Anticonvulsants modify inactivation but not activation processes of sodium channels in Myxicola axons

The effects of pronase and the anticonvulsant drugs diphenylhydantoin, bepridil, and sodium valproate on fast and slow Na+ inactivation were examined in cut-open Myxicola giant axons with loose patch-clamp electrodes applied to the internal surface. Pronase completely eliminated fast Na+ inactivation without affecting the kinetics of Na+ activation or the maximum Na+ conductance. The time and voltage dependences of slow inactivation following pronase treatment were identical to those measured before enzyme application in the same axons. All three anticonvulsants slowed the time course of recovery from fast Na+ inactivation in untreated axons, and shifted the steady-state fast inactivation curve in the hyperpolarizing direction along the voltage axis. Anticonvulsants enhanced steady-state slow inactivation and retarded recovery from slow inactivation in both untreated and pronase-treated axons. Although some quantitative differences were seen, the order of potency of the anticonvulsants on slow Na+ inactivation was the same as that for recovery from fast inactivation.     

Solvent substitution as a probe of channel gating in Myxicola. Differential effects of D2O on some components of membrane conductance.

Careful examination of effects of solvent substitution on excitable membranes offers the theoretical possibility of identifying those aspects of the gating and translocation processes which are associated with significant changes in solvent order. Such information can then be used to develop or modify moire detailed models. We have examined the effects of heavy water substitution in Cs+-and K+-dialyzed Myxicola giant axons. At temperatures of 4-6 degrees C, the rates of Na+, K+, and Na+ inactivation during a maintained depolarization were all showed by approximately 50% in the presence of D2O. In contrast, the effects of solvent substitution on the time-course of prepulse inactivation and reactivation were much larger, with slowing averaging 160%. Studies at higher temperatures yielded Q10's for Na+ activation and K+ activation which were essentially comparable (0.72) and slightly but significantly smaller than that for inactivation during a maintained depolarization (0.84). In contrast, the Q10 for the D2O effect on prepulse inactivation was approximately 0.48. Heavy water substitution decrease Gk to a significantly greater extent than G(Na), while the decrease in the conductance of the Na+ channel caused by D2O was independent of whether the current-carrying species was Na+ or Li+. Sodium channel selectivity to the alkali metal cations and NH4+ was not changed by D2O substitution.     

Sodium channel activation in the squid giant axon. Steady state properties

Treatment of giant axons from the squid, Loligo pealei, with pronase removes Na channel inactivation. It was found that the peak Na current is increased, but the activation kinetics are not significantly altered, by pronase. Measurements of the fraction of open channels as a function of voltage (F-V) showed an e-folding at 7 mV and a center point near -15 mV. The rate of e-folding implies that a minimum of 4 e-/channel must cross the membrane field to open the channel. The charge vs. voltage (Q-V) curve measured in a pronase-treated axon is not significantly different from that measured when inactivation is intact: approximately 1,850 e-/micron2 were measured over the voltage range -150 to 50 mV, and the center point was near -30 mV. Normalizing these two curves (F-V and Q-V) and plotting them together reveals that they cross when inactivation is intact but saturate together when inactivation is removed. This illustrates the error one makes when measuring peak conductance with intact inactivation and interpreting that to be the fraction of open channels. A model is described that was used to interpret these results. In the model, we propose that inactivation must be slightly voltage dependent and that an interaction occurs between the inactivating particle and the gating charge. A linear sequence of seven states (a single open state with six closed states) is sufficient to describe the data presented here for Na channel activation in pronase-treated axons.     

Single-channel recordings from two types of amiloride-sensitive epithelial Na+ channels

We report here the first evidence in intact epithelial cells of unit conductance events from amiloride-sensitive Na+ channels. The events were observed when patch-clamp recordings were made from the apical surface of cultured epithelial kidney cells (A6). Two types of channels were observed: one with a high selectivity to Na+ and one with relatively low selectivity. The characteristics of the low-selectivity channel are as follows: single-channel conductance ranged between 7 and 10 pS (mean = 8.4 +/- 1.3), the current-voltage (I-V) relationship displayed little if any nonlinearity over a range of +/- 80 mV (with respect to the patch pipette) and the channel Na+/K+ selectivity was approximately 3-4:1. Amiloride, a cationic blocker of the channel, reduced channel mean open time and increased channel mean closed times as the voltage of the cell interior was made more negative. Amiloride induced channel flickering at increased negative potentials (intracellular potential with respect to the patch) but did not alter the single-channel conductance or the I-V relationship from that observed in control patches. The characteristics of the high-selectivity channel are: a single-channel conductance of 1-3 pS (mean = 2.8 +/- 1.2), the current-voltage relationship is markedly nonlinear with a Na+/K+ selectivity greater than 20:1. The mean open and closed times for the two types of channels are quite different, the high-selectivity channel being open only about 10% of the time while the low-selectivity channel is open about 30% of the time.     

Saturation transfer electron paramagnetic resonance detection of sickle hemoglobin aggregation during deoxygenation.

In Myxicola axons, substitution of tetramethylammonium (TMA+) for Cs+ alters intramembrane charge movements (gating currents). Although the total charge moved during and following a depolarizing step remains constant, with TMA+ the ON response has additional slower component(s), and the OFF response is retarded. Concommitantly, TMA+ produces the same voltage-dependent block of Na+ inactivation in Myxicola as has been observed in other preparations. At large positive potentials as many as 70% of the Na+ channels fail to inactivate in the steady state. In addition, TMA+ slows Na+ activation, retards the inactivation of those Na+ channels that remain able to inactivate, and decreases the maximum Na+ conductance. The steady-state Na+ conductance induced by internal TMA+ or Na+ is consistent with a scheme in which these internal cations simply modify Na+ channels in an all-or-none fashion so that a fraction become incapable of inactivating.     

Modulation of single cardiac sodium channels by DPI 201-106

Single sodium channel currents were analysed in cell attached patches from single ventricular cells of guinea pig hearts in the presence of a novel cardiotonic compound DPI 201-106. The mean single channel conductance of DPI-treated Na channels was not changed by DPI (20.8 +/- 4 pS, control, 3 patches; 21.3 +/- 1 pS with DPI, 5 mumol/1,3 patches). DPI voltage-dependently prolongs the cardiac sodium channel openings by removal of inactivation at potentials positive to -40 mV. At potentials negative to -40 mV a clustering of short openings at the very beginning of the depolarizing voltage steps can be observed causing a transient time course of the averaged currents. Long openings induced an extremely slow inactivation. Short openings, long openings and nulls appeared in groups referring to a modal gating behaviour of DPI-treated sodium channels. DPI-modified Na channels showed a monotonously prolonged mean open time with increased depolarizing voltage steps, e.g. the open state probability within a sweep was increased. However, the number of non-empty sweeps was decreased with the magnitude of the depolarizing steps, e.g. the probability of the channel being open as calculated from the averaged currents was voltage-dependently decreased by DPI (50% decrease at -50.7 +/- 9 9 mV, 3 patches). Short and long openings of DPI-modified channels could be separated by variation of the holding potential. The occurrence of long Na channel openings was much more suppressed by reducing the holding potential (half maximum inactivation at -112 +/- 8 mV, 4 patches) than that of short openings (half maximum inactivation at -88 +/- 8 mV, 4 patches). Otherwise, short living openings completely disappeared at potentials positive to -40 mV where the occurrence of long openings was favoured. The differential voltage dependence of blocking and activating effects of DPI on cardiac Na channels as well as the differential voltage dependence of the appearance of short and long openings refers to a modal gating behaviour of cardiac Na channels.     

Modifications of sodium channel gating in Myxicola giant axons by deuterium oxide, temperature, and internal cations.

In dialyzed Myxicola axons substitution of heavy water (D2O) externally and internally slows both sodium and potassium kinetics and decreases the maximum conductances. Furthermore, this effect is strongly temperature dependent, the magnitude of the slowing produced by D2O substitution decreasing with increasing temperature over the range 3-14 degrees C with a Q10 of approximately 0.71. The relatively small magnitude of the D2O effect, combined with its strong temperature dependence, suggests that the rate limiting process producing a conducting channel involves appreciable local changes in solvent structure. Maximum conductances in the presence of D2O were decreased by approximately 30%, while the voltage dependences of both gNa and gK were not appreciably changed. In contrast to the effects of heavy water substitution on the ionic currents, membrane asymmetry currents were not altered by D2O, suggesting that gating charge movement may preceed by several steps the final transformation of the Na+ channel to a conducting state. In Myxicola axons the effect of temperature alone on asymmetry current kinetics can be well described via a simple temporal expansion equivalent to a Q10 of 2.2, which is somewhat less than the Q10 of GNa activation. The integral of membrane asymmetry current, representing maximum charge movement, is however not appreciably altered by temperature.     

Inhibition of sodium currents by local anesthetics in chloramine-T- treated squid axons. The role of channel activation

In order to test the requirement of Na channel inactivation for the action of local anesthetics, we investigated the inhibitory effects of quaternary and tertiary amine anesthetics on normally inactivating and noninactivating Na currents in squid axons under voltage clamp. Either the enzymatic mixture pronase, or chloramine-T (CT), a noncleaving, oxidizing reagent, was used to abolish Na channel inactivation. We found that both the local anesthetics QX-314 and etidocaine, when perfused internally at 1 mM, elicited a "tonic" (resting) block of Na currents, a "time-dependent" block that increased during single depolarizations, and a "use-dependent" (phasic) block that accumulated as a result of repetitive depolarizations. All three effects occurred in both control and CT-treated axons. As in previous reports, little time-dependent or phasic block by QX-314 appeared in pronase-treated axons, although tonic block remained. Time-dependent block was greatest and fastest at large depolarizations (Em greater than +60 mV) for both the control and CT-treated axons. The recovery kinetics from phasic block were the same in control and CT-modified axons. The voltage dependence of the steady state phasic block in CT-treated axons differed from that in the controls; an 8-10% reduction of the maximum phasic block and a steepening and shift of the voltage dependence in the hyperpolarizing direction resulted from CT treatment. The results show that these anesthetics can bind rapidly to open Na channels in a voltage-dependent manner, with no requirement for fast inactivation. We propose that the rapid phasic blocking reactions in nerve are consequences primarily of channel activation, mediated by binding of anesthetics to open channels, and that the voltage dependence of phasic block arises directly from that of channel activation.     

Veratridine modifies open sodium channels

The state dependence of Na channel modification by the alkaloid neurotoxin veratridine was investigated with single-channel and whole-cell voltage-clamp recording in neuroblastoma cells. Several tests of whole-cell Na current behavior in the presence of veratridine supported the hypothesis that Na channels must be open in order to undergo modification by the neurotoxin. Modification was use dependent and required depolarizing pulses, the voltage dependence of production of modified channels was similar to that of normal current activation, and prepulses that caused inactivation of normal current had a parallel effect on the generation of modified current. This hypothesis was then examined directly at the single-channel level. Modified channel openings were easily distinguished from normal openings by their smaller current amplitude and longer burst times. The modification event was often seen as a sudden, dramatic reduction of current through an open Na channel and produced a somewhat flickery channel event having a mean lifetime of 1.6 s at an estimated absolute membrane potential of -45 mV (23 degrees C). The modified channel had a slope conductance of 4 pS, which was 20-25% the size of the slope conductance of normal channels with the 300 mM NaCl pipette solution used. Most modified channel openings were initiated by depolarizing pulses, began within the first 10 ms of the depolarizing step, and were closely associated with the prior opening of single normal Na channels, which supports the hypothesis that modification occurs from the normal open state.     

Gating of Na channels. Inactivation modifiers discriminate among models

Macroscopic Na currents were recorded from N18 neuroblastoma cells by the whole-cell voltage-clamp technique. Inactivation of the Na currents was removed by intracellular application of proteolytic enzymes, trypsin, alpha-chymotrypsin, papain, or ficin, or bath application of N-bromoacetamide. Unlike what has been reported in squid giant axons and frog skeletal muscle fibers, these treatments often increased Na currents at all test pulse potentials. In addition, removal of inactivation gating shifted the midpoint of the peak Na conductance-voltage curve in the negative direction by 26 mV on average and greatly prolonged the rising phase of Na currents for small depolarizations. Polypeptide toxins from Leiurus quinquestriatus scorpion and Goniopora coral, which slow inactivation in adult nerve and muscle cells, also increase the peak Na conductance and shift the peak conductance curve in the negative direction by 7-10 mV in neuroblastoma cells. Control experiments argue against ascribing the shifts to series resistance artifacts or to spontaneous changes of the voltage dependence of Na channel kinetics. The negative shift of the peak conductance curve, the increase of peak Na currents, and the prolongation of the rise at small depolarization after removal of inactivation are consistent with gating kinetic models for neuroblastoma cell Na channels, where inactivation follows nearly irreversible activation with a relatively high, voltage-independent rate constant and Na channels open only once in a depolarization. As the same kind of experiment does not give apparent shifting of activation and prolongation of the rising phase of Na currents in adult axon and muscle membranes, the Na channels of these other membranes probably open more than once in a depolarization.     

BTX modification of Na channels in squid axons. I. State dependence of BTX action

The state dependence of Na channel modification by batrachotoxin (BTX) was investigated in voltage-clamped and internally perfused squid giant axons before (control axons) and after the pharmacological removal of the fast inactivation by pronase, chloramine-T, or NBA (pretreated axons). In control axons, in the presence of 2-5 microM BTX, a repetitive depolarization to open the channels was required to achieve a complete BTX modification, characterized by the suppression of the fast inactivation and a simultaneous 50-mV shift of the activation voltage dependence in the hyperpolarizing direction, whereas a single long-lasting (10 min) depolarization to +50 mV could promote the modification of only a small fraction of the channels, the noninactivating ones. In pretreated axons, such a single sustained depolarization as well as the repetitive depolarization could induce a complete modification, as evidenced by a similar shift of the activation voltage dependence. Therefore, the fast inactivated channels were not modified by BTX. We compared the rate of BTX modification of the open and slow inactivated channels in control and pretreated axons using different protocols: (a) During a repetitive depolarization with either 4- or 100-ms conditioning pulses to +80 mV, all the channels were modified in the open state in control axons as well as in pretreated axons, with a similar time constant of approximately 1.2 s. (b) In pronase-treated axons, when all the channels were in the slow inactivated state before BTX application, BTX could modify all the channels, but at a very slow rate, with a time constant of approximately 9.5 min. We conclude that at the macroscopic level BTX modification can occur through two different pathways: (a) via the open state, and (b) via the slow inactivated state of the channels that lack the fast inactivation, spontaneously or pharmacologically, but at a rate approximately 500-fold slower than through the main open channel pathway.     

Modification of inactivation in cardiac sodium channels: ionic current studies with Anthopleurin-A toxin

The site 3 toxin, Anthopleurin-A (Ap-A), was used to modify inactivation of sodium channels in voltage-clamped single canine cardiac Purkinje cells at approximately 12 degrees C. Although Ap-A toxin markedly prolonged decay of sodium current (INa) in response to step depolarizations, there was only a minor hyperpolarizing shift by 2.5 +/- 1.7 mV (n = 13) of the half-point of the peak conductance- voltage relationship with a slight steepening of the relationship from - 8.2 +/- 0.8 mV to -7.2 +/- 0.8 mV (n = 13). Increases in Gmax were dependent on the choice of cation used as a Na substitute intracellularly and ranged between 26 +/- 15% (Cs, n = 5) to 77 +/- 19% (TMA, n = 8). Associated with Ap-A toxin modification time to peak INa occurred later, but analysis of the time course INa at multiple potentials showed that the largest effects were on inactivation with only a small effect on activation. Consistent with little change in Na channel activation by Ap-A toxin, INa tail current relaxations at very negative potentials, where the dominant process of current relaxation is deactivation, were similar in control and after toxin modification. The time course of the development of inactivation after Ap-A toxin modification was dramatically prolonged at positive potentials where Na channels open. However, it was not prolonged after Ap-A toxin at negative potentials, where channels predominately inactivate directly from closed states. Steady state voltage-dependent availability (h infinity or steady state inactivation), which predominately reflects the voltage dependence of closed-closed transitions equilibrating with closed-inactivated transitions was shifted in the depolarizing direction by only 1.9 +/- 0.8 mV (n = 8) after toxin modification. The slope factor changed from 7.2 +/- 0.8 to 9.9 +/- 0.9 mV (n = 8), consistent with a prolongation of inactivation from the open state of Ap-A toxin modified channels at more depolarized potentials. We conclude that Ap-A selectively modifies Na channel inactivation from the open state with little effect on channel activation or on inactivation from closed state(s).     

Inactivation of the Na permeability in squid giant axons

Inactivation of the Na permeability has been studied in intact and perfused squid giant axons with the voltage clamp method. The main results are: 1. Upon depolarization inactivation develops along an exponential time course; the upper limit for an initial delay in the development of inactivation is 50-100 musec. 2. Adding 20-40 mM KCl to K-free external solution accelerates the development of inactivation and slows its removal. 3. Scorpion venoms increase the maintained conductance, i.e. make inactivation less complete; the voltage dependence of the maintained conductance is different from that of the peak conductance.     

Slow currents through single sodium channels of the adult rat heart

The currents through single Na+ channels from the sarcolemma of ventricular cells dissociated from adult rat hearts were studied using the patch-clamp technique. All patches had several Na+ channels; most had 5-10, while some had up to 50 channels. At 10 degrees C, the conductance of the channel was 9.8 pS. The mean current for sets of many identical pulses inactivated exponentially with a time constant of 1.7 +/- 0.6 ms at -40 mV. Careful examination of the mean currents revealed a small, slow component of inactivation at pulse potentials ranging from -60 to -30 mV. The time constant of the slow component was between 8 and 14 ms. The channels that caused the slow component had the same conductance and reversal potential as the fast Na+ currents and were blocked by tetrodotoxin. The slow currents appear to have been caused by repeated openings of one or more channels. The holding potential influenced the frequency with which such channel reopening occurred. The slow component was prominent during pulses from a holding potential of -100 mV, while it was very small during pulses from -140 mV. Ultraslow currents through the Na+ channel were observed occasionally in patches that had large numbers of channels. They consisted of bursts of 10 or more sequential openings of a single channel and lasted for up to 150 ms. We conclude that the single channel data cannot be explained by standard models, even those that have two inactivated states or two open states of the channel. Our results suggest that Na+ channels can function in several different "modes," each with a different inactivation rate.     

Characterization of the isoform-specific differences in the gating of neuronal and muscle sodium channels

Human heart (hH1), human skeletal muscle (hSkM1), and rat brain (rIIA) Na channels were expressed in cultured cells and the activation and inactivation of the whole-cell Na currents measured using the patch clamp technique. hH1 Na channels were found to activate and inactivate at more hyperpolarized voltages than hSkM1 and rIIA. The conductance versus voltage and steady state inactivation relationships have midpoints of -48 and -92 mV (hH1), -28 and -72 mV (hSkM1), and -22 and -61 mV (rIIA). At depolarized voltages, where Na channels predominately inactivate from the open state, the inactivation of hH1 is 2-fold slower than that of hSkM1 and rIIA. The recovery from fast inactivation of all three isoforms is well described by a single rapid component with time constants at -100 mV of 44 ms (hH1), 4.7 ms (hSkM1), and 7.6 ms (rIIA). After accounting for differences in voltage dependence, the kinetics of activation, inactivation, and recovery of hH1 were found to be generally slower than those of hSkM1 and rIIA. Modeling of Na channel gating at hyperpolarized voltages where the channel does not open suggests that the slow rate of recovery from inactivation of hH1 accounts for most of the differences in the steady-state inactivation of these Na channels.     

Geographical distribution and inactivation kinetics in internally perfused Myxicola giant axons.

In some preparations the time constant of Na current inactivation determined with two pulses (tau c) is larger over some range of potentials than that determined from the current decay during a single pulse (tau h), while in others tau c(V) and tau h(V) are the same. Myxicola giant axons obtained from specimens collected from coastal waters of northeastern North America display a tau c - tau h difference under all conditions we have tested. In these axons tau c(V) and tau h(V) are unchanged by reduction of Na current density, addition of K-channel blockers, or internal perfusion. Specimens of the same species, Myxicola infundibulum, collected from a different geographical location, the south coast of England, have been studied under internal perfusion with K as the major cation internally, with reduced external Na concentration and in the presence of K-channel blockers. In these axons tau c(V) and tau h(V) approximately superpose, raising the possibility that dramatic differences in Na current kinetics may not necessarily reflect basic differences in the organization of the Na channel gating machinery.     

Destruction of Sodium Conductance Inactivation in Squid Axons Perfused with Pronase

We have studied the effects of the proteolytic enzyme Pronase on the membrane currents of voltage-clamped squid axons. Internal perfusion of the axons with Pronase rather selectively destroys inactivation of the Na conductance (g_Na). At the level of a single channel, Pronase probably acts in an all-or-none manner: each channel inactivates normally until its inactivation gate is destroyed, and then it no longer inactivates. Pronase reduces _Na, possibly by destroying some of the channels, but after removal of its inactivation gate a Na channel seems no longer vulnerable to Pronase. The turn-off kinetics and the voltage dependence of the Na channel activation gates are not affected by Pronase, and it is probable that the enzyme does not affect these gates in any way. Neither the K channels nor their activation gates are affected in a specific way by Pronase. Tetrodotoxin does not protect the inactivation gates from Pronase, nor does maintained inactivation of the Na channels during exposure to Pronase. Our results suggest that the inactivation gate is a readily accessible protein attached to the inner end of each Na channel. It is shown clearly that activation and inactivation of Na channels are separable processes, and that Na channels are distinct from K channels.     

Osmotic swelling of hepatocytes increases membrane conductance but not membrane capacitance.

We have used the whole-cell patch-clamp technique to study changes in membrane conductance and membrane capacitance after osmotic swelling in rat hepatocytes. Hypoosmotic solutions induced an instantaneous increase in the volume of patch-clamped cells that was followed by a slow decline reminiscent of regulatory volume decrease as seen in intact cells. These morphological changes were associated with a transient increase in membrane conductance. The rise in conductance was not correlated with changes in capacitance, neither in time after the initiation of cell swelling nor in magnitude. Therefore we conclude that an osmotically induced increase in conductance is probably a result of the activation of existent channels in the plasmalemma and not a result of the fusion of vesicle membrane containing ionic channels.     

Cardiac Na currents and the inactivating, reopening, and waiting properties of single cardiac Na channels

Tetrodotoxin (TTX)-sensitive Na currents were examined in single dissociated ventricular myocytes from neonatal rats. Single channel and whole cell currents were measured using the patch-clamp method. The channel density was calculated as 2/micron 2, which agreed with our usual finding of four channels per membrane patch. At 20 degrees C, the single channel conductance was 20 pS. The open time distributions were fit by a single-exponential function with a mean open time of approximately 1.0 ms at membrane potentials from -60 to -40 mV. Averaged single channel and whole cell currents were similar when scaled and showed both fast and slow rates of inactivation. The inactivation and activation gating shifted quickly to hyperpolarized potentials for channels in cell-attached as well as excised patches, whereas a much slower shift occurred in whole cells. Slowly inactivating currents were present in both whole cell and single channel current measurements at potentials as positive as -40 mV. In whole cell measurements, the potential range could be extended, and slow inactivation was present at potentials as positive as -10 mV. The curves relating steady state activation and inactivation to membrane potential had very little overlap, and slow inactivation occurred at potentials that were positive to the overlap. Slow inactivation is in this way distinguishable from the overlap or window current, and the slowly inactivating current may contribute to the plateau of the rat cardiac action potential. On rare occasions, a second set of Na channels having a smaller unit conductance and briefer duration was observed. However, a separate set of threshold channels, as described by Gilly and Armstrong (1984. Nature [Lond.]. 309:448), was not found. For the commonly observed Na channels, the number of openings in some samples far exceeded the number of channels per patch and the latencies to first opening or waiting times were not sufficiently dispersed to account for the slowly inactivating currents: the slow inactivation was produced by channel reopening. A general model was developed to predict the number of openings in each sample. Models in which the number of openings per sample was due to a dispersion of waiting times combined with a rapid transition from an open to an absorbing inactivated state were unsatisfactory and a model that was more consistent with the results was identified.     

Temperature dependence of Na currents in rabbit and frog muscle membranes

The effect of temperature (0-22 degrees C) on the kinetics of Na channel conductance was determined in voltage-clamped rabbit and frog skeletal muscle fibers using the triple-Vaseline-gap technique. The Hodgkin-Huxley model was used to extract kinetic parameters; the time course of the conductance change during step depolarization followed m3h kinetics. Arrhenius plots of activation time constants (tau m), determined at both moderate (-10 to -20 mV) and high (+100 mV) depolarizations, were linear in both types of muscle. In rabbit muscle, Arrhenius plots of the inactivation time constant (tau h) were markedly nonlinear at +100 mV, but much less so at -20 mV. The reverse situation was found in frog muscle. The contrast between the highly nonlinear Arrhenius plot of tau h at +100 mV in rabbit muscle, compared with that of frog muscle, was interpreted as revealing an intrinsic nonlinearity in the temperature dependence of mammalian muscle Na inactivation. These results are consistent with the notion that mammalian cell membranes undergo thermotropic membrane phase transitions that alter lipid-channel interactions in the 0-22 degrees C range. Furthermore, the observation that Na channel activation appears to be resistant to this effect suggests that the gating mechanisms that govern activation and inactivation reside in physically distinct regions of the channel.