Heterologous expression of the human wild-type and variant Na_V1.8 (A1073V) in rat sensory neurons

Abstract

Background:

Silent inflammatory bowel disease (IBD) is a condition in which individuals with the active disease experience minor to no pain. Voltage-gated Na⁺ (Na_V) channels expressed in sensory neurons play a major role in pain perception. Previously, we reported that a Na_V1.8 genetic polymorphism (A1073V, rs6795970) was more common in a cohort of silent IBD patients. The expression of this variant (1073V) in rat sympathetic neurons activated at more depolarized potentials when compared to the more common variant (1073A). In this study, we investigated whether expression of either Na_V1.8 variant in rat sensory neurons would exhibit different biophysical characteristics than previously observed in sympathetic neurons.

Methods.

Endogenous Na_V1.8 channels were first silenced in DRG neurons and then either 1073A or 1073V human Na_V1.8 cDNA constructs were transfected. Na_V1.8 currents were recorded with the whole-cell patch-clamp technique.

Key Results.

The results indicate that 1073A and 1073V Na_V1.8 channels exhibited similar activation values. However, the slope factor (k) for activation determined for this same group of neurons decreased by 5 mV, suggesting an increase in voltage sensitivity. Comparison of inactivation parameters indicated that 1073V channels were shifted to more depolarized potentials than 1073A-expressing neurons, imparting a proexcitatory characteristic.

Conclusions & Inferences.

These findings differ from previous observations in other expression models and underscore the challenges with heterologous expression systems. Therefore, the use of human sensory neurons derived from induced pluripotent stem cells may help address these inconsistencies and better determine the effect of the polymorphism present in IBD patients.

Graphical Abstract

Introduction

Voltage-gated Na⁺ channels (Na_V) are well known to play a crucial role in cellular electrical signaling, and are expressed in both excitable and non-excitable cell types. Nine Na_V channel subtypes (Na_V1.1–1.9) have been identified, and each exhibits different kinetics for Na⁺ conduction, play specific roles in action potential production and propagation, and differs discretely in preferred expression patterns [1–4]. Of these, Na_V1.7, Na_V1.8, and Na_V1.9 are expressed primarily in the peripheral nervous system [4] and preferentially expressed in nociceptive neurons, including those projecting to the gastrointestinal system [5–10]. The electrophysiological properties of these nociceptive neurons suggest that the Na_V1.8 subtype, expressed mainly in small-diameter dorsal root ganglion (DRG) sensory neurons, is especially important for their normal function [11]. Na_V1.8 plays a key role in regulating and transmitting pain from the viscera by producing the upstroke of the action potential in these neurons [12]. Several studies have been carried out to assess the nociceptive response in animal models of colitis. The associated visceral pain caused by colitis was reduced when using either therapeutics [13] or genetic knockouts (KO) that targeted Na_V1.8 expression [14].

Several polymorphisms, found in the genes encoding Na_V1.7 (Scn9A), Na_V1.8 (Scn10A), and Na_V1.9 (Scn11A) have been associated with significant alterations in pain perception. Some result in “loss-of-function” phenotypes, whereas others have been associated with “gain-of-function” changes [15–21]. For instance, a loss-of-function phenotype is seen in the Na_V1.8 variant, rs6795970. This variant leads to a substitution from alanine (A) to valine (V) at position 1073 (A1073V), and carrier patients have been reported to exhibit reduced mechanical pain sensitivity [19]. Further, the heterologous expression of this variant in Na_V1.8 KO mouse DRG has been shown to exhibit an acceleration in inactivation, as well as a decrease in repetitive firing of the neurons following prolonged depolarizing stimuli [19]. In a previous study, we found that homozygosity for this Na_V1.8 polymorphism (1073V) is observed more commonly in patients with hypoalgesic, or “silent” inflammatory bowel disease (IBD) [20]. Notably, up to one third of IBD patients with overt intestinal inflammation do not experience any significant pain sensation and live with the “silent” disease. Silent IBD patients are also twice as likely to be hospitalized and to develop serious complications, including intra-abdominal fistulae [20, 22].

Similar to other groups [19, 23–24], we previously compared the electrophysiological properties of both the wild-type and variant Na_V1.8 channels. However, we employed rat superior cervical ganglion (SCG) neurons as our expression system [25]. Our results showed that the variant channel exhibited a shift of activation and peak potentials towards more depolarizing voltage potentials, suggesting that these neurons would be hypoexcitable. Although the steady-state inactivation characteristics for both channel subtypes were not overtly different [25], a separate study reported that these properties were shifted to more positive potentials for 1073V-expressing Neuro 2A cells when compared to cells transiently transfected with 1073A [24]. In the present study, we examined electrophysiological properties of the wild-type (1073A) and variant (1073V) Na_V1.8 heterologously expressed in rat DRG L₄-L₆ neurons (a sensory neuron model innervating the colon). Initially, the natively expressed Na_V1.8 channels were silenced. Thereafter, each channel subtype was transfected and the biophysical properties were determined.

Materials & Methods

Rat dorsal root ganglion (DRG) tissue isolation

All studies followed the National Institutes of Health guidelines and were approved by the Institutional Animal Care and Use Committee (IACUC) at the Pennsylvania State University College of Medicine (protocol number PROTO201901097. In this study, adult male Sprague-Dawley rats were employed and housed in cages under 12 h light/dark cycles. All were provided with environmental enrichment tubes, water and standard rat chow ad libitum. In some rats, the DRG neurons innervating the colon were retrogradely-labeled as follows. Seven days prior to DRG tissue isolation, the rats were anesthetized with isoflurane (3–5%) and approximately 30 μl of 3% 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate (DiI, in DMSO) were injected into the descending colon muscle. The incision was then closed with absorbable sutures (Vicryl 40). For post-operative pain relief, buprenorphine (1 mg/kg) was administered intraperitoneally. Additionally, the animals received daily antibiotics (Enrofloxacin, 10 mg/kg) for 5 days post-surgery to reduce the risk of surgical infection.

DRG transfection

On the day that DRG tissue was isolated and transfected (Figure 1), the rats were euthanized with CO₂ and quickly decapitated with a laboratory guillotine. The lumbar (L₄-L₆) DRG were dissected, cleared of connective tissue in ice-cold Hank’s balanced salt solution (HBSS, Thermo Fisher Scientific, Waltham, MA), and multiple parallel slits were then made perpendicular to the long axis.

Figure 1.

Schematic representation of DRG tissue transfection with siRNA, shRNA, and cDNA constructs in this study. The DRG tissue was isolated and transfected with either siRNA nucleotides or shRNA cDNA on Day 1 and Day 3. On Day 5, the tissue was transfected with either 1073A or 1073V cDNA constructs. The tissue was dissociated on Day 6 and electrophysiological assays were performed on Day 7. The media was replaced every 24 hr.

Silencing of native Na_V1.8 in DRG tissue was carried out by employing either of the following approaches (Day 1, Figure 1). In the first procedure, the tissue was transfected with small interference RNA (siRNA) nucleotides employing both electroporation and lipofection immediately following tissue isolation, as previously described [26]. The DRG tissue was first electroporated with the NEON Electroporator System (Thermo Fisher Scientific) in the T solution^® (provided with the electroporation kit) containing 4 siRNA sequences designed to silence rat Na_V1.8 and 2 mM 2,3-butanedione monoxime (BDM). For control experiments, we employed a scrambled siRNA sequence. The DRG ganglia were electroporated with three 20 millisecond (ms) 1000-volt pulses. Thereafter, the DRG tissue was placed in a 22 mm dish that contained either scrambled or Na_V1.8 siRNA (2 mM), BDM (2 mM) and 10 μl Lipofectamine 2000 (Thermo Fisher Scientific) in a final volume of 1 ml, and placed for 4–5 hr in a humidified incubator (5% CO₂/95% air) at 37^oC. After the incubation period in siRNA nucleotides, the tissue was then rinsed three times with minimal essential media (MEM, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS), 1% penicillin-streptomycin, and 1% glutamine and then returned to the incubator in the supplemented MEM. This protocol was repeated 48 hr after the initial transfection (Day 3, Figure 1). The siRNA sequences targeting Na_V1.8 were designed employing a macro written by Stephen R. Ikeda (National Institute on Alcohol Abuse and Alcoholism) on IGOR Pro (Wavemetrics, Inc.) and chosen based on criteria described previously [26–27]. The rat Na_V1.8 target sequences were: 5-CAT TCA TGG TGT TGA ATA A-3 corresponding to nucleotide position 272–290; 5-CAA TGG AGA TGG CCT TCA A-3 corresponding to nucleotide position 2126–2144; 5-AGA ACA ACT TGT ACA TGT A-3 corresponding to nucleotide position 4172–4190; and 5-TCA TCG TGG TCA ACA TGT A-3 corresponding to nucleotide position 5138–5156.

For the second transfection approach, the DRG tissue was transfected on Day 1 and Day 3 (Figure 1) with a mammalian shRNA knockdown vector (Vector Builder, Inc., Chicago, IL) coding for 3 shRNA nucleotides designed to silence Na_V1.8. The 3 sequences were: AGA ACA ACT TGT ACA TGT ACT CGA GTA CAT GTA CAA GTT GTT CT; CAT TCA TGG TGT TGA ATA ACT CGA GTT ATT CAA CAC CAT GAA TG; and TCA TCG TGG TCA ACA TGT ACT CGA GTA CAT GTT GAC CAC GAT GA. The expression for each shRNA was driven by the human RNA polymerase III U6. A separate vector was employed that contained a scrambled shRNA sequence (Vector Builder, Inc.). The DRG tissue was transfected with either vector employing lipofection as described above, though the amount of vector in the transfection solution ranged from 10–20 μg.

Two days prior to cell dissociation (Day 5, Figure 1), the DRG tissue was transfected via lipofection with cDNA plasmids coding for either human wild-type (1073A) or variant Na_V1.8 (1073V). Both Na_V1.8 inserts (NCBI No. NM 006514.3) were cloned in the pcDNA3.1 vector via the Xho1 and EcoR1 sites as described previously [25]. On day 6, the transfected DRG tissue was enzymatically dissociated in a culture flask with non-supplemented MEM or Opti-MEM (Thermo Fisher Scientific) containing 0.5 mg/mL collagenase D (Roche, Mannheim, Germany), 0.4 mg/mL trypsin (Worthington Biochemical, Lakewood, NJ), and 0.1 mg/mL DNase (Millipore-Sigma, St. Louis, MO). The culture flask was placed in a shaking water bath at 35°C for 40–45 minutes. Afterward, the DRG neurons were dispersed by vigorous shaking for 10 sec and centrifuged twice for 6 min at 44 g. The DRG neurons were then resuspended in MEM which was supplemented with 10% fetal bovine serum, 1% glutamine, and 1% penicillin-streptomycin. Finally, the DRG neurons were plated onto 35-mm poly-L-lysine coated dishes and stored in humidified incubator supplied with 5% CO₂ and 95% air at 37°C.

Quantitative RT-PCR (QRT-PCR) assays

Whole rat DRGs (L₄-L₆) transfected with either siRNA nucleotides or shRNA vector were isolated (Day 6, Figure 1) and pulverized in room temperature lysis buffer containing the reducing agent β-Mercaptoethanol (Millipore-Sigma), then stored at −80°C until the RNA was isolated and purified using the Nucleospin RNA/Protein kit (Macherey-Nagel). RNA samples were quantified using a Nanodrop ND-8000 instrument (Thermo Fisher Scientific), and cDNA was made from 200 ng of RNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). qRT-PCR was performed using Taqman Gene Expression Master Mix, Taqman primers for GAPDH (cat. # Rn01775763_g1) and Na_V1.8 (cat. # Rn00568393_m1) gene sequences and the QuantStudio 12K Flex Real-Time PCR System (all from Thermo Fisher Scientific).

Western blotting analysis

Protein lysates of transfected DRGs (L₄-L₆) were prepared in M-PER lysis buffer (Thermo Fisher Scientific) supplemented with a protease inhibitor cocktail (PIC, Millipore Sigma) on Day 6 (Figure 1). Specifically, isolated DRGs were weighed in a 1.5 mL centrifuge tube on a microbalance balance, then 40–50 μl of M-PER/PIC was added to the tube and the ganglia were pulverized using a plastic pestle designed to fit snugly in the bottom of the tube (Kimble, cat. # 9749520000). After thoroughly grinding the tissue, additional M-PER/PIC was mixed in to achieve a ratio of 10 μl of lysis buffer per 1 mg of tissue, and the tube was placed in a shaker at 900 RPM for 10 min at room temperature. Following incubation in the shaker, insoluble lysis products were pelleted by centrifugation at >14,000 × g for 15 min, then the supernatant was collected and stored at −80°C until use.

Western blotting was performed using the “Wes” Simple Western system (Protein Simple). On the day of the assay, protein samples were quantified using the Qubit 4.0 Fluorometer and the Qubit Protein Assay Kit (both from Thermo Fisher Scientific). Lyophilized master mix provided in the 66 – 440 kDa Simple Western Separation Module (Protein Simple) was reconstituted with kit-provided sample buffer and DTT solution, then added to the protein samples at a ratio of one part master mix to four parts protein lysate to achieve a final protein concentration of 1.5 μg/μL and a DTT concentration of 40 mM. Samples were then incubated at room temperature for 30–35 min prior to loading in the Wes plate (at 3 μg/lane) to give the DTT time to fully denature the proteins before size separation in the kit-provided capillary cartridge.

Following size separation, the presence of Na_V1.8 protein was detected through chemiluminescence using a polyclonal anti-Na_V1.8 primary antibody (Novus, cat. # NBP2–75584, 1:100 dilution) and an undiluted Horseradish Peroxidase-conjugated anti-rabbit secondary antibody provided in an accessory kit (Protein Simple, cat. # DM-001). The Wes machine performed all incubations and washes automatically, and the companion software (Compass 3.1.8) was used to align the signals from each capillary with the 66 – 440 kDa ladder via fluorescent standards contained in the master mix. Na_V1.8 chemiluminescence peaks were identified in the range of 220 – 290 kDa and their size was determined automatically by the software through the integration of a fitted Gaussian function. Peak areas were normalized across lanes based on the measurement of total protein in each sample using an accessory kit (Protein Simple, DM-TP01). Normalized peak areas were compared to determine the percent knockdown of Na_V1.8 in siRNA-transfected samples versus controls treated with scrambled siRNA.

Electrophysiological recordings and analysis

Na⁺ currents from isolated DRG neurons were acquired at room temperature employing the whole-cell patch-clamp technique with an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA), equipped with an ITC-18 data acquisition interface (HEKA Instruments, Holliston, MA). The recording electrodes were made from borosilicate glass capillaries (No. 8250; King Precision Glass, Claremont, CA) and coated with Sylgard (Dow Corning, Midland, MI). Data acquisition and voltage protocol generation were performed with custom-designed software (F6) developed by Stephen R. Ikeda (National Institute on Alcohol Abuse and Alcoholism). The current traces were filtered at 2 kHz (−3 dB) and digitized at 10 kHz with a four-pole low-pass Bessel filter. Both the cell membrane capacitance and series resistance (80–85%) were electronically compensated, while the lag was set to approximately 7 μs. It should be noted that we did not correct for liquid junction potentials. The external solution consisted of (in mM) 70 NaCl, 80 tetraethylammonium (TEA)-Cl, 5 MnCl₂, 10 HEPES, 10 glucose, and 0.0003 TTX, pH 7.4. The pipette solution contained (in mM) 120 N-methyl-d-glucamine, 20 TEA-OH, 10 sucrose, 11 EGTA, 10 HEPES, 1 CaCl₂, 4 Na₂-ATP, 0.3 Na₂GTP, and 14 Tris-creatine phosphate, pH 7.2 with methanesulfonic acid.

Na⁺ currents were analyzed with the Igor Pro software (WaveMetrics, Oswego, OR). In order to determine the cell’s current density (pA/pF), the peak Na⁺ current amplitude was divided by the membrane capacitance. This latter parameter was measured from the uncompensated capacitive currents that were elicited with a 5-ms step pulse from −80 to −70 mV. Thereafter the equation C_m = Q/V was employed to determine cell membrane capacitance (C_m), where Q is the charge stored in the capacitor/cell membrane and V is the voltage-step amplitude. The current traces were corrected for linear leakage current, which was determined from hyperpolarizing command pulses.

The chord conductance equation, G_Na+ = I_Na+/(E_M−E_Na+), was used to determine the Na⁺ channel activation curves, where I_Na+ is the peak Na⁺ current, G_Na is the peak membrane conductance at potential E_M, and E_Na is the Na⁺ reversal potential. A modified Boltzmann function equation, employing nonlinear regression, was then used to obtain a fit to the data with GraphPad Prism software (GraphPad, San Diego, CA): G_Na/G_max = 1/(1+exp[(V_H−V_M)/k], where G_Na/G_max is the fractional peak membrane conductance, V_H is the half-activation potential, V_M is the membrane potential, and k is the slope factor. Similarly, the Na⁺ current inactivation curves were acquired with a steady-state inactivation protocol which consisted of a 0.5-s conditioning pre-pulse over the potential range of −80 mV to +10 mV. This was followed by a constant test pulse to the peak test potential. A modified Boltzmann equation was also used to fit the inactivation curves with GraphPad Prism software.

It should be noted that in all DRG neurons tested in this study, no overt TTX-resistant Na_V1.9 currents were observed. Compared to Na_V1.8, this channel subtype exhibits hyperpolarizing activation and slow inactivation properties [28]. Further, the holding potential (−80 mV) in our experiments favored Na_V1.8 currents. Nevertheless, we cannot rule out that a minor component of total NaV currents originated from Na_V1.9 channels. Although the isolated tissue was not dissociated until Day 6 post-isolation (Figure 1), the whole-cell patch-clamp recordings were performed in DRG neurons devoid of processes. Those recordings obtained with poor space-clamp were excluded.

Data and statistical analysis

Statistical comparisons of the nonlinear regression fits for activation and inactivation were determined with a sum-of-squares F-test using Prism (GraphPad Software). The figures shown were obtained with Graphic (Autodesk, San Rafael, CA) software. The data are expressed as mean ± standard error (SE).

Results

Figure 2A illustrates representative native TTX-resistant Na⁺ current recordings obtained from acutely isolated DiI-labeled DRG neurons innervating the colon. The currents were elicited by depolarizing pulses from a holding potential of −80 mV. The amplitude of the peak current plotted as a function of step voltage (I-V relationship) is shown in Figure 2B. As the depolarization steps increased, the inward current increased over the potential range of −35 to 0 mV. At potentials more positive than +5 mV, the current decreased, reversing polarity near +54 mV. From the I-V curve, the reversal potential value was obtained by interpolating a linear portion of curve. In this group of neurons, the native peak current density was 62.0 ± 27.4 pA/pF (n=6). Figure 2C is the activation curve, which is plotted as peak conductance normalized to maximal peak conductance. The peak conductance value was calculated from the chord conductance equation (described in the Methods section). The activation curve depicted for this set of neurons shows that the Na⁺ current activates over the potential range of −35 mV to +20 mV. The V_H and slope factor values were −12.7 mV and 5.4 mV, respectively.

Figure 2.

A) Representative family of Na⁺ currents recorded from an acutely isolated DiI-labeledrat DRG neuron in the presence 300 nM TTX to block the endogenous TTX-sensitive Na⁺ currents. The Na⁺ currents were elicited with the voltage protocols shown, below. B) Mean current-voltage (I-V) relationship for the peak Na⁺ currents in DiI-labeled neurons. Data are means ± SE of neurons tested (n=6). C) Voltage-dependent activation curve of native rat DRG Na_V1.8 channels. Solid lines represent the nonlinear regression fits of a modified Boltzmann equation to the data (mean ± SE) obtained at each step potential.

In the remaining set of experiments to be described, the DRG tissue was transfected with siRNA designed to silence native Na_V1.8 channels (described above). As the entire ganglia were transfected, the colon was not injected with DiI. Rather, the DRG neurons to be studied were chosen randomly. Figure 3A shows that Na_V1.8 mRNA levels in transfected tissue were decreased by more than 98% when compared to the tissue transfected with scrambled siRNA. Further, the Western blotting analysis (Figure 3B) indicated that Na_V1.8 expression levels were decreased compared to those transfected with scrambled siRNA by 76% (n=6 ganglia/rat, 3 rats). Two bands of approximately 100-kDa (in siRNA-transfected lanes) and 140-kDa (in scrambled siRNA-transfected lane) were detected, which suggest non-specific detection by the antibody. The scatter plot depicted in Figure 3C illustrates the peak Na_V1.8 current density in a group of DRG neurons in which natively expressed Na_V1.8 channels were silenced. The mean current density was 4.4 ± 2.3 pA/pF (n=25). No overt Na_V1.8 currents were observed in 17/25 DRG neurons.

Figure 3.

Quantitative assessment of Na_V1.8 mRNA expression in DRG tissue post-Na_V1.8 siRNA transfection by qRT-PCR (A), Western blot analysis (B) and whole-cell patch-clamp (C). A) qRT-PCR analysis indicating Na_V1.8 mRNA expression levels in L₄-L₆ DRG tissue transfected with Na_V1.8 siRNA 4–5 days post-transfection. QRT-PCR was carried out with total RNA from scrambled- and Na_V1.8 siRNA-transfected DRG tissue. The fold-differences were calculated with the Ct values for the Na_V1.8 probe and corrected for GAPDH expression levels (reference gene) in each sample. The corrected Na_V1.8 expression level was then normalized to its corresponding value obtained from DRG tissue transfected with scrambled siRNA (i.e. ΔΔCt) and % change is shown (each point represents one animal). B) Na_V1.8 protein expression levels in DRG tissue transfected with either scrambled or Na_V1.8 siRNA. The predicted molecular mass for the channel is approximately 280 kDa. The lanes were loaded with 3 μg protein. Each lane represents one animal. C) Measurement of the Na_V1.8 current amplitude at the peak depolarizing potential. The Na⁺ current density was calculated from the peak Na⁺ current amplitude at the peak test pulse and normalized to membrane capacitance. The cell membrane capacitance (C) was determined from the numerical integration of a transient elicited with a depolarizing pulse from −80 mV to −70 mV prior to electronic compensation.

Next, we recorded TTX-resistant Na⁺ currents in DRG neurons in which natively expressed Na_V1.8 channels were initially silenced and subsequently transfected with the wild-type, human Na_V1.8 (1073A, Figure 1). As described for Figure 2 above, Figure 4A shows superimposed inward Na⁺ currents that were elicited by depolarizing steps from a holding potential of −80 mV. The currents began to activate near −30 mV and the peak current reached a peak near −5 mV. The I-V curve is depicted in Figure 4B and the peak current density for this group of neurons was 41.6 ± 15.6 pA/pF (n=10). Figure 4C is the activation curve, which indicates that the Na⁺ current for 1073A-transfected neurons activates over the potential range of −35 mV to +20 mV. The V_H and slope factor values were −7.3 mV and 8.7 mV, respectively.

Figure 4.

A) Na⁺ currents carried by heterologously expressed wild-type (1073A) human Na_V1.8 channels in rat DRG neurons. Representative family of Na⁺ currents evoked with the voltage protocol shown in a DRG neuron expressing wild-type human Na_V1.8 channels. The recordings were elicited in the presence of 300 nM TTX to block endogenous Na_V channels. B) Mean current-voltage (I-V) relationships for peak TTX-resistant Na⁺ currents. Each point indicates the mean ± SE (n=16) of DRG neurons tested. C) Voltage-dependent activation curve of wild-type (1073A, black circles), mutant (1073V, red circles) human Na_V1.8, and natively expressed rat Na_V1.8 (blue circles). Solid lines represent the nonlinear regression fits of a modified Boltzmann equation to the data (mean ± SE) obtained at each step potential. The data for endogenous rat Na_V1.8 (also shown in Figure 2C) is included for the purpose of comparison to data obtained with both human cDNA constructs.

We next recorded TTX-resistant Na⁺ currents in DRG neurons transfected with the minor variant (1073V). Figure 5A shows superimposed Na⁺ currents which began to activate near −35 mV and reached a peak current at 0 mV. The I-V relationship shown in Figure 5B indicates that the channel peak current was reached at +5 mV and started to reverse at potentials more positive than +10 mV. From the activation curve in Figure 5C, it can be observed that the variant channels activate over the range of −25 mV to +20 mV. For the variant channels, the V_H value was −9.0 mV while the slope factor value was 3.5 mV. The activation curve for 1073A-expressing neurons (Fig. 4C) exhibited a shallower voltage dependence than that obtained in 1073V-expressing cells (Fig. 4C). We compared the fits for both groups of neurons and they were significantly different (p=0.0043). Moreover, the peak current density for the 1073V-expressing neurons was 19.7 ± 5.2 pA/pF (n=5). A statistical comparison of current density showed there was no statistical difference between 1073A- and 1073V-expressing DRG neurons (p=0.35, unpaired t test). Nevertheless, the current density for both groups of neurons was lower than that measured in acutely isolated DRG neurons described above.

Figure 5.

A) Na⁺ currents carried by heterologously expressed mutant (1073V) human Na_V1.8 channels in rat DRG neurons. Representative traces of whole-cell Na⁺ currents evoked with the voltage protocol shown recorded from a DRG neuron transfected with the mutant human Na_V1.8 channel. The currents were acquired in the presence of 300 nM TTX to block endogenous Na_V channels. B) Mean current-voltage (I-V) relationships for peak TTX-resistant Na⁺ currents. Each point indicates the mean ± SE of DRG neurons tested (n=5).

In the next set of experiments the steady-state inactivation parameters were determined for both groups of neurons with the voltage paradigm shown in Figure 6A. The superimposed traces illustrated in Figures 6A and 6B were obtained from 1073A- and 1073V-transfected DRG neurons, respectively. The nonlinear least square fits of the averaged data are represented by the solid lines for each group. Comparison of the best fits for both groups indicated they were significantly different (p<0.0001). The V_H values for the 1073A- and 1073V-transfected neurons were −34.6 and −26.3 mV, respectively. Also, the slope factor (k) for the 1073A- and 1073V-expressing DRG neurons were −7.1 and −6.7 mV, respectively.

Figure 6.

Na⁺ currents in rat DRG neurons transfected with either the wild-type (1073A) or variant 1073V human Na_V1.8 channel. A) Family of Na⁺ currents in 1073A-expressing DRG neurons obtained with the voltage protocol (middle). B) Na⁺ current traces acquired with the voltage protocol shown in A. C) Superimposed inactivation curves both 1073A- and 1073V-transfected DRG neurons. The solid lines represent the best fit of a modified Boltzmann equation for 1073A- (A, black circles, n=6) and 1073V-expressing (B, red circles, n=5) neurons. The data were normalized to the fitted maximum current.

Discussion

TTX-resistant Na_V1.8 channels, expressed primarily in nociceptive neurons [1], have been reported to undergo gain- or loss-of-function polymorphisms in painful disorders. The putative loss-of-function Na_V1.8 (1073V) variant, for instance, has been previously shown to be associated with silent IBD [20, 25], diminished mechanical pain sensitivity [19], diagnosis of cardiac arrhythmias [29], early onset of atrial fibrillation [24] and Brugada syndrome [23]. The affected amino acid lies in the intracellular loop between transmembrane domains II and III. Furthermore, when heterologously expressed in rat superior cervical ganglion (SCG), the 1073V variant activated at more depolarized potentials when compared to neurons expressing the wild-type channel [25]. In the present study, the I-V relationships indicated that the peak membrane potential occurred at approximately +5 mV. In this follow-up study, we found that when either channel was expressed within a sensory neuron environment (rat DRG neurons) there was a modest difference in V_H activation values. That is, there was an approximately 1.8 mV difference when comparing activation parameters of both groups of neurons. This observation is similar to that observed by Waxman and colleagues who likewise heterologously expressed both Na_V1.8 channel isoforms in mouse DRG neurons and observed a 4.3 mV hyperpolarizing shift in V_H [19]. In the same study, the 1073V-expressing neurons showed an acceleration of inactivation [19]. Both of these effects could potentially modulate pain pathways. Additionally, this finding is unlike our previous report where we expressed both variants in rat SCG neurons and observed an approximate 10 mV shift toward depolarizing potentials of V_H by 1073V-expressing cells [25]. Two earlier studies also examined the biophysical properties of both variants expressed in ND7/23 [23] and Neuro 2A cells [24] and reported that the activation V_H was shifted to more hyperpolarized membrane potentials. In the present study we also observed a decrease of 5 mV of the activation slope factor (k) in DRG neurons expressing the 1073V variant channel. This decrease in the k value resulted in a steeper activation-voltage relationship, suggesting the variant exhibits a greater sensitivity to changes in membrane potential. Our results show that the activation V_H for natively expressed Na_V1.8 was hyperpolarized approximately 5 mV to that measured in 1073A-expressing neurons. This suggests that we did not obtain exact recapitulation of the phenotype. These findings may have been a result of posttranscriptional processes, posttranslational modifications, and channel trafficking. Additionally, there may have been some time-dependent changes that occurred as neurons spent time in culture.

We also examined whether the steady-state inactivation parameters would differ between both groups of neurons transfected with the channel variants. Comparing the inactivation properties revealed that the nonlinear regression fits were significantly different and there was a depolarizing shift in V_H of approximately 8 mV in 1073V-expressing DRG neurons when compared to neurons expressing 1073A channels. We observed a similar depolarizing shift in SCG neurons, though the magnitude was approximately 3.3 mV [25]. Conversely, an approximately 2 mV hyperpolarization shift of the steady-state inactivation V_H in mouse DRG neurons expressing 1073V channels was reported [19]. Similarly, it was also shown that ND7/23 [23] and Neuro 2A [24] cells expressing the 1073V variant exhibited almost a 22 mV and 0.4 mV hyperpolarizing shift, respectively. The differences observed in all these studies is likely a result of: i) expression models (sensory neurons vs. sympathetic neurons vs. cell lines), ii) animal model (rat vs. mouse), iii) transfection approach (electroporation vs lipofection vs stable cell line expression), iv) variability in channel expression levels. We also employed neurons in which native Na_V1.8 channels were silenced, while it is possible to employ sensory neurons from Na_V1.8 KO mice [19].

The depolarizing shift of the steady-state inactivation observed in the current study suggests that neurons might exhibit a hyperexcitable state. Indeed, electrophysiological recordings of another Na_V1.8 polymorphism, discovered in a patient with small fiber neuropathy (G1662S), indicated that the observed hyperexcitability was linked to a 7 mV depolarizing shift of the steady-state inactivation [18]. Nevertheless, our finding is difficult to reconcile with the hypoalgesic phenotype observed in patients with “silent” IBD [20]. However, it is also possible that other properties, not tested in the present study, may have been altered such that the net effect would result in neuronal hypoexcitability. For example, current clamp experiments were not performed in the present study as the focus was to examine Na_V1.8 biophysical properties. Since the endogenous Na_V1.8 were silenced, it is uncertain whether there would be a compensatory expression of other Na_V channels, which would potentially effect the excitability of the neurons. For instance, a recent transcriptomic study of three sensory, including DRG, and one sympathetic rat ganglia showed that the sensory ganglia expressed Na_V1.6-Na_V1.9 [30].

A potential limitation to the results described is that the expression environment may not necessarily possess conditions that recapitulate those available to the native human channel. This includes a binding protein that modulates channel activity or other processes including trafficking or post-translational modification of these channels. Furthermore, in the present study, we initially silenced natively expressed Na_V1.8 channels. The knockdown period occurred three to four days post-siRNA transfection and one to two days post-1073A or −1073V cDNA transfection. Consequently, the time-dependent changes observed may be responsible for the observed differences. It is also possible that during the time the DRG tissue remained in culture, the expression levels of other ion channels and/or accessory proteins may have been altered. As mentioned above, current clamp experiments were not carried out in this study. Finally, the results of the present study and those discussed above highlight the challenges posed by different expression systems in performing reconstitution experiments. With the advent of CRISPR technology, it is tempting to speculate that the creation of an animal model that carries the 1073V variant would provide us with key mechanistic insights of the channel’s role in pain transmission processes.

In summary, natively expressed Na_V1.8 channels were silenced in rat DRG neurons and this was followed by heterologous expression of either human wild-type 1073A or polymorphic (1073V) channels. The biophysical characterization of the variant channels exhibited a slight (~1.8 mV) hyperpolarized activation V_H, and a lower slope factor (k) indicative of a higher voltage dependence. In addition, the steady-state inactivation V_H for the 1073V channels was shifted by 8 mV in a depolarization direction when compared to the wild-type channels. This depolarizing shift suggests that heterologous expression of the mutant channels may alter the neurons’ phenotype to an hyperexcitable condition. Considering the differences observed in studies evaluating these parameters, it will be necessary to perform electrophysiological experiments in induced pluripotent stem cell-derived human sensory neurons or employing animal models carrying the cognate variant to further elucidate the impact of this polymorphism on the biophysical function in nociceptive neurons.