Size Matters: Erythromelalgia Mutation S241T in Nav1.7 Alters Channel Gating*

The Nav1.7 sodium channel is preferentially expressed in most nociceptive dorsal root ganglion neurons and in sympathetic neurons. Inherited erythromelalgia (IEM, also known as erythermalgia), an autosomal dominant neuropathy characterized by burning pain in the extremities in response to mild warmth, has been linked to mutations in Nav1.7. Recently, a substitution of Ser-241 by threonine (S241T) in the domain I S4-S5 linker of Nav1.7 was identified in a family with IEM. To investigate the possible causative role of this mutation in the pathophysiology of IEM, we used whole-cell voltage-clamp analysis to study the effects of S241T on Nav1.7 gating in HEK293 cells. We found a hyperpolarizing shift of activation midpoint by 8.4 mV, an accelerated time to peak, slowing of deactivation, and an increase in the current in response to small, slow depolarizations. Additionally, S241T produced an enhancement of slow inactivation, shifting the midpoint by -12.3 mV. Because serine and threonine have similar biochemical properties, the S241T substitution suggested that the size of the side chain at this position affected channel gating. To test this hypothesis, we investigated the effect of S241A and S241L substitutions on the gating properties of Nav1.7. Although S241A did not alter the properties of the channel, S241L mimicked the effects of S241T. We conclude that the linker between S4 and S5 in domain I of Nav1.7 modulates gating of this channel, and that a larger side chain at position 241 interferes with its gating mechanisms.

The Nav1.7 IEM mutations, which have been investigated to date, shift the midpoint of activation to hyperpolarized potentials, except for F1449V, all slow channel deactivation time constants, and increase the response to small, slow depolarizations (ramp currents) (8, 11, 16 -18). Expression of F1449V and L858H mutant Nav1.7 channels in dorsal root ganglion neurons leads to an increase in excitability as measured by a reduced threshold and increased firing frequency, as would be expected from the changes in gating properties (8,17).
The seven missense mutations in Nav1.7, which have been described to date (8 -12, 17), alter residues that are highly conserved among all sodium channels. Except for two mutations in S4 and S6 of domain I (DI), the other five mutations change residues in cytoplasmic linkers, with one mutation in the S4-S5 linker of domain I (DI S4-S5), three mutations in the DII S4-S5, and one mutation in the loop between DIII and DIV (7). Ser-241 to threonine (S241T) mutation in DI S4-S5 was identified in several generations of a Flemish family with IEM (10).
The S241T substitution might be considered a conservative change because of the similar chemical properties of the side chain, which can serve as a phosphoacceptor site. We reasoned that the increase in size of the side chain by one methyl group might alter the local folding of the linker or cause a long-range allosteric effect in other structures, which could affect properties of the channel leading to neuronal hyperexcitability. We used whole-cell patch-clamp methods to study the biophysical effects of the naturally occurring S241T mutation and then tested the hypothesis that these effects are attributable to the longer side chain of threonine by investigating the effect of substituting Ser-241 by alanine (S241A) and leucine (S241L). Our data show that residue 241 altered channel gating and suggest that channel function was sensitive to the size of the amino acid residue at that position.

EXPERIMENTAL PROCEDURES
Plasmids and Stable Cell Lines-The S241T, S241A, and S241L mutations were introduced individually into hNav1.7 R , * This work was supported in part by the Medical Research Service and Rehabilitation Research Service, Department of Veterans Administration Rehabilitation Research and Development Service, the National Multiple Sclerosis Society, and the Erythromelalgia Association. The Center for Neuroscience and Regeneration Research is a collaboration of the Paralyzed Veterans of America and the United Spinal Association with Yale University. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 (20) of HEK293 cells stably expressing the sodium channels Nav1.7 R , S241T, S241A, or S241L derivatives were performed with an EPC-9 amplifier (HEKA Electronics, Lambrecht/Pfalz, Germany) using fire polished 1-1.5 M⍀ electrodes (World Precision Instruments, Inc., Sarasota, FL). The pipette solution contained (in mM): 140 CsF, 10 NaCl, 1 EGTA, and 10 HEPES; 302 mosmol (pH 7.4, adjusted with CsOH), and the extracellular bath contained (in mM): 140 NaCl, 3 KCl, 10 glucose, 10 HEPES, 1 MgCl 2 , 1 CaCl 2 , 0.0003 tetrodotoxin; 310 mosmol (pH 7.4, adjusted with NaOH). Tetrodotoxin was added to the bath solution to block all endogenous voltage-gated sodium currents, which might be present in HEK293 cells (21) and thereby study Nav1.7 R in isolation. All recordings were conducted at room temperature (ϳ21°C). The pipette potential was adjusted to zero before seal formation, and the voltages were not corrected for liquid junction potential. Capacity transients were cancelled, and series resistance was compensated by 65-95%. Leakage current was subtracted digitally online using hyperpolarizing potentials applied after the test pulse (P/4 procedure), except for ramp current recordings. Currents were acquired using Pulse software (HEKA Electronics), filtered at 2.9 kHz, and sampled at a rate of 20 kHz.
Voltage protocols were carried out 4 min after establishing cell access. Briefly, standard current-voltage (I-V) families were obtained using 40 ms pulses from a holding potential of Ϫ100 mV to a range of potentials (Ϫ100 to ϩ60 mV) in 10 mV steps with 10 s between pulses. The peak value at each potential was plotted to form I-V curves. Activation curves were obtained by calculating the conductance (G) at each voltage (V).
with V rev being the reversal potential, determined for each cell individually. Activation curves were fitted with the following Boltzmann distribution equation, where G Na is the voltage-dependent sodium conductance, G Na,max is the maximal sodium conductance, V1 ⁄ 2 is the potential at which activation is half-maximal, V m is the membrane potential, and k is the slope factor. Ramp currents were elicited by slowly depolarizing voltage ramps, ranging from Ϫ100 to ϩ20 mV at a rate of 0.2 mV/ms. Size of absolute ramp currents was measured as the difference between first current downward reflection and the peak of the ramp current. Protocols for assessing steady-state fast inactivation consisted of a series of prepulses (Ϫ130 to Ϫ10 mV) lasting 500 ms from the holding potential of Ϫ100 mV followed by a 40 ms depolarization to Ϫ10 mV to assess the noninactivated transient current. The normalized curves were fitted using a Boltzmann distribution equation, where I Na,max is the peak sodium current elicited after the most hyperpolarized prepulse, V m is the preconditioning pulse potential, V1 ⁄ 2 is the half-maximal sodium current, and k is the slope factor. The decay time constants were obtained by fitting a single exponential function to the current traces. Steady-state slow inactivation was determined using 30 s prepulses ranging from Ϫ130 mV to ϩ10 mV followed by a brief hyperpolarization to Ϫ120 mV for 100 ms to remove fast inactivation. Available channels were then tested with a 20 ms test pulse to Ϫ10 mV. Elicited currents were normalized to the maximal peak current and fitted using a Boltzmann distribution with an added constant (offset).
The rate of deactivation was measured using a short (0.5 ms) depolarizing pulse to Ϫ20 mV followed by a 50 ms repolarizing pulse to potentials ranging from Ϫ100 to Ϫ40 mV. Decaying current was then fitted with a single exponential function using PulseFit software (HEKA Electronics).
Statistical analysis was carried out using SPSS software (SPSS Inc., Chicago, IL) performing one-way analysis of variance, and significance for multiple comparisons was tested using Tamhane T2 post hoc analysis. Results revealing p Ͻ 0.05 were considered significant. All data are presented as mean Ϯ S.E.

S241T Lowers Threshold for Activation, Slows Deactivation, and Increases Ramp
Responses of hNa v 1.7 R -The amino acid threonine is larger than serine by a methyl (CH 3 ) group, but both have similar chemical reactivity at their reactive OH group on the side chain, for example, phosphorylation. Nevertheless, the S241T mutation has been linked to IEM (10). Therefore, we investigated the effects of the S241T substitution on the biophysical properties of Nav1.7.
To characterize the effects of substitution of Ser-241 with threonine on the biophysical properties of Nav1.7, we per-formed whole-cell patch-clamp recordings from HEK293 cell lines stably transfected with wild type Nav1.7 R and the mutant derivative S241T. Voltage steps from a holding potential of Ϫ100 mV produced a fast activating and fast inactivating current for both constructs (Fig.  1A). The V1 ⁄ 2 of activation ( Fig. 1B and Table 1) was Ϫ25.6 Ϯ 0.9 mV (n ϭ 47) for Nav1.7 R and Ϫ34.0 Ϯ 1.1 mV (n ϭ 39) for the S241T mutant channels, showing a shift of Ϫ8.4 mV. The V1 ⁄ 2 of steady-state fast inactivation was not changed by this mutation (Fig. 1B and Table 1). Thus, the predicted window current would be larger, close to resting membrane potential of neurons.
Kinetics of activation were investigated by measuring the time to peak of the transient current. Considering the observed shift in V1 ⁄ 2 of activation induced by the S241T mutation, we expected a faster activation depending on the step voltage. At Ϫ30 mV, activation of S241T was significantly faster than that of Nav1.7 R (Fig. 1C; p Ͻ 0.05, one-way analysis of variance, Tamhane post hoc analysis). The same statistical analysis was performed for all indicated p values. Deactivation of S241T was significantly slower at Ϫ40 and Ϫ50 mV (Nav1.7 R , 0.34 Ϯ 0.03 ms; S241T, 1.55 Ϯ 0.15 ms at Ϫ40 mV; p Ͻ 0.01 (Fig. 1D)). The inactivation decay time constants at these potentials were 12.0 Ϯ 3.6 ms (Ϫ40 mV) and 5.3 Ϯ 0.4 ms (Ϫ50 mV) for Nav1.7 R and 8.4 Ϯ 1 ms (Ϫ40 mV) and 3.2 Ϯ 0.4 ms (Ϫ50 mV) for S241T. As the inactivation time constants were significantly slower than the deactivation time constants, contamination of channel closing by inactivation could be considered low.
Small, slow depolarizations of Nav1.7 evoke ramp currents that mimicked weak natural stimuli, for example, generator potentials. The slow closed-state inactivation of Nav1.7 permits this channel to produce relatively larger ramp currents compared with other sodium channels (19,22). HEK293 cells which express Nav1.7 R channels produced FIGURE 1. S241T lowers the threshold for activation of Nav1.7 R by shifting steady-state activation and delaying deactivation. A, representative I-V families recorded from HEK293 cells stably expressing Nav1.7 R (left) or the mutation S241T (right). Cells were held at Ϫ100 mV, and depolarizing steps were applied to membrane potentials ranging from Ϫ100 to ϩ40 mV in 10 mV steps. B, voltage dependence of steady-state fast inactivation, and activation is shown for Nav1.7 R (gray squares; activation, n ϭ 47 and inactivation, n ϭ 51) and S241T (pink squares; activation, n ϭ 39 and inactivation, n ϭ 29). Steady-state fast inactivation was obtained by measuring the current evoked by a 40 ms test pulse to Ϫ10 mV after inducing fast inactivation with a 500 ms prepulse to potentials ranging from Ϫ130 to Ϫ10 mV. Conductance curves were deduced from current-voltage families, normalized, and fitted with a Boltzmann equation, as described under "Experimental Procedures." C, voltage dependence of time to peak is shown for step depolarizations from a holding potential of Ϫ100 mV to the indicated voltages. The mutation S241T (pink squares, n ϭ 24) opens faster than Nav1.7 R (gray squares, n ϭ 61) at Ϫ30 mV. D, S241T (pink squares, n ϭ 34) deactivates significantly slower at Ϫ50 and Ϫ40 mV repolarizing potentials than Nav1.7 R (gray squares, n ϭ 44). Deactivation time constants were obtained by a single exponential fit of tail currents elicited by repolarization to the indicated potentials from a brief depolarization of 0.5 ms to Ϫ20 mV; see inset for protocol. E, S241T increases current elicited by slow ramp depolariziations. Cells were held at Ϫ100 mV and stimulated with a depolarizing voltage ramp that increased to 20 mV within 600 ms. The bar graph shows the current recorded during the ramp expressed as percentage of transient peak current obtained during initial I-V families, 4 min after breaking into the cell. Representative current traces of Nav1.7 R (gray line) and S241T (pink line) are shown. F, The mutation S241T (pink squares, n ϭ 12) shifts voltage-dependence of slow inactivation of Nav1.7 R (gray squares, n ϭ 21). Steady-state slow inactivation was assessed using a 20 ms pulse to Ϫ10 mV after a 30 s prepulse to potentials ranging from Ϫ140 to 10 mV followed by a 100 ms pulse to Ϫ120 mV to remove fast inactivation. See inset for protocol. ***, significant difference to Nav1.7 R with p Ͻ 0.001; **, p Ͻ 0.01; *, p Ͻ 0.05. Data points are connected with straight lines and represent means Ϯ S.E. NOVEMBER 24, 2006 • VOLUME 281 • NUMBER 47 a ramp current as expected (peak ramp current as percentage of transient peak is 0.13% Ϯ 0.01%, n ϭ 46 (Fig. 1E)). S241T substitution caused a 3.6-fold increase in the ramp current (S241T, 0.49% Ϯ 0.07% of peak transient current, n ϭ 38; p Ͻ 0.001). The voltage at which the ramp current peaked was hyperpolarized for S241T as expected from the negative shift of the V1 ⁄ 2 of activation.
Size of the Side Chain at Position 241 Modulates Gating Properties-Serine and threonine side chains differ by one methyl group ( Fig. 2A), which could interfere with the packing of the S4-S5 linker secondary structure when the two amino acids are exchanged. To address the possibility that the different size of the side chain of the residue at position 241 contributes to the functional changes in the S241T mutant channels, we substituted Ser-241 by the relatively smaller residue alanine (S241A), which carried one less oxygen atom or the significantly larger residue leucine (S241L) (Fig. 2A). S241L substitution was chosen because it had a bulkier side chain compared with serine; also, a S246L mutation in Nav1.4 (Ser-246 in Nav1.4 corresponds to Ser-241 in Nav1.7) has been described in a patient with myasthenic syndrome (23). HEK293 cells, stably transfected with S241A and S241L, produced fast activating and inactivating voltage-gated sodium currents, similar to Nav1.7 R and S241T channels ( Fig. 2A).
We investigated the effect of S241A and S241L on voltage-dependence of activation and steady-state fast inactivation. S241A did not significantly alter activation properties of the channel (Fig.  2B, blue squares and Table 1). In addition, the midpoint of steadystate fast inactivation was not significantly different compared with Nav1.7 R . The fast inactivation slope decreased to 6.2 Ϯ 0.1 mV (n ϭ 25; p Ͻ 0.05; slope for Nav1.7 R , 6.6 Ϯ 0.1 mV).
S241L currents, however, showed a significant hyperpolarizing shift in activation midpoint compared with Nav1.7 R (by 4.9 mV; n ϭ 21; p Ͻ 0.05 (Fig. 2B, yellow squares and Table 1)), similar to the effect of S241T; the activation midpoints of S241L and S241T were not significantly different. The slope of activa- Currents were evoked by depolarizing the membrane from the holding potential of Ϫ100 mV to potentials ranging from Ϫ100 to 40 mV in 10 mV steps. In this figure and Figs. 3 and 4, data for Nav1.7 R and S241T are the same as in Fig. 1 and are included for comparison. The structure of the amino acid at position 241 in the S4-S5 linker in domain I is shown as structural model, the stick model highlights the side chain of serine for Nav1.7 R , threonine for S241T, alanine for S241A, and leucine for S241L. B, voltage dependence of steady-state fast inactivation and activation are shown for Nav1.7 R (gray squares; activation, n ϭ 47 and inactivation, n ϭ 51), S241T (pink squares; activation, n ϭ 39 and inactivation, n ϭ 29), S241A (blue squares; activation, n ϭ 35 and inactivation, n ϭ 25), and S241L (yellow squares; activation, n ϭ 21 and inactivation, n ϭ 20). Steadystate fast inactivation was obtained by measuring the current evoked by a 40 ms pulse to Ϫ10 mV after inducing fast inactivation with a 500 ms prepulse to potentials ranging from Ϫ130 to Ϫ10 mV. Conductance curves were deduced from current-voltage families normalized and fitted with a Boltzmann equation, as described under "Experimental Procedures." Data points are connected with straight lines and represent means Ϯ S.E.

TABLE 1 Parameters of Boltzmann fits and values for ramp currents
All values are in mV, except for the data of the ramp currents, which are presented as percentage of transient peak currents. To investigate the changes in activation kinetics in more detail, we compared the time to peak and deactivation time constants of the mutant channels. S241A had no significant influence on the time needed to reach the peak of the transient current or on the deactivation time constants (Fig. 3, A and B). S241L, however, made the channel open faster (significant shorter time to peak in the voltage range from Ϫ30 mV to ϩ30 mV; Nav1.7 R , 0.8 Ϯ 0.02 ms; S241L, 0.6 Ϯ 0.02 ms at Ϫ30 mV; p Ͻ 0.001; Fig. 3A) and deactivate slower (S241L, 0.69 Ϯ 0.10 ms at Ϫ40 mV; p Ͻ 0.05, not significant at Ϫ50 mV) (Fig. 3B); both effects were similar to those induced by the naturally occurring mutation S241T. The decay time constants were much larger than the deactivation time constants for each mutation (6.9 Ϯ 1.4 ms for S241A and 3.2 Ϯ 0.3 ms for S241L inactivation time constant at Ϫ40 mV).
Changes in slow inactivation induced by an increase in size of side chain in the mutant S241T channels were mimicked by S241L: the V1 ⁄ 2 was shifted by Ϫ10.8 mV to Ϫ89.2 Ϯ 2.1 mV (n ϭ 7; p Ͻ 0.05), and the slope was reduced to 6.6 Ϯ 0.3 mV (p Ͻ 0.001) so that the slow inactivation kinetics were almost indistinguishable for these two mutations ( Fig. 4 and Table 1). S241A induced a smaller change in slope (9.5 Ϯ 0.6 mV, n ϭ 13; p Ͻ 0.05) but did not alter the V1 ⁄ 2 (Ϫ83.5 Ϯ 2.5 mV). For both mutations the offset did not change significantly (9.8% Ϯ 1.2% for S241L and 9.6% Ϯ 1.8% for S241A).

DISCUSSION
We show in this study that the IEM mutation S241T of the human voltage-gated sodium channel Nav1.7 alters the biophysical properties of the channel, consistent with the pathophysiology of this disorder. Activation of S241T was shifted to hyperpolarized potentials, time to peak was faster, deactivation was slower, and response to slow, small depolarization (ramp currents) was increased. All these changes would be expected to increase the excitability of nociceptive dorsal root ganglion neurons harboring the mutation, thus contributing to the pain in IEM (8,17). Slow inactivation, on the other hand, was enhanced in the S241T mutant channels, similar to previously reported IEM mutations (8,16,18), but its contribution to the pathophysiology of erythromelalgia is not known at this time. This change would theoretically render mutations expressing DRG neurons hypoexcitable. Current clamp investigations on the F1449V mutation in DRG neurons, however, showed that the expression of the mutant channel in DRG neurons, although enhancing slow inactivation, renders the neurons hyperexcitable, as reflected by a lower threshold for firing and an increased frequency of firing of DRG neurons expressing the mutant channel compared with those expressing WT channels (8). Similarly, L858H mutation, which also shows enhanced slow inactivation (16), renders DRG neurons expressing the mutant channel hyperexcitable (17). Thus, the effect of slow inactivation on the channel appears to be outweighed by other gating changes.
Serine and threonine are polar amino acids that share similar biochemical characteristics because of their reactive OH group in their side chains. Thus, substitutions such as S241T are generally expected to maintain the function of the protein.
Although Ser-241 in the context of the S4-S5 linker (SVKK) is a potential phosphoacceptor site for protein kinase C, it is reasonable to expect that TVKK would be an equally probable substrate. However, Ser-241 is a highly conserved residue in the DI S4-S5 linker in all the vertebrate sodium channels, and because S241T has been identified in patients in whom sensory neurons are hyperexcitable, a substitution at this position would be predicted to affect the gating properties of the channel, similar to other IEM mutations. Threonine side chain, however, is larger than that of serine by a methyl group. To assess the role of the size of the side chain at position 241 in hNav1.7, we mutated serine to the slightly smaller amino acid alanine (S241A) and to the larger amino acid leucine (S241L), both of which are nonpolar, unlike serine or threonine. S241A showed characteristics similar to wild type Nav1.7 R , whereas S241L induced changes comparable with those observed for the naturally occurring mutation S241T. Thus, polarity of the res-  , n ϭ 35). B, the mutations S241L (yellow squares, n ϭ 19) and S241T (pink squares, n ϭ 34) deactivate significantly slower at Ϫ40 mV repolarizing potentials than Nav1.7 R (gray squares, n ϭ 44) or S241A (blue squares, n ϭ 28). Deactivation time constants of a single exponential fit of tail currents elicited by repolarization to the indicated potentials from a brief depolarization of 0.5 ms to Ϫ20 mV; see inset in Fig. 1D for protocol. Data points are connected with straight lines and represent means Ϯ S.E. *, significant differences to Nav1.7 R with p Ͻ 0.05; **, p Ͻ 0.01; ***, p Ͻ 0.001. NOVEMBER 24, 2006 • VOLUME 281 • NUMBER 47 idue at position 241 does not appear to contribute to the mutant phenotype, although the size of the side chain at this position is a factor that has an important impact on the functional properties of the channel.

Nav1.7 Mutation S241T Alters Channel Gating
Gating and kinetic properties of Nav1.7 channels are similar when the channel is expressed in DRG neurons or in the stable HEK293 cell line (19). Thus, the properties of WT and mutant channel derivatives in stable HEK293 cells reported in this study are likely to resemble those in native DRG neurons. There is no experimental evidence that implicates the S4-S5 linker in interactions with ␤-subunits. Thus, whereas we cannot formally rule out the possibility that an allosteric effect of the Ser-241 substitutions might influence interactions with other subunits, we would argue that it is not likely, if present, to have a major effect.
The S4-S5 linker of ion channels has been implicated in the regulation of gating properties of channels, and mutations in these linkers cause a number of neurological and muscular disorders. Sodium channel mutations in the S4-S5 linkers in domains II-IV have been shown to affect gating of several channels: Nav1.7 (16), Nav1.6 (11, 24), Nav1.5 (25), Nav1.4 (26 -32), and Nav1.2 (33). There is also evidence for an important functional role of the S4-S5 linker from mutations of other ion channels. The S218L mutation in the DI S4-S5 linker of voltage-gated calcium channel Cav2.1 induces a gain of function and causes familial hemiplegic migraine (34). Interaction between the residue Asp-540 in S4-S5 and Arg-665 in S6 is suggested to contribute to stabilization of the closed conformation in hERG channels (35). The hyperpolarization-activated anion channel HCN2 is shown to display disrupted gating because of mutations in S4-S5 linker (36). Thus, the S4-S5 linker appears to modulate gating properties of a diverse group of ion channels.
Recently published models of voltage-gated ion channels (37-45) depict a voltage sensor consisting of segments S1-S4, whereas segments S5 and S6 form the pore, control ion selectivity, and opening of the channel. The voltage sensor complex is connected to the pore region via the S4-S5 linker. A model for the potassium channel based on x-ray crystal structure of Kv1.2 shows the S4-S5 linker located parallel to the membrane inner surface, facing aqueous and lipids on opposite sides of the ␣-helix, and crossing the C-terminal part of S6 making amino acid contact with it (38). Ferrer et al. verified the proximity of the S4-S5 linker to S6 in the shaker potassium channel by introducing cysteine residues and testing the effect of oxidation (46). Recent computer modeling of Kv1.2 and an insect sodium channel support the idea that S4-S5 linker is located close to and influences bending of S6, which is important for subsequent opening of the pore (47,48). Increasing the size of a side chain within the S4-S5 linker that possibly faces S6 might be expected to facilitate bending of S6 and thus lower the threshold for channel activation.
The midpoint of steady-state fast inactivation was not affected by S241T, S241A, or S241L. This is consistent with the idea that the S4-S5 linkers in DI and DII play only a minor role in fast inactivation, whereas the S4-S5 linkers of DIII and DIV are postulated to be the receptor site for the fast inactivation gate (30, 33, 49 -52). Several mutations in S4-S5 linker of domain III of sodium channels Nav1.4 (53), Nav1.5 (25), and Nav1.6 (24) have been shown to affect fast inactivation. The suggested minor role of the DI S4-S5 linker in fast inactivation might explain why we did not observe significant changes in steady-state fast inactivation of S241T; only S241L, which replaced serine with a much larger amino acid, produced a small negative shift, suggesting that the function of the receptor site for the inactivation gate might have been enhanced by a larger residue.
The data presented here show that S241T alters Nav1.7 gating in a manner that could underlie dorsal root ganglion neuron hyperexcitability (8,17). We have also shown that S241L induces gating changes similar to those of S241T and would thus be predicted to cause a mutant phenotype in patients who might carry this mutation. Paradoxically, the S246L substitution in Nav1.4 (which is equivalent to Ser-241 in Nav1.7) observed in a family with myasthenic syndrome, did not in itself induce a disease phenotype, which instead was linked to the mutation V1442E in the DIV S3-S4 extracellular linker of the channel (23). Nav1.4/S246L caused a 7 mV hyperpolarizing shift of steady-state fast inactivation of Nav1.4 and shifted the V1 ⁄ 2 of steady-state slow inactivation to the left, comparable with our results for Nav1.7 R /S241L. Interestingly, it also increased the population of channels that did not undergo slow inactivation at positive potentials and did not cause a change in V1 ⁄ 2 of activation (23). The difference in the effects on activation of analogous mutations in Nav1.7 and Nav1.4 and in the clinical effects of these mutations (disease-inducing for Nav1.7 R /S241L FIGURE 4. S241L enhances slow inactivation of Nav1.7 R , similar to S241T. The mutations S241L (yellow squares, n ϭ 7) and S241T (pink squares, n ϭ 12) shift voltage-dependence of slow inactivation to hyperpolarized potentials compared with Nav1.7 R (gray squares, n ϭ 21) and S241A (blue squares, n ϭ 13). Steady-state slow inactivation was assessed using a 20 ms pulse to Ϫ10 mV after a 30 s prepulse at potentials ranging from Ϫ140 to 10 mV followed by a 100 ms pulse to Ϫ120 mV to remove fast inactivation. See inset in Fig. 1F for protocol. Data points are connected with straight lines and represent means Ϯ S.E. Right panels, bar graphs show the results of a Boltzmann fit (midpoint of slow inactivation and slope). *, significant difference to Nav1.7 R with p Ͻ 0.05; ***, p Ͻ 0.001.
but not for Nav1.4/S246L) suggests an isoform-specific role of the S4-S5 linker and cautions against over-generalization of the limited experimental results from heterologous expression systems. Because Nav1.4 is mainly expressed in skeletal muscle cells (54,55), whereas Nav1.7 is expressed in sensory neurons and sympathetic neurons (13)(14)(15), it is not surprising that these analogous mutations in Nav1.4 and Nav1.7 have different functional effects. As shown recently, cell background can have a significant influence on the effects of mutant ion-channel electrogenesis (17).