were constructed from data obtained following TEA block of the endogenous K_V and the TASK-3-mediated K_2P conductance. The data were well-described with a modified Hill equation (dashed lines) to give an estimate of the IC₅₀ for this drug–receptor interaction. From these fits, it is clear that the endogenous K _V is sensitive to TEA with an IC₅₀ of 0.7 mM, whereas the TASK-3 conductance is relatively insensitive to this blocker with an extrapolated IC₅₀ of 300 mM. d In order to examine the relationship between G _K and the RMP, results were pooled from all cells transfected with either TASK-3 (filled circles) or TREK-1 (filled triangles), and for un-transfected cells (open circles). The data set was well described by a single exponential function (dashed line)

Fig. 3

Two-pore domain potassium channel expression is sufficient to repolarise the AP. a, b and c Single AP generated in response to a brief 1-ms current injection in three individual cells expressing TREK-1 channels. The grey dashed line illustrates the 0 mV level. The magnitude of G _K was smallest in a (0.4 nS/pF) and largest in c (3.1 nS/pF). The magnitude of G _K correlates with the RMP of the HEK cell, the current required to reach AP threshold, and the duration of the AP. The sub-threshold voltage response is also shown for each cell (grey line) to further demonstrate the faster membrane time constant that is apparent at larger values of G _K

Fig. 4

K_2P channel expression alters the current required to reach AP threshold. a Continuous voltage recording (black trace) obtained during a conductance simulation experiment from a HEK cell expressing TASK-3 channels. A series of long-duration 300-ms depolarising current pulses (grey trace) were applied at increasing current intensities. The relationship between the magnitude of the injected current and the AP frequency for all cells examined in this way is plotted in panel b. Analysis of a single cells AP frequency response to injected current is shown in panel c. This relationship is well described by a simple Boltzmann-type function (dashed line). From this analysis, we can estimate the current required to reach AP threshold (i _AP), the slope of the relationship (k) and the maximum AP frequency (fAP_max). d The results of fitting linear regressions to the relationship between G _K and the three variables (i _AP, k, and fAP_max) extracted from the data shown in b. It is clear that the only correlation is between the current required to reach AP threshold (i _AP) and the magnitude of the potassium leak (G _K)

Fig. 5

Relationship between the magnitude of the K_2P-mediated conductance and the AP rise and decay. a Two threshold APs from separate cells are superimposed to illustrate the difference in AP rise and decay. The first AP (i) comes from a cell with a G _K of 0.1 nS/pF whereas the second AP (ii) comes from a cell with a G _K of 0.5 nS/pF. b Phase plane plots of the two APs shown in a. The maximum AP rise and decay was calculated from these plots. c Plot of the maximum AP rise (open circles) and decay (open squares) for all threshold APs elicited during conductance simulation procedure in relation to the G _K for each cell examined

Fig. 6

Actions of halothane on K_2P channels and the consequence of this enhancement for AP properties. a Results from a voltage-clamp experiment illustrating the current–voltage relationship for a TASK-3-expressing cell during a standard ramp protocol in the presence (grey trace) and absence (black trace) of 2 % halothane. The insert shows the time course of the halothane enhancement for this cell calculated at a command potential of −20 mV. b A single threshold AP elicited from a different TASK-3-expressing cell in the presence (grey trace) and absence (black trace) of 2 % halothane. The dashed line illustrates the 0 mV level. c Phase plane plots for the APs shown in b illustrating the modest speeding of the AP rise and decay in the presence (grey trace) of 2 % halothane