Abstract

Sodium channel Nav1.9 is expressed in peripheral nociceptive neurons, as well as visceral afferents, and has been shown to act as a threshold channel. Painful peripheral neuropathy represents a significant public health challenge and may involve gain-of-function variants in sodium channels that are preferentially expressed in peripheral sensory neurons. Although gain-of-function variants of peripheral sodium channels Nav1.7 and Nav1.8 have recently been found in painful small fibre neuropathy, the aetiology of peripheral neuropathy in many cases remains unknown. We evaluated 459 patients who were referred for possible painful peripheral neuropathy, and confirmed the diagnosis of small fibre neuropathy in a cohort of 393 patients (369 patients with pure small fibre neuropathy, and small fibre neuropathy together with large fibre involvement in an additional 24 patients). From this cohort of 393 patients with peripheral neuropathy, we sequenced SCN11A in 345 patients without mutations in SCN9A and SCN10A, and found eight variants in 12 patients. Functional profiling by electrophysiological recordings showed that these Nav1.9 mutations confer gain-of-function attributes to the channel, depolarize resting membrane potential of dorsal root ganglion neurons, enhance spontaneous firing, and increase evoked firing of these neurons. Our data show, for the first time, missense mutations of Nav1.9 in individuals with painful peripheral neuropathy. These genetic and functional observations identify missense mutations of Nav1.9 as a cause of painful peripheral neuropathy.

Introduction

Painful peripheral neuropathy represents an unmet medical challenge (Hoeijmakers et al., 2012a). Recently, Faber et al. (2012a, b) reported gain-of-function missense variants in sodium channels Nav1.7 and Nav1.8, which are abundantly expressed in spinal sensory (dorsal root ganglion, DRG) neurons and their axons within peripheral nerves (Black and Waxman, 2002; Persson et al., 2010; Black et al., 2012), in patients with a diagnosis of painful small fibre neuropathy according to the international criteria (Tesfaye et al., 2010; Hoeijmakers et al., 2012a). Although the majority of cases with small fibre neuropathy are of unknown aetiology, the involvement of Nav1.7 and Nav1.8 suggests that other sodium channels that are expressed within peripheral sensory neurons may also harbour variants that may be linked to painful neuropathy.

Nav1.9, initially termed NaN (Dib-Hajj et al., 1998) or SNS2 (Tate et al., 1998), is preferentially expressed within peripheral DRG, trigeminal and myenteric neurons (Dib-Hajj et al., 1998, 2002; Rugiero et al., 2003), and has been found within free nerve terminals in skin and cornea (Black and Waxman, 2002; Dib-Hajj et al., 2002; Persson et al., 2010). Nav1.9 current is resistant to tetrodotoxin (TTX), with a distinctive hyperpolarized voltage-dependence of activation close to the resting membrane potential of most neurons (−60 to −70 mV) and broad overlap between activation and steady-state inactivation predicting that it should produce a persistent current (Cummins et al., 1999). Current-clamp recordings and computer simulation studies have shown that Nav1.9 acts as a threshold channel that regulates excitability of DRG neurons (Herzog et al., 2001; Baker et al., 2003; Ostman et al., 2007).

Animal studies have suggested that Nav1.9 plays a role in pain signalling (Dib-Hajj et al., 2010). Recently, a gain-of-function mutation in SCN11A, the gene that encodes Nav1.9, was reported to be linked to two rare genetic disorders: congenital insensitivity to pain (Leipold et al., 2013), and familial episodic pain (Zhang et al., 2013). Here we describe missense mutations in Nav1.9 in patients with painful peripheral neuropathy. We show by voltage-clamp, current-clamp and dynamic clamp electrophysiology that these mutations alter gating properties of the channel and increase spontaneous firing and evoked firing of DRG neurons. These genetic and functional observations link gain-of-function Nav1.9 variants to a common neurological disorder, painful peripheral neuropathy.

Materials and methods

All human studies were approved by an institutional review board at Maastricht University Medical Centre. All aspects of the study were explained to all patients, and written informed consent was obtained before study initiation.

SCN11A mutation analysis

Genomic DNA was extracted from blood using the Puregene® genomic DNA isolation kit (Gentra-Systems). All coding exons and flanking intronic sequences were amplified and sequenced as described previously (Faber et al., 2012a). Genomic sequences were compared with reference Nav1.9 complementary DNA (NM_014139.2) to identify sequence variations (Dib-Hajj et al., 1999) using Alamut Mutation-Interpretation Software (Interactive-Biosoftware). A control panel of DNA from healthy 224 and 470 chromosomes from The Netherlands and the UK, respectively, were screened for all new mutations. The NCBI SNP database, the Human Gene Mutation Database (HGMD®), the 1000 Genomes project (www.1000genomes.org/data), and the Human EVS database were also screened.

Plasmids

Human Nav1.9 was amplified from human DRG neuron complementary DNA template (Dib-Hajj et al., 1999), and cloned into a modified pcDNA3 vector designed to be a low-copy number plasmid (Klugbauer et al., 1995). Using Mega mutagenesis protocol, a sequence encoding enhanced green fluorescent protein was cloned upstream of the Nav1.9 ATG with a ‘stopGo’ 33 amino acid 2A linker, such that the GFP-2A adaptor and the Nav1.9 channel proteins are produced as independent proteins from the same messenger RNA (Ryan and Drew, 1994; Atkins et al., 2007; Luke et al., 2008). In pilot experiments, we confirmed the presence of GFP-2A and Nav1.9 channels as independent proteins on western blots, and demonstrated that the channel produces the expected TTX-resistant ultra-slow inactivating current (data not shown). The c.1142T>C (p.Ile381Thr) and c.3473T>C (p.Leu1158Pro) mutations were introduced into the construct using QuikChange® II XL site-directed mutagenesis (Stratagene) and referred to as I381T and L1158P channels hereinafter.

Dorsal root ganglion neuron isolation and transfection for voltage-clamp studies

Animal studies followed a protocol approved by the Veterans Administration Connecticut Healthcare System Institutional Animal Care and Use Committees. For voltage-clamp recording, DRG neurons were isolated, as previously reported (Dib-Hajj et al., 2009), from 4–8 week old Nav1.9 knock-out mice (Ostman et al., 2007) and transfected by electroporation. Briefly, DRGs were harvested, incubated at 37°C for 20 min in complete saline solution (in mM: 137 NaCl, 5.3 KCl, 1 MgCl2, 25 sorbitol, 3 CaCl2, and 10 HEPES, adjusted to pH 7.2 with NaOH) containing 0.5 U/ml Liberase TM (Roche Diagnostics) and 0.6 mM EDTA before 15 min incubation at 37°C in complete saline solution containing 0.5 U/ml Liberase TL (Roche Diagnostics), 0.6 mM EDTA, and 30 U/ml papain (Worthington Biochemical). Tissue was then centrifuged and triturated in 0.5 ml of DRG media: Dulbecco's Modified Eagle Medium/F12 (1:1) with 100 U/ml penicillin, 0.1 mg/ml streptomycin (Invitrogen) and 10% foetal bovine serum (Hyclone), containing 1.5 mg/ml bovine serum albumin (low endotoxin) and 1.5 mg/ml trypsin inhibitor (Sigma). After trituration, the cell suspension was transfected with wild-type or mutant Nav1.9 constructs with a Nucleofector IIS device (Lonza), using the Amaxa® Basic Neuron SCN Nucleofector™ Kit (VSPI-1003) and SCN Basic Neuro Protocol 6. Briefly, the cell suspension was centrifuged (100 g for 3 min), the supernatant was carefully removed, and the cell pellet was resuspended in 20 μl Nucleofector™ solution, mixed with 2.5 μg of wild-type human Nav1.9 or mutant construct, and transfected. After electroporation, 100 μl of calcium-free Dulbecco’s modified Eagle medium (Invitrogen) was added and cells were incubated at 37°C for 5 min in a 95% air/5% (vol/vol) CO2 incubator to allow neurons to recover. The cell mixture was then diluted with DRG media containing 1.5 mg/ml bovine serum albumin (low endotoxin) and 1.5 mg/ml trypsin inhibitor, seeded onto poly-D-lysine/laminin-coated coverslips (BD) and incubated at 37°C to allow DRG neurons to attach to the coverslips. After 40 min, DRG media was added into each well to a final volume of 1.4 ml and the DRG neurons were maintained at 37°C in a 95% air/5% (vol/vol) CO2 incubator for 40 h before voltage-clamp recording.

Voltage-clamp recordings

Voltage-clamp recordings were obtained at 22 ± 1°C, 40–48 h after transfection using an EPC-10 amplifier (HEKA Electronics). Small DRG neurons (<25 μm) with robust green fluorescence and no apparent neurites were selected for voltage-clamp recording. Pipette potential was adjusted to zero before seal formation, and liquid junction potential was not corrected. Capacity transients were cancelled and voltage errors minimized with ∼90% series resistance compensation. Voltage-dependent currents were acquired with Patchmaster at 5 min after establishing whole-cell configuration, sampled at 50 kHz, and filtered at 2.9 kHz. The pipette solution contained (in mM): 140 CsF, 10 NaCl, 1 EGTA, and 10 HEPES, pH 7.3 with CsOH (adjusted to 315 mOsmol/l with dextrose). The extracellular bath solution contained (in mM): 140 NaCl, 3 KCl, 1 MgCl2, 1 CaCl2, 10 HEPES, 5 CsCl, 20 tetraethylammonium chloride (TEAċCl), pH 7.3 with NaOH (327 mOsmol/l). TTX (500 nM), CdCl2 (0.1 mM), and 4-aminopyridine (1 mM) were added in the bath solution to block endogenous TTX-sensitive (sensitive) voltage-gated sodium currents, calcium currents, and potassium currents, respectively.

Initially, neurons were held at −100 mV, and a series of 100-ms step depolarizations from −100 to −20 mV were applied in 5 mV increments at 5 s intervals to record the total TTX-resistant current. The neurons were then held at −60 mV and depolarized through a series of test pulses following a 100 ms step to −100 mV to inactivate Nav1.9, leaving the Nav1.8 current intact. The Nav1.8 current could then be subtracted from the total TTX-resistant current in order to isolate the Nav1.9 current (Cummins et al., 1999). Activation curves were obtained by converting I to conductance (G) at each voltage (V) using the equation G = I / (V − Vrev), where Vrev is the reversal potential, which was determined for each cell individually. Activation curves were then fit with Boltzmann functions in the form of G = Gmax / {1 + exp[(V1/2,act − V) / k]}, where Gmax is the maximal sodium conductance, V1/2,act is the potential at which activation is half-maximal, V is the test potential and k is the slope factor.

Steady-state fast-inactivation was assessed with a series of 500 ms prepulses (−140 to 0 mV in 10-mV increments) and the remaining non-inactivated channels were activated by a 50-ms step depolarization to −45 mV (a voltage at which Nav1.8 channel does not open). Steady-state slow-inactivation was determined with 30 s prepulses ranging from −120 to −30 mV followed by a 100 ms pulse to −100 mV to remove fast-inactivation; prepulse was from −130 for I381T mutant as the plateau was not reached at −120 mV (Fig. 4E). Remaining available channels were activated by a 50 ms test pulse to −45 mV. Peak inward currents obtained from steady-state fast-inactivation and slow-inactivation protocols were normalized to the maximal peak current (Imax) and fit with Boltzmann functions: I / Imax = A + (1 − A) / {1 + exp[(V − V1/2,inact) / k]}, where V represents the inactivating prepulse potential, V1/2,inact represents the midpoint of the steady-state fast-inactivation or slow-inactivation.

Deactivation was estimated from current decay, using a 25 ms short depolarizing pulse to −45 mV followed by a 150 ms repolarizing pulse to potentials ranging from −110 to −70 mV with 5-mV increments. Deactivation kinetics were calculated by fitting the decaying currents with a single exponential function.

Dorsal root ganglion neuron isolation and transfection for current-clamp studies

DRGs from 4- to 8-week-old Sprague Dawley rats were harvested and dissociated as described previously (Dib-Hajj et al., 2009). Briefly, DRG were dissociated with a 20 min incubation at 37°C in complete saline solution containing 1.5 mg/ml Collagenase A (Roche Diagnostics) and 0.6 mM EDTA, followed by a 20-min incubation at 37°C in complete saline solution containing 1.5 mg/ml Collagenase D (Roche Diagnostics), 0.6 mM EDTA, and 30 U/ml papain. A suspension of DRG neurons was then prepared as described above, 2 ml of DRG media was added to the suspension and the suspension filtered using a 70 µm nylon cell strainer (Becton Dickinson). The cell strainer was washed with 2 × 2 ml of DRG media, the washes were combined with the suspension and then centrifuged at 100 g for 3 min, the supernatant removed and cells transfected as described above using the Amaxa® Basic Neuron SCN Nucleofector™ Kit (VSPI-1003) and SCN Basic Neuro Protocol 6. After electroporation, cells were treated as described above except that 30–40 min after seeding DRG media was added to each well to a final volume of 1.0 ml containing nerve growth factor (50 ng/ml) and glial cell line-derived neurotrophic factor (50 ng/ml).

Current-clamp recordings

Current-clamp recordings were obtained as previously described (Faber et al., 2012b) from small (<30-µm diameter) green fluorescent protein-labelled DRG neurons 40–48 h after transfection using EPC-10 (HEKA) amplifier. Electrodes had a resistance of 1–3 MΩ when filled with the pipette solution, which contained (in mM): 140 KCl, 0.5 EGTA, 5 HEPES, and 3 Mg-ATP, pH 7.3 with KOH (adjusted to 315 mOsm with dextrose). The extracellular solution contained (in mM): 140 NaCl, 3 KCl, 2 MgCl2, 2 CaCl2, 10 HEPES, pH 7.3 with NaOH (adjusted to 320 mOsm with dextrose). Whole-cell configuration was obtained in voltage-clamp mode before proceeding to current-clamp recording mode. Cells with action potential overshoot of <40 mV were excluded for data collection. Threshold was determined by the first action potential elicited by a series of 200-ms depolarizing current injections that increased in 5 pA increments. For comparison with our previous current-clamp studies on neuropathy-associated mutations of Nav1.7 (Faber et al., 2012a) and Nav1.8 (Faber et al., 2012b), the response to a graded series of 500 ms current stimuli from 25 to 500 pA, in 25 pA increments, was assessed.

Data analysis

Electrophysiological data were analysed using Fitmaster (HEKA Electronics) and Origin 8.5.1 (Microcal), and presented as means ± standard error (SE). Statistical significance was determined by unpaired Student’s t-tests (current-clamp except firing frequency and spontaneous activity), paired Student’s t-test [measurement of changes in current threshold corresponding to resting membrane potential (RMP) alterations], Mann-Whitney test (firing frequency) or two-proportion z-test (comparison of proportion of cells producing spontaneous activity).

Controls for voltage- and current-clamp

To minimize inherent culture-to-culture variability and to overcome the possibility of error induced by pooling the results of experiments on DRG neurons harvested from multiple animals and cultured over many months, each mutant was compared by voltage- and current-clamp with contemporaneous controls of wild-type human Nav1.9 expressed in cultures of cells prepared, transfected and recorded under identical conditions by the same physiologist.

Dynamic clamp recording

DRG neurons obtained from adult Nav1.9 knockout mice were enzymatically isolated as described previously (Dib-Hajj et al., 2009; Ahn et al., 2013) and were used for dynamic clamp recording within 12 h in vitro. Small DRG neurons (soma diameter <25 µm) were dynamically clamped (Sharp et al., 1993; Kemenes et al., 2011; Samu et al., 2012) in whole-cell configuration using patch pipettes pulled from glass capillaries (World Precision Instruments; catalogue number PG10165-4). Pipette resistance was 2–3 MΩ when filled with the intracellular solution (in mM): 140 KCl; 3 MgATP, 0.5 EGTA, 5 HEPES; 10 glucose, pH 7.3 with KOH (adjusted to 325 mOsm with sucrose). The extracellular solution was Hank's Balanced Salt Solution (Invitrogen, catalogue number 14025) (in mM): 1.3 CaCl2, 0.5 MgCl2, 0.4 MgSO4, 5.3 KCl, 0.4 KH2PO4, 4.2 NaHCO3, 138 NaCl, 0.3 Na2HPO4, 5.6 glucose (adjusted to 325 mOsm with sucrose). Liquid junction potential (+3.8 mV) between pipette and bath solutions was compensated. Membrane voltages and currents were recorded in dynamic-clamp using MultiClamp 700B amplifier (Molecular Devices) interfaced with CED Power 1401 mk II DAI and Signal software (CED), digitized by Digidata 1440A DAC and stored on the hard disk using pCLAMP 10 software (Molecular Devices). Voltage and current traces were filtered at 10 kHz and digitized at 50 kHz. Recordings were made at room temperature (22–23°C). All data are presented as means ± SE. Data were analysed using pCLAMP 10 (Molecular Devices), and Origin 8.5 (OriginLab) software. The hypothesis that population means are significantly different was assessed using Student’s t-test or Mann-Whitney non-parametric test, when appropriate (*P < 0.05; **P < 0.01).

Kinetic model of Nav1.9 channel

Nav1.9 channel model used for dynamic clamp was based on Hodgkin-Huxley equations dm/dt = αm(1 − m) − βmm; dh/dt = αh(1 − h) − βhh; ds/dt = αs(1 − s) − βss, where m, h and s are channel activation, fast- and slow-inactivation variables, and α (β) are forward (backward) rate constants, respectively. The following rate constants were used for the wild-type Nav1.9 channel model:  

formula
 
formula
 
formula

I381T mutant was described by the following rate constants:  

formula
 
formula
 
formula

Sodium current was described by INa = gmax × m × h × S × (Vm − ENa), where gmax is maximal conductance, Vm is membrane voltage, ENa = 65 mV is sodium reversal potential (Herzog et al., 2001). Steady-state fast- and slow-inactivation, steady-state activation, time constants for activation and deactivation for both wild-type and I381T mutant channels were from our voltage-clamp recordings; fast-inactivation and repriming time constants of wild-type channels were from Vanoye et al. (2013); slow-inactivation time constant was from Herzog et al. (2001); gmax was set to 30 nS (Rush and Waxman, 2004) and our voltage-clamp recordings, which show ∼2 nA of current. We assumed a voltage-dependent shift of I381T channel inactivation and repriming kinetics to be equivalent to the V1/2 shift of I381T channel steady-state inactivation; slow-inactivation time constant was from Herzog et al. (2001). All parameters were adjusted for the respective junction potentials.

In dynamic-clamp recordings we fixed slow-inactivation at −60 mV RMP (S = 0.027 resulting in CED dynamic-clamp g1max = S × gmax = 0.8 nS, referred to thereafter as 100% NaV1.9 conductance) because we limited stimulation to 2 s trains at 5 Hz where the development of slow-inactivation is not appreciable. Currents evoked by voltage-clamp and action potential-clamp protocols were calculated in 10 µs precision using custom program written in Origin8.5 LabTalk.

Results

DNA analysis

Two mutations in SCN11A, the gene encoding Nav1.9, substituting highly conserved residues in membrane-spanning segments, were selected for detailed functional analysis from eight SCN11A mutations. Of 459 patients referred to the Maastricht University Medical Centre for possible painful peripheral neuropathy between September 2009 and October 2013, the diagnosis of predominantly small fibre neuropathy was confirmed in 393 patients (Table 1). In 369 of these 393 patients, the diagnosis of definite small fibre neuropathy was made, and this was considered to be pure small fibre neuropathy based on clinical symptoms, abnormal quantitative sensory testing and/or abnormal skin biopsy, in combination with normal nerve conduction studies, in accordance with published criteria (Lauria et al., 2012). In an additional 24 of these 393 patients, a diagnosis of small fibre neuropathy was made, but on the basis of clinical signs of large fibre involvement or abnormal nerve conduction studies the diagnosis of predominantly small fibre neuropathy together with large fibre dysfunction was made. Thus our overall cohort of 393 patients with a definite diagnosis of painful peripheral neuropathy consisted of 369 patients with pure small fibre neuropathy and 24 patients with small fibre neuropathy together with large fibre dysfunction. In this paper we describe four patients with painful peripheral neuropathy: two patients with pure small fibre neuropathy and two patients with predominantly small fibre neuropathy together with large fibre involvement. From this cohort of 393 patients, SCN9A mutations were found in 34 patients (Faber et al., 2012a), and SCN10A mutations were found in 15 patients (Faber et al., 2012b). SCN11A was sequenced in those patients without mutations in SCN9A and SCN10A (Fig. 1A). Additionally, in one patient the variant found in SCN9A did not confer hyperexcitability in DRG neurons when assessed by electrophysiology; thus SCN10A and SCN11A were subsequently tested, which identified a variant in SCN11A. Twelve of 345 patients tested, carried eight variants in SCN11A: exon 9, c.1142T>C (p.Ile381Thr) (n = 2); exon 9, c.1257G>T (p.Lys419Asn) (n = 2); exon 12, c.1744G>A (p.Ala582Thr) (n = 2); exon 14, c.2042C>A (p.Ala681Asp); exon 15, c.2524G>C (p.Ala842Pro); exon 20, c.3473T>C (p.Leu1158Pro) (n = 2); intron 24, c.4057-1G>A; exon 26, c.5067C>G (p.Phe1689Leu). All of the subjects carrying these variants were heterozygous for these new mutations. Of the seven missense variants found in SCN11A, four (c.1142T>C, p.Ile381Thr; c.1744G>A (p.Ala582Thr); c.2042C>A (p.Ala681Asp); c.3473T>C, p. Leu1158Pro), in exons 9, 12, 14, and 20, respectively, substituted amino acids in membrane-spanning segments of the channel; the I381T mutation located in domain I transmembrane 6 (DI/S6), and the L1158P mutation located in DIII/S4, were chosen for functional analysis (Fig. 1B).

Figure 1

Scheme for patient selection and localization of Nav1.9 variants. (A) Scheme for selecting patients with small fibre neuropathy for SCN11A screening. *SCN9A variant c.3799C>G (p.Leu1267Val) did not show abnormalities in cell electrophysiology; therefore, SCN10A and SCN11A were tested. (B) Schematic of sodium channel polypeptide showing the locations of the missense mutations in Nav1.9 in individuals with painful peripheral neuropathy and no mutations in Nav1.7 or Nav1.8. The mutations are: (1) I381T; (2) K419N; (3) A582T; (4) A681D; (5) A842P; (6) L1158P; and (7) F1689L. The mutations I381T and L1158P, which were profiled in this study, were each present in two patients. One additional mutation was found at the 3’ acceptor splice site in intron 24.

Figure 1

Scheme for patient selection and localization of Nav1.9 variants. (A) Scheme for selecting patients with small fibre neuropathy for SCN11A screening. *SCN9A variant c.3799C>G (p.Leu1267Val) did not show abnormalities in cell electrophysiology; therefore, SCN10A and SCN11A were tested. (B) Schematic of sodium channel polypeptide showing the locations of the missense mutations in Nav1.9 in individuals with painful peripheral neuropathy and no mutations in Nav1.7 or Nav1.8. The mutations are: (1) I381T; (2) K419N; (3) A582T; (4) A681D; (5) A842P; (6) L1158P; and (7) F1689L. The mutations I381T and L1158P, which were profiled in this study, were each present in two patients. One additional mutation was found at the 3’ acceptor splice site in intron 24.

Table 1

Patient characteristics and additional investigations in a cohort of 393 patients diagnosed with painful peripheral neuropathy

Age Mean 49 years (range 1–80, SD 14.1) 
Family history Positive n = 60, negative n = 262, unknown n = 71 
Abnormal intraepidermal nerve fibre density n = 135 
Abnormal quantitative sensory testing n = 363 
Abnormal nerve conduction studies n = 24 
Age Mean 49 years (range 1–80, SD 14.1) 
Family history Positive n = 60, negative n = 262, unknown n = 71 
Abnormal intraepidermal nerve fibre density n = 135 
Abnormal quantitative sensory testing n = 363 
Abnormal nerve conduction studies n = 24 

SD = standard deviation.

The I381T and L1158P mutations were not found in 224 and 470 chromosomes from control populations in The Netherlands and the UK, respectively. The I381T variant was not reported in the dbSNP, EVS and 1000 Genome databases; the L1158P was reported with a low minor allele frequency in the EVS (n = 8/13006) and the 1000 Genome (n = 1/2184) databases, but was not reported in the dbSNP database (n = 0/4552). The mutation I381T substitutes a hydrophobic highly conserved isoleucine with a polar threonine in DI/S6, a pore-lining segment. The mutation L1158P substitutes a hydrophobic residue with a long sidechain, leucine with a proline residue, which introduces a turn in the DIII/S4 of domain 3, the voltage-sensor of this domain (Fig. 1B). In silico analysis suggested these mutations to be deleterious.

Clinical description

Patient 1: c.3473T>C; p.Leu1158Pro

A 73-year-old male was referred to our neurology outpatient clinic with a 2-year history of numbness and tingling in both legs, which started after intestinal surgery for colorectal carcinoma, with the symptoms gradually disseminating to his hands. He was not treated with chemotherapy. He also developed cramps and stabbing pain in his feet, and contact with sheets was irritating, and sometimes his hands discoloured white, accompanied with sharp pain. The complaints were not associated with temperature changes or activity. Family history was unremarkable. The patient also complained of hyperhidrosis, dry eyes and mouth, orthostatic dizziness, palpitations and occasionally hot flushes. Pregabalin initially gave some pain relief, but only for a few months.

Physical examination showed no abnormalities, and nerve conduction studies were unremarkable. Quantitative sensory testing showed abnormal thresholds for warmth sensation in both feet and for cold sensation in the right foot. Skin biopsy demonstrated an intraepidermal nerve fibre density of 3.1 per mm, which was normal compared with gender- and age-matched normative values (5th percentile: 2.1 per mm). The patient was diagnosed with small fibre neuropathy. DNA analysis showed no mutations in the SCN9A and SCN10A genes, however, a new variant was found in the SCN11A gene: c.3473T>C; p.Leu1158Pro.

Patient 2: c.3473T>C; p.Leu1158Pro

A 61-year-old female presented with a 2-year history of numbness, tingling and a dull burning pain in the toes and foot soles. She also complained of restless legs and occasional calf cramps, with standing and walking aggravating the complaints. The toes intermittently showed a blue-purple discolouration. The patient also complained of hyperhidrosis, diarrhoea, a dry mouth, dry eyes, palpitations and hot flushes. A low dose of gabapentin (300 mg/day) resulted in pain relief. Medical history showed primary biliary cirrhosis, well-treated hyperthyroidism and a laryngeal carcinoma, treated 10 years before with surgery and radiation therapy. The patient also had a history of herpes zoster attacks of her buttocks. Family history is unremarkable.

Physical examination showed decreased vibration sense at both ankles. Laboratory tests revealed positive anti-mitochondrion antibodies, appropriate to the primary biliary cirrhosis. Blood tests demonstrated elevated liver enzymes, weak bands in the gamma region and vitamin B12 deficiency. Nerve conduction studies were normal. Quantitative sensory testing showed abnormal thresholds for warmth and cold sensation in both feet. A skin biopsy demonstrated an intraepidermal nerve fibre density of 1.0 per mm, which was low compared with gender- and age-matched normative values (5th percentile: 3.2 per mm), which confirmed a diagnosis of small fibre neuropathy. Based on the impaired vibration sense, a diagnosis of predominantly small fibre neuropathy together with large fibre involvement was made. DNA analysis showed no mutations in the SCN9A and SCN10A genes. A new variant was found in the SCN11A gene: c.3473T>C; p.Leu1158Pro.

Patient 3: c.1142T>C; p.Ile381Thr

A 50-year-old male presented with a 10-year history of numbness of the hands, arms and feet, and a family history of joint pain. Low ambient temperatures triggered burning pain in his fingers and toes, sometimes with a red discolouration. In addition, he complained of arthralgia. The pain tended to be more severe in the morning. The patient also complained of hyperhidrosis, diarrhoea, a dry mouth, orthostatic dizziness, palpitations and hot flushes. A combination of nortriptyline, arthrotec and acetaminophen gave some pain relief.

Physical examination showed slightly reduced vibration sense at the toes. Nerve conduction studies demonstrated signs of a moderate peripheral polyneuropathy. Quantitative sensory testing demonstrated abnormal thresholds for warmth sensation in both hands and the left foot. The thresholds for cold sensation were abnormal in the left foot. A skin biopsy demonstrated an intraepidermal nerve fibre density of 4.8 per mm, which was normal (5th percentile: 3.5 per mm; 10th percentile: 4.9 per mm). The patient was diagnosed with predominantly small fibre neuropathy together with large fibre dysfunction. DNA analysis showed no mutations in the SCN9A and SCN10A genes, however, a variant was found in the SCN11A gene: c.1142T>C; p.Ile381Thr.

Patient 4: c.1142T>C; p.Ile381Thr

A 56-year-old female presented with a 5-year history of burning, tingling sensations and a dull, pulsing pain of the fingers and toes. The complaints gradually extended to the upper legs, pubic region, ears and tip of the tongue. Bathing in lukewarm water and a cool breeze diminished the complaints. Contact with sheets and wearing shoes was painful. The patient also complained of hyperhidrosis, diarrhoea, micturation problems, palpitations and hot flushes. Several anti-epileptic drugs caused side effects and were discontinued, and amitriptyline gave no pain relief.

The patient suffered from an autosomal dominant form of Alport’s syndrome, located on chromosome 2, characterized by glomerulonephritis, end-stage kidney disease and hearing loss, but at the time of evaluation had normal renal function. Both her mother and sister were also diagnosed with Alport’s syndrome. Her mother experienced the same pain complaints as the patient. DNA from the mother was not available for analysis.

Physical examination showed no abnormalities, and nerve conduction studies were normal. Quantitative sensory testing showed abnormal thresholds for warmth sensation in both feet. The thresholds for cold sensation were abnormal in the left foot. A skin biopsy demonstrated an intraepidermal nerve fibre density of 8.1 per mm, which was normal (5th percentile: 4.3 per mm). The patient was diagnosed with small fibre neuropathy. DNA analysis showed a variant (c.3799C>G; p.Leu1267Val) in the SCN9A gene. However, this variant did not confer hyperexcitability in DRG neurons when assessed by electrophysiology (unpublished data); thus SCN10A and SCN11A were subsequently tested, which identified a variant in SCN11A:c.1142T>C; p.Ile381Thr.

Functional assessment of L1158P: patch-clamp

Wild-type human Nav1.9 and mutant channels L1158P were each transiently expressed in DRG neurons from Nav1.9 knock-out mice. Nav1.9 currents were obtained by subtracting Nav1.8 sodium currents from the total TTX-resistant sodium currents. To record the total TTX-resistant sodium currents, cells were held at −100 mV with 500 nM TTX in the bath and stepped to membrane potentials from −100 to −20 mV for 100 ms in 5 mV increments. Subsequently, the holding potential was depolarized to −60 mV to inactivate Nav1.9 channels (Cummins et al., 1999), and Nav1.8 currents were elicited by the same protocol as recording the total TTX-resistant currents following a prepulse of 100 ms at −100 mV. The Nav1.9 current was obtained by subtracting the Nav1.8 current from the total TTX-resistant current. Representative voltage-dependent inward currents of wild-type and L1158P displaying the ultra-slow inactivation characteristic of Nav1.9 are presented in Fig. 2A and B, respectively. Although the nominal Nav1.9 channel current density of L1158P mutant channel (97 ± 18 pA/pF; n = 13) was smaller than that of wild-type channel (135 ± 36 pA/pF; n = 16), the difference did not reach statistical significance (P > 0.05). Figure 2C shows the voltage-dependence of activation for wild-type and L1158P channels. The midpoint of activation for L1158P (estimated by fitting with a Boltzmann function) was significantly shifted by −6.7 mV (wild-type: −49.6 ± 2.0 mV, n = 12; L1158P: −56.3 ± 1.1 mV, n = 10; P < 0.05). However, the slope factor was not significantly different between wild-type and L1158P channels (wild-type: 7.3 ± 0.5, n = 12; L1158P: 7.2 ± 0.5, n = 10; P > 0.05). Steady-state fast-inactivation was measured using a series of 500 ms prepulses from −140 to 0 mV followed by 50 ms test pulse to −45 mV. As shown in Fig. 2C, the midpoint of fast-inactivation for L1158P mutant channels (−48.8 ± 1.1 mV, n = 10) was not significantly different from that of wild-type channels (−50.5 ± 1.3 mV, n = 10; P > 0.05). The slope factor of fast-inactivation was not significantly different between wild-type and L1158P channels (wild-type: 10.7 ± 0.8, n = 10; L1158P: 11.1 ± 0.8, n = 10; P > 0.05).

Figure 2

Voltage-clamp analysis of L1158P mutant channels. (A and B) Representative current traces recorded from DRG neurons expressing wild-type (A) or L1158P (B) mutant channels. (C) Comparison of activation for wild-type (WT) and L1158P mutant channels. L1158P shifts activation by −6.7 mV (wild-type: −49.6 ± 2.0 mV, n = 12; L1158P: −56.3 ± 1.1 mV, n = 10; P < 0.05). Comparison of steady-state fast-inactivation for wild-type and L1158P mutant channels. L1158P does not affect steady-state fast-inactivation (wild-type: −50.5 ± 1.3 mV, n = 10; L1158P: −48.8 ± 1.1 mV, n = 10; P > 0.05). (D) Comparison of deactivation time constants. L1158P (n = 15) deactivates significantly slower than wild-type (n = 12), *P < 0.05. Inset: Representative currents recorded from two different neurons expressing wild-type (black) or L1158P (red) sodium channels at a repolarization voltage of −70 mV. The wild-type current peak was scaled to match the L1158P current peak for comparing the kinetics of current decay. (E) Steady-state slow-inactivation curves for wild-type and L1158P. Steady-state slow-inactivation was not altered by the L1158P mutation (wild-type: −79.0 ± 1.7 mV, n = 12; L1158P: −78.0 ± 1.3 mV, n = 10; P > 0.05).

Figure 2

Voltage-clamp analysis of L1158P mutant channels. (A and B) Representative current traces recorded from DRG neurons expressing wild-type (A) or L1158P (B) mutant channels. (C) Comparison of activation for wild-type (WT) and L1158P mutant channels. L1158P shifts activation by −6.7 mV (wild-type: −49.6 ± 2.0 mV, n = 12; L1158P: −56.3 ± 1.1 mV, n = 10; P < 0.05). Comparison of steady-state fast-inactivation for wild-type and L1158P mutant channels. L1158P does not affect steady-state fast-inactivation (wild-type: −50.5 ± 1.3 mV, n = 10; L1158P: −48.8 ± 1.1 mV, n = 10; P > 0.05). (D) Comparison of deactivation time constants. L1158P (n = 15) deactivates significantly slower than wild-type (n = 12), *P < 0.05. Inset: Representative currents recorded from two different neurons expressing wild-type (black) or L1158P (red) sodium channels at a repolarization voltage of −70 mV. The wild-type current peak was scaled to match the L1158P current peak for comparing the kinetics of current decay. (E) Steady-state slow-inactivation curves for wild-type and L1158P. Steady-state slow-inactivation was not altered by the L1158P mutation (wild-type: −79.0 ± 1.7 mV, n = 12; L1158P: −78.0 ± 1.3 mV, n = 10; P > 0.05).

The kinetics of deactivation, which reflect the transition from the open to the closed state, were measured using single exponential fits to estimate the time of current decay at potentials from −110 to −70 mV after briefly activating the channels at −45 mV for 25 ms. As shown in Fig. 2D, the rates of current decay of L1158P mutant channels were significantly slower than those of wild-type channels across all deactivation potentials tested. The inset in Fig. 2D shows the slowed deactivation for L1158P at −70 mV.

We also measured steady-state slow-inactivation, a process that inactivates the channel much more slowly than fast-inactivation, with 30 s prepulses, followed by 100 ms pulses to −100 mV to allow recovery from fast-inactivation, and then a 50 ms test pulse to −45 mV to determine the fraction of Nav1.9 channel current available. As shown in Fig. 2E, the L1158P mutation has no significant effect on steady-state slow-inactivation. The midpoint was −79.0 ± 1.7 mV for wild-type (n = 12) and −78.0 ± 1.3 mV for L1158P mutant channels (n = 10; P > 0.05), respectively. The slope factor of L1158P channels (5.3 ± 0.1, n = 10) was not significantly different from that of wild-type channels (5.4 ± 0.3, n = 12; P > 0.05).

To assess the effect of the L1158P mutation on excitability, we expressed wild-type and L1158P mutant channels in rat small DRG neurons and performed current-clamp recordings (Fig. 3). The input resistance was similar between DRG neurons expressing wild-type channels (702 ± 81 MΩ, n = 26) and those expressing L1158P mutant channels (712 ± 95 MΩ, n = 22; P > 0.05). However, the RMP of DRG neurons that expressed L1158P mutant channels (−50.9 ± 1.0 mV, n = 22) was significantly depolarized by 3.5 mV, compared with DRG neurons expressing wild-type channels (−54.4 ± 1.2 mV, n = 26; P < 0.05) (Fig. 3C).

Figure 3

Current-clamp analysis of L1158P mutation. (A) Responses of representative DRG neurons expressing wild-type (upper) and L1158P (lower) channels, respectively, to 500 ms depolarization current steps which are 1×, 1.5× and 2× (left, centre and right traces, respectively) the current threshold for action potential generation. (B) Comparison of repetitive action potential firing spike number between DRG neurons expressing wild-type (n = 26) and L1158P (n = 22) across a range of 500 ms step current injections from 25 to 500 pA; *P < 0.05. (C) L1158P significantly depolarized resting membrane potential by 3.5 mV compared with wild-type channels (wild-type: −54.4 ± 1.2 mV, n = 26; L1158P: −50.9 ± 1.0 mV, n = 22; *P < 0.05). (D) Current threshold was reduced significantly after expression of L1158P (165 ± 22 pA, n = 22), compared with wild-type (250 ± 21 pA, n = 26; **P < 0.01). AP = action potential.

Figure 3

Current-clamp analysis of L1158P mutation. (A) Responses of representative DRG neurons expressing wild-type (upper) and L1158P (lower) channels, respectively, to 500 ms depolarization current steps which are 1×, 1.5× and 2× (left, centre and right traces, respectively) the current threshold for action potential generation. (B) Comparison of repetitive action potential firing spike number between DRG neurons expressing wild-type (n = 26) and L1158P (n = 22) across a range of 500 ms step current injections from 25 to 500 pA; *P < 0.05. (C) L1158P significantly depolarized resting membrane potential by 3.5 mV compared with wild-type channels (wild-type: −54.4 ± 1.2 mV, n = 26; L1158P: −50.9 ± 1.0 mV, n = 22; *P < 0.05). (D) Current threshold was reduced significantly after expression of L1158P (165 ± 22 pA, n = 22), compared with wild-type (250 ± 21 pA, n = 26; **P < 0.01). AP = action potential.

Current threshold was significantly reduced after expression of L1158P (165 ± 22 pA, n = 22), compared with wild-type channels (250 ± 21 pA, n = 26; P < 0.01) (Fig. 3D). However, there was no significant difference between these two groups of DRG neurons for either voltage threshold (the voltage at which the take-off occurs) (wild-type: −16.8 ± 1.1 mV; L1158P: −16.4 ± 0.9 mV, P > 0.05), action potential width at 0 mV (wild-type: 3.30 ± 0.30 ms; L1158P: 3.27 ± 0.25 ms, P > 0.05) or action potential amplitude (wild-type: 106.8 ± 1.3 mV; L1158P: 104.9 ± 2.2 mV, P > 0.05) between neurons expressing wild-type (n = 26) or L1158P (n = 22) channels.

To evaluate repetitive action potential firing, we assessed the response of DRG neurons to a series of 500 ms current injections from 25 to 500 pA in 25 pA increments. Figure 3A shows the responses of a representative DRG neuron which expressed wild-type or L1158P mutant channels, to 500 ms current steps at 1×, 1.5× and 2× current threshold. The DRG neuron expressing wild-type channels generated a single spike in response to the current injections at both 1× and 1.5× current threshold, and two spikes to the injection of 2× current threshold. In contrast, the neuron expressing L1158P mutant channels fired two or more action potentials in response to the stimuli of 1.5× current threshold, and five action potentials in response to stimulation at 2× current threshold. Figure 3B shows data for the population of transfected DRG neurons, and clearly demonstrates that neurons expressing the L1158P channels fired significantly more action potentials compared to neurons expressing wild-type channels. Additionally, we found that for both groups of DRG neurons, a small proportion of neurons fired spontaneously; however, we did not detect a significant difference for the portion of spontaneously firing neurons between the two groups (wild-type: 5 of 31 cells, 16%; L1158P: 7 of 29 cells, 24%; P > 0.05, z-test).

Functional assessment of I381T: patch-clamp

Representative voltage-dependent inward currents of wild-type and I381T, which again display the ultra-slow inactivation characteristic of Nav1.9, are presented in Fig. 4A and B, respectively. The average peak inward current amplitude was not significantly different between wild-type Nav1.9 and I381T (wild-type, −2.39 ± 0.45 nA, n = 13; I381T, −3.10 ± 0.81 nA, n = 14; P > 0.05). When normalized for capacitance, similarly, the average peak current density of I381T (−198 ± 47 pA/pF, n = 14) did not differ from that of wild-type (−231 ± 50 pA/pF, n = 13; P > 0.05).

Figure 4

Voltage-clamp analysis of I381T mutant channels. Inward currents recorded from a small-diameter DRG neuron expressing wild-type Nav1.9 (A) or I381T mutant (B) sodium channels. (C) Comparison of the voltage dependence of activation (wild-type, −55.3 ± 1.8 mV, n = 12; I381T, −62.2 ± 1.4 mV, n = 10; P < 0.01) and steady-state fast-inactivation (wild-type, −53.3 ± 1.7 mV, n = 11; I381T, −40.0 ± 1.6 mV, n = 14; P < 0.001) for wild-type and I381T channels. (D) Voltage-dependence of deactivation time constants for wild-type (n = 12) and I381T (n = 8) channels. Inset: Representative currents recorded from two different neurons expressing wild-type (black) or I381T (red) sodium channels at a repolarization voltage of −70 mV. The wild-type current peak was scaled to match the I381T current peak for comparing the kinetics of current decay. (E) Voltage-dependence of steady-state slow-inactivation for wild-type (−81.9 ± 4.1 mV, n = 5) and I381T channels (−82.4 ± 2.1 mV, n = 6, P > 0.05).

Figure 4

Voltage-clamp analysis of I381T mutant channels. Inward currents recorded from a small-diameter DRG neuron expressing wild-type Nav1.9 (A) or I381T mutant (B) sodium channels. (C) Comparison of the voltage dependence of activation (wild-type, −55.3 ± 1.8 mV, n = 12; I381T, −62.2 ± 1.4 mV, n = 10; P < 0.01) and steady-state fast-inactivation (wild-type, −53.3 ± 1.7 mV, n = 11; I381T, −40.0 ± 1.6 mV, n = 14; P < 0.001) for wild-type and I381T channels. (D) Voltage-dependence of deactivation time constants for wild-type (n = 12) and I381T (n = 8) channels. Inset: Representative currents recorded from two different neurons expressing wild-type (black) or I381T (red) sodium channels at a repolarization voltage of −70 mV. The wild-type current peak was scaled to match the I381T current peak for comparing the kinetics of current decay. (E) Voltage-dependence of steady-state slow-inactivation for wild-type (−81.9 ± 4.1 mV, n = 5) and I381T channels (−82.4 ± 2.1 mV, n = 6, P > 0.05).

The voltage-dependence of activation and steady-state fast-inactivation for wild-type and I381T are illustrated in Fig. 4C. Compared with wild-type, the midpoint of activation (V1/2,act) was shifted 6.9 mV more negative for I381T (wild-type, −55.3 ± 1.8 mV, n = 12; I381T, −62.2 ± 1.4 mV, n = 10; P < 0.01). The slope factor of I381T, however, remained unaffected (wild-type, 7.93 ± 0.61; I381T, 8.49 ± 0.53, P > 0.05). Steady-state fast-inactivation also exhibited a robust change in voltage-dependence. The midpoint voltage (V1/2,fast) of I381T was depolarized by 13.3 mV compared to that of wild-type (wild-type, −53.3 ± 1.7 mV, n = 11; I381T, −40.0 ± 1.6 mV, n = 14; P < 0.001). In contrast, the slope factor of I381T was not significantly different from that of wild-type (wild-type, 10.6 ± 1.6; I381T, 12.1 ± 1.0; P > 0.05). The combinatorial effect of a hyperpolarizing shift of activation and depolarizing shift of steady-state fast-inactivation markedly increases the predicted window current for I381T mutant channel, compared with wild-type channels.

The deactivation properties of I381T are displayed in Fig. 4D. The I381T mutation (n = 8) produced a significant decrease in the rate of current decay across all deactivation potentials tested, compared to wild-type (n = 12). For example, the deactivation time constant at −70 mV was 4.84 ± 0.14 ms (n = 8) for I381T, increasing by 23% compared to that for wild-type (3.93 ± 0.2 6 ms, n = 12, P < 0.05). Representative current traces at a repolarizing potential of −70 mV are presented in the inset of Fig. 4D to illustrate slowed deactivation in I381T transfected DRG neuron.

As a final assessment of the effects of the I381T mutation on voltage-gated sodium currents mediated by Nav1.9, slow-inactivation of the channels was examined using a 50 ms depolarizing pulse at −45 mV after a series of 30 s prepulses from −130 to −30 mV, which are followed by 100 ms pulses to −100 mV to remove fast-inactivation. The peak current at the testing potential of −45 mV was normalized and plotted against the prepulse membrane potential. Slow-inactivation for wild-type and I381T channels are superimposed in Fig. 4E. Fit with a Boltzmann function, the midpoint voltage of slow-inactivation of wild-type was −81.9 ± 4.1 mV (n = 5), with a slope factor of 6.98 ± 0.92 (n = 5). The I381T mutant channel had a midpoint of slow-inactivation of −82.4 ± 2.1 mV (n = 6, P > 0.05), and a slope factor of 7.88 ± 1.1 (n = 6, P > 0.05), neither of which differ significantly from that of wild-type channels.

To assess the effects of I381T on DRG neuronal excitability, we performed current-clamp studies in two groups of rat small DRG neurons transfected with wild-type or I381T channels. There were no significant differences in input resistance (wild-type, 588 ± 54 MΩ; I381T, 673 ± 77 MΩ, P > 0.05), action potential amplitude (wild-type, 97.5 ± 3.3 mV; I381T, 102 ± 3.8 mV, P > 0.05), action potential width at 0 mV (wild-type, 3.99 ± 0.23 ms; I381T, 3.90 ± 0.33 ms, P > 0.05) or voltage threshold at which an all-or-none action potential takes off (wild-type, −18.5 ± 2.0 mV; I381T, −19.6 ± 1.7 mV, P > 0.05) between neurons expressing wild-type (n = 26) or I381T (n = 22) channels. The RMP of neurons expressing I381T, however, was depolarized by 5.5 mV (wild-type, −50.6 ± 1.6 mV; I381T, −45.1 ± 1.4 mV, P < 0.05) (Fig. 5C). Current threshold, at which an all-or-none action potential is elicited, was assessed by injecting a series of currents in increments of 5 pA over 200 ms to neurons expressing either wild-type or I381T channels. There is a nearly 50% reduction of current threshold for single action potentials in neurons expressing I381T (wild-type, 236 ± 30 pA; I381T, 138 ± 23 pA, P < 0.05) (Fig. 5D).

Figure 5

Current-clamp analysis of I381T mutation. (A) Responses of a neuron expressing wild-type or I381T channels to 500 ms depolarizing current steps that are 1×, 1.5× and 2× of the current threshold. (B) Summary of firing frequency data. The total number of action potentials (APs) elicited by the indicated depolarizing current steps from 25 to 500 pA in 25 pA increments was compared between two groups of neurons expressing wild-type (n = 26) or I381T (n = 22) channels. (C) RMP of neurons expressing I381T (−45.1 ± 1.4 mV) was depolarized by 5.5 mV as compared to that of wild-type channels (−50.6 ± 1.6 mV, P < 0.05). (D) Current threshold was significantly reduced after expressing I381T channels (wild-type, 236 ± 30 pA; I381T, 138 ± 23 pA; P < 0.05). (E) Representative recording of spontaneous firing in a DRG neuron expressing I381T mutant channels. Trace was recorded for 30 s without current injection. Inset: Bar graph showing the proportion of spontaneous firing cells for DRG neurons expressing I381T (red) is >5-fold compared to that of wild-type (black); numbers to the right of the bar graph show values for wild-type (10%, lower value in parentheses) and I381T (52%, upper value). *P < 0.001 z-test.

Figure 5

Current-clamp analysis of I381T mutation. (A) Responses of a neuron expressing wild-type or I381T channels to 500 ms depolarizing current steps that are 1×, 1.5× and 2× of the current threshold. (B) Summary of firing frequency data. The total number of action potentials (APs) elicited by the indicated depolarizing current steps from 25 to 500 pA in 25 pA increments was compared between two groups of neurons expressing wild-type (n = 26) or I381T (n = 22) channels. (C) RMP of neurons expressing I381T (−45.1 ± 1.4 mV) was depolarized by 5.5 mV as compared to that of wild-type channels (−50.6 ± 1.6 mV, P < 0.05). (D) Current threshold was significantly reduced after expressing I381T channels (wild-type, 236 ± 30 pA; I381T, 138 ± 23 pA; P < 0.05). (E) Representative recording of spontaneous firing in a DRG neuron expressing I381T mutant channels. Trace was recorded for 30 s without current injection. Inset: Bar graph showing the proportion of spontaneous firing cells for DRG neurons expressing I381T (red) is >5-fold compared to that of wild-type (black); numbers to the right of the bar graph show values for wild-type (10%, lower value in parentheses) and I381T (52%, upper value). *P < 0.001 z-test.

Representative action potentials triggered by current injections 1×, 1.5× and 2× of current threshold in neurons transfected with either wild-type or I381T channels are illustrated in Fig. 5A. The average numbers of action potentials in wild-type and I381T elicited by graded current injections in 25 pA increments are plotted in Fig. 5B. DRG neurons transfected with I381T fire at a significantly higher frequency than DRG neurons transfected with wild-type in response to similar stimuli from 225 pA to 500 pA.

Expression of the I381T channels also led to an increase in the proportion of spontaneously firing DRG neurons. A representative recording showing sustained spontaneous firing from a DRG neuron expressing I381T mutant channels is shown in Fig. 5E. Only 10% (3 of 29 cells) of DRG neurons expressing wild-type channels fired spontaneously. Expression of I381T, in contrast, resulted in >5-fold increase in the proportion of spontaneously firing cells (52%, 24 of 46 cells), significantly more than for wild-type (P < 0.001, z-test).

Depolarization of resting membrane potential and dorsal root ganglion neuron excitability

The L1158P and I381T mutations depolarized RMP of DRG neurons by 3.5 mV and 5.5 mV, respectively. Although most current-clamp studies show that depolarized RMP due to Nav1.7 mutations result in DRG neuron hyperexcitability (Rush et al., 2006; Faber et al., 2012a; Han et al., 2012b), Leipold et al. (2013) have suggested that depolarization can reduce the excitability of DRG neurons. To examine the hypothesis that the depolarization of RMP contributes to the decrease in current threshold for single action potential generation, we measured changes in current thresholds when injected currents were used to hold DRG neurons at membrane potentials 5 mV more depolarized than their RMP, similar to the shift produced by the I381T channel. The RMP for adult rat DRG neurons was −58.6 ± 1.1 mV (n = 14). We found that depolarization of RMP by 5 mV decreased current threshold from 252 ± 46 pA to 189 ± 39 pA (n = 14, P < 0.001, paired t-test). We also performed the converse experiment and found that hyperpolarization of RMP by 5 mV increased current threshold to 319 ± 72 pA (n = 14, P < 0.01, paired t-test). In all cells tested, there is a decrease in current threshold when RMP is depolarized by 5 mV and an increase in current thresholds when RMP is hyperpolarized by 5 mV. On average, a 5 mV depolarization of RMP led to a significant reduction in current threshold of action potential in DRG neurons by 27.2 ± 3.9% while a 5 mV hyperpolarization of RMP resulted in an increase in current threshold by 26.5 ± 3.5% (Fig. 6).

Figure 6

Depolarization of RMP of DRG neurons by 5 mV reduces current threshold for action potential. (A) Representative action potentials recorded from a native adult rat DRG neuron at its RMP (black), at membrane potential depolarized by current injection 5 mV (red) or hyperpolarized by 5 mV (blue). Arrows and numbers indicate current thresholds at different membrane potentials. (B) The current threshold was reduced when the neurons were held at membrane potentials 5 mV depolarized compared to RMP and increased when they were held at membrane potentials 5 mV hyperpolarized compared to RMP. *P < 0.001, paired t-test.

Figure 6

Depolarization of RMP of DRG neurons by 5 mV reduces current threshold for action potential. (A) Representative action potentials recorded from a native adult rat DRG neuron at its RMP (black), at membrane potential depolarized by current injection 5 mV (red) or hyperpolarized by 5 mV (blue). Arrows and numbers indicate current thresholds at different membrane potentials. (B) The current threshold was reduced when the neurons were held at membrane potentials 5 mV depolarized compared to RMP and increased when they were held at membrane potentials 5 mV hyperpolarized compared to RMP. *P < 0.001, paired t-test.

Assessment of physiological levels of I381T by dynamic clamp recordings

Our current-clamp observations indicate that transfection of DRG neurons with mutant Nav1.9 channels renders them hyperexcitable, compared to similar neurons transfected with wild-type Nav1.9 channels. However, transfection of DRG neurons by wild-type or mutant channels does not permit precise control of the level of expression of the transfected channels and, by definition, produces supra-physiological levels of the channel. Given the contention by Leipold et al. (2013) that gain-of-function mutations of Nav1.9 can impair action potential generation in DRG neurons, we decided to use dynamic clamp (Sharp et al., 1993; Kemenes et al., 2011; Samu et al., 2012) to study the effects of expressing wild-type and mutant Nav1.9 at physiological levels. Although currently available dynamic clamp software only permits inclusion of two kinetic particles (and thus is not strictly comparable to current-clamp), it provides the opportunity to assess the effect of the expression of precisely calibrated, physiologically relevant levels of wild-type and mutant Nav1.9 channels on DRG neuron excitability.

In the present study we used dynamic clamp recordings to introduce physiologically relevant, calibrated levels of wild-type and I381T conductances into small DRG neurons. As human subjects carrying the I381T mutation are heterozygous, we started with Nav1.9-null DRG neurons and added back 100% wild-type or 50%/50% mixture of wild-type and I381T current (WT:IT). First we performed in silico evaluation of wild-type and I381T mutant channels. Figure 7A shows computer simulation of current traces obtained from wild-type (blue) and I381T (orange) Nav1.9 channel model INa = gmax × m × h × s × (Vm − ENa) which we used for these experiments. Figure 7C shows the Nav1.9 kinetic model steady-state fast-inactivation (solid), steady-state activation (solid) and steady-state slow-inactivation (dashed) (left panel); activation time constant (middle panel) and inactivation time constant (right panel). The wild-type model is shown in blue and I381T is shown in orange. Computed currents were evoked in response to voltage steps ranging from −90 mV to 60 mV (in 10 mV increments) applied from a −110 mV holding potential. The model cell was equipotential 15 pF sphere with 0.0018 S/cm2 current density. We used spontaneous firing from a small DRG neuron (Fig. 7B, top) as a voltage command to study modelled wild-type (blue) and I381T (orange) Nav1.9 currents associated with the action potential shown (Fig. 7B, bottom).

Figure 7

Dynamic clamp recording of DRG neuron firing using physiological levels of wild-type or mutant Nav1.9 currents. (A) Current traces obtained from wild-type (blue) and I381T (orange) Nav1.9 channel model INa = gmax × m × h × s × (Vm − ENa). Currents were evoked by voltage steps ranged from −90 mV to 60 mV (in 10 mV increment) applied from −110 mV holding potential. The model cell was equipotential 15 pF sphere with 0.0018 S/cm2 current density. (B, top) Repetitive firing of a spontaneously active small DRG neuron. (B, bottom) Modelled wild-type (blue) and I381T (orange) Nav1.9 currents obtained in response to the voltage command shaped in the form of the action potential shown in B (top). (C) Nav1.9 kinetic model steady-state fast inactivation (solid), steady-state activation (solid) and steady-state slow inactivation (dashed) are presented on the left panel; activation time constant and inactivation time constant are shown on the middle and right panel, respectively. Wild-type model is shown in blue and I381T is shown in orange. (D) Dynamic-clamp recordings of action potentials evoked by current pulses of threshold intensity in control (black), after dynamic-clamp addition of 100% wild-type conductance (blue) and (50% wild-type + 50% I381T) conductance (red). Stimulation current pulses are shown on top panel, dynamic-clamp currents are shown on the middle panel and action potentials are presented on the bottom panel. (E) Action potential repetitive firing evoked by a 5 Hz train of current pulses in control (top, black) and after dynamic-clamp addition of 100% wild-type (middle, blue) and (50% wild-type + 50% I381T) conductance (bottom, red). (F) ΔRMP: 9.4 mV ± 0.7 mV (wild-type); 12.7 mV ± 0.7 mV (WT:IT) (P < 0.01, t-test, n = 13). RMP: −61.7 mV ± 0.7 mV (control); −52.3 mV ± 0.8 (wild-type); −49.1 mV ± 0.7 mV (WT:IT); −62.1 mV ± 0.8 mV (control). Wild-type versus WT:IT (P < 0.01, t-test, n = 13). Current threshold change: −37.8% ± 3.3% (wild-type); −48.7% ± 3.5% (WT:IT) (P < 0.05, t-test, n = 13). Current threshold: 116.9 pA ± 9.70 pA (control); 71.5 pA ± 5.6 pA (wild-type); 59.2 pA ± 5.3 pA (WT:IT); 116.2 pA ± 10.0 pA (control). Wild-type versus WT:IT (P > 0.05, t-test, n = 13). Number of action potentials: 4.2 ± 0.6 (wild-type); 6.6 ± 0.7 (WT:IT) (P < 0.05, Mann-Whitney, n = 11).

Figure 7

Dynamic clamp recording of DRG neuron firing using physiological levels of wild-type or mutant Nav1.9 currents. (A) Current traces obtained from wild-type (blue) and I381T (orange) Nav1.9 channel model INa = gmax × m × h × s × (Vm − ENa). Currents were evoked by voltage steps ranged from −90 mV to 60 mV (in 10 mV increment) applied from −110 mV holding potential. The model cell was equipotential 15 pF sphere with 0.0018 S/cm2 current density. (B, top) Repetitive firing of a spontaneously active small DRG neuron. (B, bottom) Modelled wild-type (blue) and I381T (orange) Nav1.9 currents obtained in response to the voltage command shaped in the form of the action potential shown in B (top). (C) Nav1.9 kinetic model steady-state fast inactivation (solid), steady-state activation (solid) and steady-state slow inactivation (dashed) are presented on the left panel; activation time constant and inactivation time constant are shown on the middle and right panel, respectively. Wild-type model is shown in blue and I381T is shown in orange. (D) Dynamic-clamp recordings of action potentials evoked by current pulses of threshold intensity in control (black), after dynamic-clamp addition of 100% wild-type conductance (blue) and (50% wild-type + 50% I381T) conductance (red). Stimulation current pulses are shown on top panel, dynamic-clamp currents are shown on the middle panel and action potentials are presented on the bottom panel. (E) Action potential repetitive firing evoked by a 5 Hz train of current pulses in control (top, black) and after dynamic-clamp addition of 100% wild-type (middle, blue) and (50% wild-type + 50% I381T) conductance (bottom, red). (F) ΔRMP: 9.4 mV ± 0.7 mV (wild-type); 12.7 mV ± 0.7 mV (WT:IT) (P < 0.01, t-test, n = 13). RMP: −61.7 mV ± 0.7 mV (control); −52.3 mV ± 0.8 (wild-type); −49.1 mV ± 0.7 mV (WT:IT); −62.1 mV ± 0.8 mV (control). Wild-type versus WT:IT (P < 0.01, t-test, n = 13). Current threshold change: −37.8% ± 3.3% (wild-type); −48.7% ± 3.5% (WT:IT) (P < 0.05, t-test, n = 13). Current threshold: 116.9 pA ± 9.70 pA (control); 71.5 pA ± 5.6 pA (wild-type); 59.2 pA ± 5.3 pA (WT:IT); 116.2 pA ± 10.0 pA (control). Wild-type versus WT:IT (P > 0.05, t-test, n = 13). Number of action potentials: 4.2 ± 0.6 (wild-type); 6.6 ± 0.7 (WT:IT) (P < 0.05, Mann-Whitney, n = 11).

As shown in Fig. 7D, dynamic-clamp recordings of action potentials evoked by current pulses of threshold intensity in control DRG neurons from Nav1.9 knockout mice with no added Nav1.9 (black), and after dynamic-clamp addition of 100% wild-type conductance (blue) and (50% wild-type + 50% I381T) conductance (red) demonstrate that wild-type Nav1.9 reduces current threshold, with I831T mutant channels reducing threshold even further. Figure 7E shows action potential repetitive firing evoked by a 5 Hz train of current pulses in control (top, black) and after dynamic-clamp addition of 100% wild-type (middle, blue) and (50% wild-type + 50% I381T) conductance (bottom, red), and illustrates increased firing frequency due to expression of the mutant channel.

Finally, we found by dynamic clamp (Fig. 7F) that addition of wild-type Nav1.9 depolarizes DRG RMP by 9.4 mV ± 0.7 mV (wild-type), and that substitution of I381T for 50% of wild-type Nav1.9 depolarizes RMP by12.7 mV ± 0.7 mV. Thus, we found, in a series of 13 Nav1.9-null DRG neurons in which we first measured control RMP (no Nav1.9), then added 100% wild-type Nav1.9, or substituted a 50%/50% mixture of wild-type Nav1.9 and I381T, and finally returned to control conditions (no Nav1.9) and again measured RMP: −61.7 mV ± 0.7 mV (control); −52.3 mV ± 0.8 mV (wild-type); −49.1 mV ± 0.7 mV (WT:IT); −62.1 mV ± 0.8 mV (control). Wild-type versus WT:IT (P < 0.01, t-test, n = 13). Similarly, we found that expression of wild-type Nav1.9 reduces current threshold by −37.8% ± 3.3% (wild-type); with 50% substitution by I381T producing a reduction of −48.7% ± 3.5% (WT:IT) (P < 0.05, t-test, n = 13) [current threshold: 116.9 pA ± 9.70 pA (control); 71.5 pA ± 5.6 pA (wild-type); 59.2 pA ± 5.3 pA (WT:IT); 116.2 pA ± 10.0 pA (control). Wild-type versus WT:IT (P > 0.05, t-test, n = 13)]. The number of action potentials evoked by a 10-pulses stimulus was increased from 4.2 ± 0.6 (wild-type), to 6.6 ± 0.7 (WT:IT) (P < 0.05, Mann-Whitney, n = 11).

Discussion

Our results identified eight mutations of SCN11A, the gene encoding sodium channel Nav1.9, in 12 individuals from a cohort of 345 individuals with painful peripheral neuropathy, who did not carry mutations of SCN9A or SCN10A. Four of these mutations substituted highly conserved amino acids in membrane-spanning segments of the channel, while three were located within cytoplasmic loops and the C-terminus, and one affected the 3’ splice acceptor site of intron 24. We assessed two of these mutations by voltage- and current-clamp, and studied one of these using dynamic-clamp, and found that they confer gain-of-function attributes on Nav1.9 and are proexcitatory in DRG neurons. Similar to Nav1.7 and Nav1.8, Nav1.9 is expressed by DRG neurons with a preferential distribution in small-diameter, non-peptidergic neurons (Dib-Hajj et al., 1998, 2002), most of which are nociceptors (Fang et al., 2002), and has been found within free nerve terminals in the skin and cornea (Black and Waxman, 2002; Dib-Hajj et al., 2002; Persson et al., 2010), as well as nodose ganglion neurons (Schild and Kunze, 1997; Matsumoto et al., 2007) including aortic baroreceptors (Qiao et al., 2009) and visceral afferent neurons that innervate the intestine (Rugiero et al., 2003; Padilla et al., 2007) and, under some conditions, the urinary bladder (Black et al., 2003). Thus gain-of-function mutations in this channel might be expected to contribute to pain, autonomic dysfunction and axonal degeneration in patients with peripheral neuropathy.

In addition to prominent complaints of numbness, tingling and pain in the distal extremities, the patients noted a number of complaints usually associated with autonomic dysfunction, including a sensation of dry eyes, diarrhoea, urinary problems, palpitations, and orthostatic dizziness. Consistent with these clinical complaints, Nav1.9 channels are present within free nerve terminals in the cornea (Black and Waxman, 2002) and it has been proposed that hyperactivity of corneal nociceptors can trigger a sensation of ocular dryness (Belmonte and Gallar, 2011; Rosenthal and Borsook, 2012). Nav1.9 is also present within nodose ganglion neurons (Schild and Kunze, 1997; Matsumoto et al., 2007) including aortic baroreceptors (Qiao et al., 2009) and visceral afferent neurons innervating the intestine (Rugiero et al., 2003; Padilla et al., 2007) and, in some injury models, the urinary bladder (Black et al., 2003). Although the effects of mutant Nav1.9 channels in these cell-types has not be studied, it is possible that altered patterns of activity in these neurons contribute to the autonomic complaints in the patients we describe

Although families were not available for study, and thus pathogenicity cannot be inferred from segregation of phenotype with carriers of these alleles, each of the two functionally-profiled mutations was identified in two independent patients. The rare occurrence in control populations and genomic databases and the gain-of-function attributes demonstrated by our functional characterization strongly suggest that these mutations are pathogenic. The two substitutions produced pro-excitatory changes in the gating properties of the mutant Nav1.9 channels as revealed by voltage-clamp recordings in DRG neurons where this channel is normally expressed. By current-clamp, we found that both mutations reduce current threshold and increase firing frequency in response to suprathreshold stimuli, which would be expected to lower threshold for, or increase intensity of, evoked pain. We also observed that the I381T mutation induced spontaneous firing of small DRG neurons, which include nociceptors, and would be expected to contribute to spontaneous pain.

Previous studies of DRG neuron excitability via current-clamp recording has been conducted following the expression of mutant sodium channels, compared to the excitability of neurons expressing the corresponding wild-type channel (for a recent review, see Dib-Hajj et al., 2013). However, these studies cannot control for the level of expression of the recombinant channel, which is generally added to the full complement of the endogenous sodium currents in the transfected neurons. In an effort to study the effect of a mutant Nav1.9 channel on the excitability of DRG neurons using physiological levels of the mutant current in relation to the endogenous Nav1.9 current we used dynamic clamp (Sharp et al., 1993; Kemenes et al., 2011; Samu et al., 2012). Using this approach, we showed that substituting 50% of wild-type Nav1.9 current by the corresponding I381T current (a manoeuvre that mimics the presence of one mutant allele) causes a significant depolarization of the RMP neurons and induces hyperexcitability of DRG neurons, providing additional support for the contention that the nociceptors of an individual heterozygous for this mutation can become hyperexcitable, consistent with the pain phenotype.

Our findings are consistent with those of a recent study that linked gain-of-function mutations in Nav1.9 to familial episodic pain (Zhang et al., 2013). Zhang et al. (2013) reported that the mutations R225C and A808G, which segregated with pain phenotype in individuals from two multi-generation families, cause an increase in the current density of Nav1.9 without affecting the channel’s gating properties, and increased excitability of DRG neurons expressing the mutant channel without affecting RMP of these neurons. Together with the current results, this study establishes a link between Nav1.9 mutations, which increase current density or confer gain-of-function attributes onto gating properties of the Nav1.9 channel and painful disorders, including the rare disease familial episodic pain as shown by Zhang et al. (2013), and the more common disorder painful peripheral neuropathy as shown here.

In contrast with our observations on painful neuropathies, and those of Zhang et al. (2013) in familial episodic pain, another study reported that a gain-of-function mutation of Nav1.9 causes a form of congenital insensitivity to pain (Leipold et al., 2013). Leipold et al. (2013) reported that the de novo mutation L811P in Nav1.9, identified in two unrelated individuals with this form of congenital insensitivity to pain using whole exome sequencing, causes a −28 mV shift in voltage-dependence of activation, leading to a depolarization of ∼6.7 mV of RMP of DRG neurons; they argued that this depolarization causes resting inactivation of most voltage-gated sodium channels in these neurons and thus leads to hypoexcitability, consistent with the phenotype of pain insensitivity. However, as shown here and in many studies (Rush et al., 2006; Faber et al., 2012a, b; Han et al., 2012b), depolarization of RMP of DRG neurons causes hyperexcitability of these neurons. Additionally, specifically mimicking a 5 mV depolarization in RMP produced by the Nav1.7 A863P erythromelalgia mutation or the Nav1.9 I381T by direct current injection into DRG neurons, instead of that produced by the transfection of mutant channels, leads to hyperexcitability of DRG neurons (Harty and Waxman, 2007) (Fig. 6). Thus, the hypoexcitability of mouse DRG neurons that express the Nav1.9 L811P mutation cannot be explained by the shift in RMP of these neurons.

Hyperexcitability of small DRG neurons, most of which are nociceptors, upon depolarization of their RMP in the range of 5–6 mV is due to the presence in these cells of Nav1.8, a channel that is unique by possessing activation and steady-state inactivation with voltage-dependencies that are depolarized by 20–40 mV with respect to other voltage-gated sodium channels (Catterall et al., 2005). Indeed, depolarization of the RMP of neurons that do not produce Nav1.8, for example sympathetic neurons of superior cervical ganglion, causes hypoexcitability of these neurons (Rush et al., 2006; Han et al., 2012b), which can be reversed by expressing Nav1.8 in these neurons (Rush et al., 2006). The pain insensitivity phenotype reported by Leipold et al. (2013) might have been explained by a lack of Nav1.8 in the DRG neurons that they studied, but this seems unlikely since Nav1.8 is expressed in >90% of small DRG neurons in the mouse (Shields et al., 2012). An alternative possibility is that extreme hyperexcitability of Nav1.9-expressing DRG neurons, due to the −28 mV shift in voltage-dependence of activation produced by the Nav1.9 L811P mutation, leads to sustained subnormality or fatigue of this group of cells, while leaving the population of hypoexcitable DRG neurons that do not express Nav1.8 intact, but Leipold et al. (2013) did not present any evidence for this type of functional abnormality.

Our results confirm the observation that gain-of-function mutations of Nav1.9 can produce painful syndromes (Zhang et al., 2013) and extend the link between gain-of-function mutations in Nav1.9 and pain from rare genetic model diseases to a more common disorder, painful peripheral neuropathy. Together with our earlier studies linking gain-of-function mutations of Nav1.7 (Estacion et al., 2011; Faber et al., 2012a; Han et al., 2014; Han et al., 2012a, b; Hoeijmakers et al., 2012b) and Nav1.8 (Faber et al., 2012b; Huang et al., 2013) and peripheral neuropathy, the present results suggest the potential for treatment of pain associated with peripheral neuropathy via the targeting of peripheral sodium channels.

Acknowledgements

We thank Lawrence Macala, Peng Zhao, Shujun Liu, Fadia Dib-Hajj, Palak Shah, Ilse Driesmans, Aline Kosten, Linda Meekels and Diane Merckx for their technical assistance in this project.

Funding

This work was supported in part by grants from the Rehabilitation Research Service and Medical Research Service, Department of Veterans Affairs (SGW and SDH), and grant #602273 from the European Union Seventh Framework Programme FP7/2007-2013. The Center for Neuroscience & Regeneration Research is a Collaboration of the Paralyzed Veterans of America with Yale University.

Abbreviations

    Abbreviations
  • DRG

    dorsal root ganglion

  • RMP

    resting membrane potential

  • TTX

    tetrodotoxin

References

Ahn
HS
Vasylyev
DV
Estacion
M
Macala
LJ
Shah
P
Faber
CG
, et al.  . 
Differential effect of D623N variant and wild-type Na1.7 sodium channels on resting potential and interspike membrane potential of dorsal root ganglion neurons
Brain Res
 , 
2013
, vol. 
1529
 (pg. 
165
-
77
)
Atkins
JF
Wills
NM
Loughran
G
Wu
CY
Parsawar
K
Ryan
MD
, et al.  . 
A case for “StopGo": reprogramming translation to augment codon meaning of GGN by promoting unconventional termination (Stop) after addition of glycine and then allowing continued translation (Go)
RNA
 , 
2007
, vol. 
13
 (pg. 
803
-
10
)
Baker
MD
Chandra
SY
Ding
Y
Waxman
SG
Wood
JN
GTP-induced tetrodotoxin-resistant Na+ current regulates excitability in mouse and rat small diameter sensory neurones
J Physiol
 , 
2003
, vol. 
548
 
Pt 2
(pg. 
373
-
82
)
Belmonte
C
Gallar
J
Cold thermoreceptors, unexpected players in tear production and ocular dryness sensations
Invest Ophthalmol Vis Sci
 , 
2011
, vol. 
52
 (pg. 
3888
-
92
)
Black
JA
Cummins
TR
Yoshimura
N
de Groat
WC
Waxman
SG
Tetrodotoxin-resistant sodium channels Na(v)1.8/SNS and Na(v)1.9/NaN in afferent neurons innervating urinary bladder in control and spinal cord injured rats
Brain Res
 , 
2003
, vol. 
963
 (pg. 
132
-
8
)
Black
JA
Frezel
N
Dib-Hajj
SD
Waxman
SG
Expression of Nav1.7 in DRG neurons extends from peripheral terminals in the skin to central preterminal branches and terminals in the dorsal horn
Mol Pain
 , 
2012
, vol. 
8
 pg. 
82
 
Black
JA
Waxman
SG
Molecular identities of two tetrodotoxin-resistant sodium channels in corneal axons
Exp Eye Res
 , 
2002
, vol. 
75
 (pg. 
193
-
9
)
Catterall
WA
Goldin
AL
Waxman
SG
International Union of Pharmacology. XLVII. Nomenclature and structure-function relationships of voltage-gated sodium channels
Pharmacol Rev
 , 
2005
, vol. 
57
 (pg. 
397
-
409
)
Cummins
TR
Dib-Hajj
SD
Black
JA
Akopian
AN
Wood
JN
Waxman
SG
A novel persistent tetrodotoxin-resistant sodium current in SNS-null and wild-type small primary sensory neurons
J Neurosci
 , 
1999
, vol. 
19
 pg. 
RC43
 
Dib-Hajj
S
Black
JA
Cummins
TR
Waxman
SG
NaN/Nav1.9: a sodium channel with unique properties
Trends Neurosci
 , 
2002
, vol. 
25
 (pg. 
253
-
9
)
Dib-Hajj
SD
Choi
JS
Macala
LJ
Tyrrell
L
Black
JA
Cummins
TR
, et al.  . 
Transfection of rat or mouse neurons by biolistics or electroporation
Nat Protoc
 , 
2009
, vol. 
4
 (pg. 
1118
-
26
)
Dib-Hajj
SD
Cummins
TR
Black
JA
Waxman
SG
Sodium channels in normal and pathological pain
Annu Rev Neurosci
 , 
2010
, vol. 
33
 (pg. 
325
-
47
)
Dib-Hajj
SD
Tyrrell
L
Black
JA
Waxman
SG
NaN, a novel voltage-gated Na channel, is expressed preferentially in peripheral sensory neurons and down-regulated after axotomy
Proc Natl Acad Sci USA
 , 
1998
, vol. 
95
 (pg. 
8963
-
8
)
Dib-Hajj
SD
Tyrrell
L
Cummins
TR
Black
JA
Wood
PM
Waxman
SG
Two tetrodotoxin-resistant sodium channels in human dorsal root ganglion neurons
FEBS Lett
 , 
1999
, vol. 
462
 (pg. 
117
-
20
)
Dib-Hajj
SD
Yang
Y
Black
JA
Waxman
SG
The NaV1.7 sodium channel: from molecule to man
Nat Rev Neurosci
 , 
2013
, vol. 
14
 (pg. 
49
-
62
)
Estacion
M
Han
C
Choi
JS
Hoeijmakers
JG
Lauria
G
Drenth
JP
, et al.  . 
Intra- and interfamily phenotypic diversity in pain syndromes associated with a gain-of-function variant of NaV1.7
Mol Pain
 , 
2011
, vol. 
7
 pg. 
92
 
Faber
CG
Hoeijmakers
JG
Ahn
HS
Cheng
X
Han
C
Choi
JS
, et al.  . 
Gain of function NaV1.7 mutations in idiopathic small fiber neuropathy
Ann Neurol
 , 
2012a
, vol. 
71
 (pg. 
26
-
39
)
Faber
CG
Lauria
G
Merkies
IS
Cheng
X
Han
C
Ahn
HS
, et al.  . 
Gain-of-function Nav1.8 mutations in painful neuropathy
Proc Natl Acad Sci USA
 , 
2012b
, vol. 
109
 (pg. 
19444
-
9
)
Fang
X
Djouhri
L
Black
JA
Dib-Hajj
SD
Waxman
SG
Lawson
SN
The presence and role of the tetrodotoxin-resistant sodium channel Nav1.9 (NaN) in nociceptive primary afferent neurons
J Neurosci
 , 
2002
, vol. 
22
 (pg. 
7425
-
33
)
Han
C
Hoeijmakers
JG
Ahn
HS
Zhao
P
Shah
P
Lauria
G
, et al.  . 
NaV1.7-related small fiber neuropathy: impaired slow-inactivation and DRG neuron hyperexcitability
Neurology
 , 
2012a
, vol. 
78
 (pg. 
1635
-
43
)
Han
C
Hoeijmakers
JG
Liu
S
Gerrits
MM
Te Morsche
RH
Lauria
G
, et al.  . 
Functional profiles of SCN9A variants in dorsal root ganglion neurons and superior cervical ganglion neurons correlate with autonomic symptoms in small fibre neuropathy
Brain
 , 
2012b
, vol. 
135
 
Pt 9
(pg. 
2613
-
28
)
Han
C
Vasylyev
D
Macala
LJ
Gerrits
MM
Hoeijmakers
JG
Bekelaar
KJ
, et al.  . 
The G1662S NaV1.8 mutation in small fibre neuropathy: impaired inactivation underlying DRG neuron hyperexcitability
J Neurol Neurosurg Psychiatry
 , 
2014
, vol. 
85
 (pg. 
499
-
505
)
Harty
TP
Waxman
SG
Inactivation properties of sodium channel Nav1.8 maintain action potential amplitude in small DRG neurons in the context of depolarization
Mol Pain
 , 
2007
, vol. 
3
 pg. 
12
 
Herzog
RI
Cummins
TR
Waxman
SG
Persistent TTX-resistant Na+ current affects resting potential and response to depolarization in simulated spinal sensory neurons
J Neurophysiol
 , 
2001
, vol. 
86
 (pg. 
1351
-
64
)
Hoeijmakers
JG
Faber
CG
Lauria
G
Merkies
IS
Waxman
SG
Small-fibre neuropathies–advances in diagnosis, pathophysiology and management
Nat Rev Neurol
 , 
2012a
, vol. 
8
 (pg. 
369
-
79
)
Hoeijmakers
JG
Han
C
Merkies
IS
Macala
LJ
Lauria
G
Gerrits
MM
, et al.  . 
Small nerve fibres, small hands and small feet: a new syndrome of pain, dysautonomia and acromesomelia in a kindred with a novel NaV1.7 mutation
Brain
 , 
2012b
, vol. 
135
 
Pt 2
(pg. 
345
-
58
)
Huang
J
Yang
Y
Zhao
P
Gerrits
MM
Hoeijmakers
JG
Bekelaar
K
, et al.  . 
Small-fiber neuropathy Nav1.8 mutation shifts activation to hyperpolarized potentials and increases excitability of dorsal root ganglion neurons
J Neurosci
 , 
2013
, vol. 
33
 (pg. 
14087
-
97
)
Kemenes
I
Marra
V
Crossley
M
Samu
D
Staras
K
Kemenes
G
, et al.  . 
Dynamic clamp with StdpC software
Nat Protoc
 , 
2011
, vol. 
6
 (pg. 
405
-
17
)
Klugbauer
N
Lacinova
L
Flockerzi
V
Hofmann
F
Structure and functional expression of a new member of the tetrodotoxin-sensitive voltage-activated sodium channel family from human neuroendocrine cells
EMBO J
 , 
1995
, vol. 
14
 (pg. 
1084
-
90
)
Lauria
G
Merkies
IS
Faber
CG
Small fibre neuropathy
Curr Opin Neurol
 , 
2012
, vol. 
25
 (pg. 
542
-
9
)
Leipold
E
Liebmann
L
Korenke
GC
Heinrich
T
Giesselmann
S
Baets
J
, et al.  . 
A de novo gain-of-function mutation in SCN11A causes loss of pain perception
Nat Genet
 , 
2013
, vol. 
45
 (pg. 
1399
-
404
)
Luke
GA
de Felipe
P
Lukashev
A
Kallioinen
SE
Bruno
EA
Ryan
MD
Occurrence, function and evolutionary origins of ‘2A-like' sequences in virus genomes
J Gen Virol
 , 
2008
, vol. 
89
 
Pt 4
(pg. 
1036
-
42
)
Matsumoto
S
Yoshida
S
Ikeda
M
Tanimoto
T
Saiki
C
Takeda
M
, et al.  . 
Effect of 8-bromo-cAMP on the tetrodotoxin-resistant sodium (Nav 1.8) current in small-diameter nodose ganglion neurons
Neuropharmacology
 , 
2007
, vol. 
52
 (pg. 
904
-
24
)
Ostman
JA
Nassar
MA
Wood
JN
Baker
MD
GTP up-regulated persistent Na+ current and enhanced nociceptor excitability require NaV1.9
J Physiol
 , 
2007
, vol. 
586
 (pg. 
1077
-
87
)
Padilla
F
Couble
ML
Coste
B
Maingret
F
Clerc
N
Crest
M
, et al.  . 
Expression and localization of the Nav1.9 sodium channel in enteric neurons and in trigeminal sensory endings: implication for intestinal reflex function and orofacial pain
Mol Cell Neurosci
 , 
2007
, vol. 
35
 (pg. 
138
-
52
)
Persson
AK
Black
JA
Gasser
A
Fischer
T
Waxman
SG
Sodium-calcium exchanger and multiple sodium channel isoforms in intra-epidermal nerve terminals
Mol Pain
 , 
2010
, vol. 
6
 pg. 
84
 
Qiao
GF
Li
BY
Zhou
YH
Lu
YJ
Schild
JH
Characterization of persistent TTX-R Na+ currents in physiological concentration of sodium in rat visceral afferents
Int J Biol Sci
 , 
2009
, vol. 
5
 (pg. 
293
-
7
)
Rosenthal
P
Borsook
D
The corneal pain system. Part I: the missing piece of the dry eye puzzle
Ocul Surf
 , 
2012
, vol. 
10
 (pg. 
2
-
14
)
Rugiero
F
Mistry
M
Sage
D
Black
JA
Waxman
SG
Crest
M
, et al.  . 
Selective expression of a persistent tetrodotoxin-resistant Na+ current and NaV1.9 subunit in myenteric sensory neurons
J Neurosci
 , 
2003
, vol. 
23
 (pg. 
2715
-
25
)
Rush
AM
Dib-Hajj
SD
Liu
S
Cummins
TR
Black
JA
Waxman
SG
A single sodium channel mutation produces hyper- or hypoexcitability in different types of neurons
Proc Natl Acad Sci USA
 , 
2006
, vol. 
103
 (pg. 
8245
-
50
)
Rush
AM
Waxman
SG
PGE2 increases the tetrodotoxin-resistant Nav1.9 sodium current in mouse DRG neurons via G-proteins
Brain Res
 , 
2004
, vol. 
1023
 (pg. 
264
-
71
)
Ryan
MD
Drew
J
Foot-and-mouth disease virus 2A oligopeptide mediated cleavage of an artificial polyprotein
EMBO J
 , 
1994
, vol. 
13
 (pg. 
928
-
33
)
Samu
D
Marra
V
Kemenes
I
Crossley
M
Kemenes
G
Staras
K
, et al.  . 
Single electrode dynamic clamp with StdpC
J Neurosci Methods
 , 
2012
, vol. 
211
 (pg. 
11
-
21
)
Schild
JH
Kunze
DL
Experimental and modeling study of Na+ current heterogeneity in rat nodose neurons and its impact on neuronal discharge
J Neurophysiol
 , 
1997
, vol. 
78
 (pg. 
3198
-
209
)
Sharp
AA
O'Neil
MB
Abbott
LF
Marder
E
Dynamic clamp: computer-generated conductances in real neurons
J Neurophysiol
 , 
1993
, vol. 
69
 (pg. 
992
-
5
)
Shields
SD
Ahn
HS
Yang
Y
Han
C
Seal
RP
Wood
JN
, et al.  . 
NaV1.8 expression is not restricted to nociceptors in mouse peripheral nervous system
Pain
 , 
2012
, vol. 
153
 (pg. 
2017
-
30
)
Tate
S
Benn
S
Hick
C
Trezise
D
John
V
Mannion
RJ
, et al.  . 
Two sodium channels contribute to the TTX-R sodium current in primary sensory neurons
Nat Neurosci
 , 
1998
, vol. 
1
 (pg. 
653
-
5
)
Tesfaye
S
Boulton
AJ
Dyck
PJ
Freeman
R
Horowitz
M
Kempler
P
, et al.  . 
Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity, and treatments
Diabetes Care
 , 
2010
, vol. 
33
 (pg. 
2285
-
93
)
Vanoye
CG
Kunic
JD
Ehring
GR
George
AL
Jr
Mechanism of sodium channel NaV1.9 potentiation by G-protein signaling
J Gen Physiol
 , 
2013
, vol. 
141
 (pg. 
193
-
202
)
Zhang
XY
Wen
J
Yang
W
Wang
C
Gao
L
Zheng
LH
, et al.  . 
Gain-of-Function mutations in SCN11A cause familial episodic pain
Am J Hum Genet
 , 
2013
, vol. 
93
 (pg. 
957
-
66
)

Author notes

*These authors contributed equally to this work.