NaV1.4 DI-S4 periodic paralysis mutation R222W enhances inactivation and promotes leak current to attenuate action potentials and depolarize muscle fibers

Abstract

Hypokalemic periodic paralysis is a skeletal muscle disease characterized by episodic weakness associated with low serum potassium. We compared clinical and biophysical effects of R222W, the first hNa_V1.4 domain I mutation linked to this disease. R222W patients exhibited a higher density of fibers with depolarized resting membrane potentials and produced action potentials that were attenuated compared to controls. Functional characterization of the R222W mutation in heterologous expression included the inactivation deficient IFM/QQQ background to isolate activation. R222W decreased sodium current and slowed activation without affecting probability. Consistent with the phenotype of muscle weakness, R222W shifted fast inactivation to hyperpolarized potentials, promoted more rapid entry, and slowed recovery. R222W increased the extent of slow inactivation and slowed its recovery. A two-compartment skeletal muscle fiber model revealed that defects in fast inactivation sufficiently explain action potential attenuation in patients. Molecular dynamics simulations showed that R222W disrupted electrostatic interactions within the gating pore, supporting the observation that R222W promotes omega current at hyperpolarized potentials. Sodium channel inactivation defects produced by R222W are the primary driver of skeletal muscle fiber action potential attenuation, while hyperpolarization-induced omega current produced by that mutation promotes muscle fiber depolarization.

Figure 1

R222W affects resting and action potentials in muscle fibers. The density of muscle fibers at various resting membrane potentials shows peaks at P₁ (hyperpolarized) and P₂ (depolarized) for control and R222W (A). P₁ peaks are −80.9 ± 0.1 mV (control) and −74.5 ± 0.4 mV (R222W), with P₂ peaks at −60.2 ± 0.4 mV (control) and −58.4 ± 0.4 mV (R222W). Action potentials in R222W (n = 10) fibers have a maximum amplitude of −8 ± 5 mV, significantly (P ≤ 0.05) less than that for control fibers (n = 6) at +11 ± 8 mV (B).

Figure 2

R222W slows activation and deactivation. Traces of sodium currents in response to step depolarizations from −90 to +60 mV for 20 ms are shown in (A), along with a schematic of the protocol for activation. I/V relationships are plotted for WT (n = 126) and R222W (n = 73) in (B) and compared in (C) utilizing the IFM/QQQ background (n = 17 and n = 21, respectively). G/V relationships fit to a Boltzmann function with the asymptote held at one are shown in (D, native background) and (E, IFM/QQQ background). R222W produces a +10.4 mV shift compared to WT in the native (n = 81 and n = 88, respectively), but not IFM/QQQ background (n = 63 and n = 48, respectively). Times for 10–90% rise to peak activation are shown in (F) for WT and mutant channels with the native (n = 48 and n = 73, respectively) or IFM/QQQ background (n = 43 and n = 45, respectively). Activation is slowed by R222W in the absence of IFM-mediated inactivation. Deactivation time constants are shown in (G) for WT/QQQ (n = 26) and R222W/QQQ (n = 29). R222W/QQQ slows deactivation as determined from tail currents (shown in inset) elicited by depolarizing channels to the empirical reversal potential for 30 ms, followed by 20 ms hyperpolarizing commands from −190 to −120 mV. Values are shown in Table 1.

Figure 3

R222W stabilizes the fast-inactivated state. Steady-state fast inactivation is shown in (A). Channels were conditioned for 300 ms from −140 mV to +20 mV, and channel availability tested with 20 ms, −20 mV test pulses (inset). R222W (n = 53) produces a −27 mV shift in the midpoint of the steady state fast inactivation curve compared to WT (n = 61). Time course of closed-state inactivation is shown in (B) for a variable-duration conditioning potential of −65 mV (protocol used shown in inset); R222W (n = 40) elicits more complete inactivation than WT (n = 28). Onset of fast inactivation from the open state is shown in (C). Compared to WT (n = 53), R222W (n = 50) accelerates open-state fast inactivation as determined from the decay of sodium currents in response to 20 ms step depolarizations from −40 mV to +20 mV. Recovery from fast inactivation is shown in (D). R222W (n = 5) slows recovery compared to WT (n = 17), measured by a double pulse protocol in which channels were inactivated for 100 ms and hyperpolarized (−100 mV shown) for increasing durations, followed by a 20 ms, −20 mV test pulse to test channel availability. Values are shown in Table 1.

Figure 4

R222W enhances slow inactivation. In all experiments, channels were recovered from fast inactivation at −120 mV (20 ms) prior to a test pulse at −20 mV to assess channel availability. Steady-state slow inactivation is shown in (A) from experiments in which channels were conditioned to voltages shown for 90 s prior to test. R222W (n = 34) significantly increases the completion and probability of slow inactivation compared to WT (n = 32). In (B), onset of slow inactivation was determined from loss of channel availability following a variable duration (0 to 360 s) depolarization to 0 mV. R222W (n = 17) did not significantly affect onset of slow inactivation compared to WT (n = 22). For recovery (C), a 0 mV, 90 s command to promote slow inactivation was followed by variable duration (0 to 360 s) hyperpolarization at −120 mV and test. R222W (n = 27) slows recovery from slow inactivation compared to WT (n = 25). Values are shown in Table 1.

Figure 5

R222W promotes a cationic leak current. Current traces from TTX-blocked channels in response to step commands to voltages ranging from −140 mV to +40 mV are shown in (A) for WT and R222W, with 115 mM Na⁺ in the external bath solution. Ohmic current from −30 mV to +20 mV was subtracted from raw current values and normalized against anodic gating current recorded in response to +40 mV step depolarization. Plot of the normalized leak current as a function of gating charge at +40 mV is shown in (B, n = 8 to 12). With 115 mM Na⁺ external, R222W channels show an inwardly directed current at hyperpolarized voltages, not observed for WT channels. Ohmic responses are observed for R222W or WT when NMDG is used as the external cation.

Figure 6

R222W action potential height is attenuated in simulation. A schematic of the main parameters used in the action potential model is shown in (A). The sarcolemma and t-tubule were both modeled, as well as potassium accumulation in the t-tubule. Barrier models of inactivation kinetics are shown in (B) for WT and R222W. Time constants were obtained at voltages from −190 mV to −70 mV (recovery), −60 mV to −30 mV (closed-state fast inactivation) and −25 mV to 20 mV (open-state fast inactivation). The fractional barrier distance is 0.50 for WT and 0.35 for R222W. The midpoint of the barrier is −55.1 mV for WT and −80.6 mV for R222W. These kinetic parameters were used to simulate wild type and R222W (attenuated) action potentials shown in (C), with peak height of the R222W action potential at −9.6 mV and that for WT at +38.7 mV. A phase diagram of the rate of voltage change is shown in (D). The peak rate of rise (0.8 V/ms) is 50% slower for R222W action potentials compared to WT (1.6 V/ms).

Figure 7

W222 distrupts interactions within the gating charge transfer center. As R222 (black) or W222 (red) enters the gating pore (A top view, C side view), the native arginine appears to interact with Y168 (yellow) while the subsituted tryptophan side chain moves away from Y168. Interaction and distance measurements are recorded in Supplementary Table S3. As R222 or W222 moves down through the gating pore (B top view, D side view), R222 interacts with E171 (green) across the gating pore, whereas W222 does not cross the gating pore to interact with E171. The gating charge transfer center asparate (D197, blue) does not interact with R222 or W222 in simulation.