Na_v1.8 in small dorsal root ganglion neurons contributes to vincristine-induced mechanical allodynia

Abstract

Vincristine-induced peripheral neuropathy is a common side effect of vincristine treatment, which is accompanied by pain and can be dose-limiting. The molecular mechanisms that underlie vincristine-induced pain are not well understood. We have established an animal model to investigate pathophysiological mechanisms of vincristine-induced pain. Our previous studies have shown that the tetrodotoxin-sensitive voltage-gated sodium channel Na_v1.6 in medium-diameter dorsal root ganglion (DRG) neurons contributes to the maintenance of vincristine-induced allodynia.

In this study, we investigated the effects of vincristine administration on excitability in small-diameter DRG neurons and whether the tetrodotoxin-resistant (TTX-R) Na_v1.8 channels contribute to mechanical allodynia. Current-clamp recordings demonstrated that small DRG neurons become hyper-excitable following vincristine treatment, with both reduced current threshold and increased firing frequency. Using voltage-clamp recordings in small DRG neurons, we now show an increase in TTX-R current density and a −7.3 mV hyperpolarizing shift in the half-maximal potential (V_1/2) of activation of Na_v1.8 channels in vincristine-treated animals, which likely contributes to the hyperexcitability that we observed in these neurons.

Notably, vincristine treatment did not enhance excitability of small DRG neurons from Na_v1.8 knockout mice, and the development of mechanical allodynia was delayed but not abrogated in these mice. Together, our data suggest that sodium channel Na_v1.8 in small DRG neurons contributes to the development of vincristine-induced mechanical allodynia.

Introduction

The microtubule-disrupting agent vincristine often produces a dose-dependent peripheral neuropathy with sensory, autonomic and motor symptoms.^1-3 Sensory symptoms are most frequently seen in vincristine-induced peripheral neuropathy (VIPN), including early manifestation of numbness and tingling of hands and feet, distal hyperesthesia and neuropathic pain which can be severe.⁴ Vincristine-induced pain appears to be associated with a poly-modal neuropathy that involves Aβ, Aδ, and C fibres,⁵ but its underlying molecular mechanisms are still not well understood.

Multiple mechanisms may contribute to the pathophysiology of VIPN.^6-8 Vincristine destabilizes axonal microtubules^9-11 and disrupts both kinesin-mediated anterograde and dynein-mediated retrograde microtubule-dependent axonal transport.¹² In addition, vincristine treatment induces a strong inflammatory response,¹³ including the infiltration of monocytes/macrophages into sciatic nerve and dorsal root ganglia (DRG), which leads to release of interleukin (IL)-6,¹⁴ and production of reactive oxygen species (ROS) and activation of neuronal TRPA1 channels.¹⁵ Minocycline inhibits activation of immune cells and has a protective effect against vincristine-induced mechanical allodynia.¹⁶ At peak pain severity, vincristine treatment is associated with a significant increase in spontaneous discharge in A- and C-fibres¹⁷ and with central sensitization in spinal cord wide dynamic range (WDR) neurons,¹⁸ which suggests the involvement of multiple classes of ligand-gated ion channels and receptors,^19-21 as well as voltage-gated sodium (Na_v) channels.^22-25

Na_v channels are essential for the initiation and propagation of action potentials in spontaneously firing neurons and stimulus-evoked activity of these neurons, and previous studies have suggested the involvement of peripheral Na_v channels in VIPN. Vincristine has been reported to induce an increase in immunoreactivity for tetrodotoxin-sensitive (TTX-S) sodium channel isoforms Na_v1.3 and Na_v1.6, as well as the tetrodotoxin-resistant (TTX-R) isoform Na_v1.8.²³ Na_v1.8 is particularly important for the different pain modalities, because it is the major contributor to the upstroke of action potential and repetitive firing in C-fibre nociceptors.^26-28 Early pharmacological evidence suggests that TTX-R Na_v channels may play a key role in vincristine-induced mechanical allodynia; systemic sodium channel blockers such as lidocaine and mexiletine attenuate tactile allodynia in rodent models of VIPN, while TTX has no effect.^24,25 However, neither anti-sense oligodeoxynucleotides targeting Na_v1.8 channels nor pharmacological inhibition of Na_v1.8 with the blocker A-803467 produce any antinociceptive effects in vincristine-induced mechanical allodynia.^24,29 These pharmacological results have been interpreted as suggesting a Na_v1.8-independent mechanism for vincristine-induced pain but have not been reconciled with the published immunohistochemical data reporting an increase in the expression of this channel.

We recently developed a mouse model of vincristine-induced mechanical allodynia and showed a significant upregulation of Na_v1.6-mediated TTX-S Na⁺ currents in medium DRG neurons, but not in small DRG neurons, and an attenuation of maintenance but not induction of mechanical allodynia in mice in which Na_v1.6 channels are knocked out in Na_v1.8-positive neurons [Na_v1.6^Nav1.8 conditional knockout (KO)].²² Interestingly, we also showed in the same study that the KO of Na_v1.7 channels in Na_v1.8-positive neurons (Na_v1.7^Nav1.8 conditional KO), which accounts for the majority (70%) of TTX-S current in these small DRG neurons,^30,31 does not ameliorate vincristine-induced mechanical allodynia. However, it has been shown that C-fibre nociceptors (small-diameter DRG neurons) become mechanically hyper-responsive after systemic administration of vincristine.^32,33 Thus, the mechanism underlying the hyperactivity of these neurons, and the contribution of Na_v1.8 remain to be determined.

In the present study, we investigated whether vincristine in our animal model causes hyper-excitability of small diameter DRG neurons and whether TTX-R sodium channel Na_v1.8 contributes to these changes. We showed that vincristine treatment increases the Na_v1.8 TTX-R current density and induces a hyperpolarizing shift in the activation of these channels in small DRG neurons, consistent with the increased activity in these neurons. We also demonstrated a delay in the onset of vincristine-induced mechanical allodynia in Na_v1.8 KO mice, supporting a role for this channel in the development of vincristine-induced pain.

Materials and methods

Animals

Seventy adult mice (3–4 months old) of both sexes (31 males and 39 females) were used in this study. Wild-type (WT) C57BL/6 mice (n = 46) were obtained from Charles River for the first behaviour experiment. For the second experiment, we used Na_v1.8 KO mice (n = 12) and their WT littermates (n = 11), which carry a biallelic Cre insertion in the ATG-carrying exon of Scn10a³⁴ and a Cre-reporter cassette (loxP-stop-loxP-tdTomato fluorescent protein), which produces a red fluorescent protein in neurons that express a functional Cre recombinase.³⁰ Animals were group housed (maximum five per cage) in a controlled environment (22 ± 2°C temperature and 55%–65% humidity) and in artificially lighted rooms on a well-defined 12 h light/dark cycle (lights on at 6:00 a.m.). Food and water were provided ad libitum throughout the experiment except during behavioural testing. Mice were acclimated to the behavioural testing environment for 2 weeks before experimental procedures. The experiments were performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee of the Veterans Administration Connecticut Healthcare System. All animals were weighed every 3 days and the weight gain was calculated. The experimenter was blinded to treatment and genotype.

Vincristine treatment

Vincristine sulfate (V8388, Sigma-Aldrich) was dissolved in ddH₂O (degassed with nitrogen sparging) to a stock solution of 2.5 mg/ml and stored at −20°C. The stock solution was used within 1 month. The stock solution was diluted in saline to 0.25 mg/ml immediately before injections. Mice were injected intraperitoneally (i.p.) with vincristine at 3 ml/kg to a final dose of 0.75 mg/kg twice a week for 4 weeks (accumulated dose 6 mg/kg). Control mice received an equal volume of saline.

Behavioural studies

von Frey test

Animals were placed on an elevated wire grid and the plantar surface of both hind paws was presented with a series of calibrated von Frey filaments (Stoeling). Each filament was applied for 5 s. A nocifensive response was recorded when flinching, withdrawal, paw licking or toe spreading was observed. At least 5 min of rest were included between each stimulus. The 50% withdrawal threshold was determined using the ‘up-down’ method, which requires four stimuli straddling the threshold.

Hargreaves test

Animals were placed on a glass plate, and the plantar paw surface was exposed to radiant heat using a Hargreaves apparatus (IITC) and the paw withdrawal latency was measured. The heat stimulation was repeated three times on both hind paws. At least 5 min of rest were included consecutive tests. The cut-off value was set at 30 s to prevent tissue damage.

Gait analysis

Animals were placed in the tunnel of a Catwalk apparatus (Noldus Information Technology) and could walk freely from one end to the other. Walking patterns were recorded by a high-speed camera collecting reflexive fluorescent light from the contact surface between the paws and glass platform. All experiments were performed in a dark room. Analysis was performed using the Catwalk XT 10.1 software. Three trials were recorded for each animal and time point. The parameters used to analyse gait were stride length and print area.

Electron microscopy analyses of axons of sciatic nerve

We followed our previously described method for preparing sciatic nerve for structural analysis.²² Briefly, mice (two males and two females each for the control and vincristine groups) were anaesthetized with ketamine/xylazine (100/10 mg/kg, i.p.) and transcardially perfused with 0.01 M PBS (pH 7.4) followed by ice-cold 4% paraformaldehyde in 0.14 M Sorensen’s phosphate buffer (pH 7.4). Sciatic nerves at mid-thigh level (distal to the trifurcation) were dissected and transferred to 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) at 4°C overnight. Samples were then post-fixed with 1% osmium tetroxide, 0.8% potassium ferrocyanide in 0.1 M cacodylate buffer, post-stained with 2% uranyl acetate, dehydrated and embedded in fresh EPON. Tissues were sectioned with a Leica UC7 ultramicrotome in 50 nm sections and imaged on an FEI Tecnai G2 Spirit Biotwin transmission electron microscope at 80 kV with an AMT NanoSprint 15 camera. To quantify the number of myelinated nerves, six images were investigated per mouse. Axon and myelin areas of myelinated nerves were measured, and the g-ratio was calculated by dividing the area of the axon by the outer area of the myelinated fibre. All measurements were carried out using ImageJ v.1.50. Statistical significance was determined (P < 0.05) using unpaired Student’s t-tests. All results are presented as mean ± standard error of the mean (SEM).

Primary mouse dorsal root ganglia neuron isolation

DRG neurons from thoracic and lumbar regions were isolated from mice (both male and female) using protocols previously described.^35,36 Briefly, DRG neurons were harvested and first incubated at 37°C for 20 min in complete saline solution (CSS) (in mM: 137 NaCl, 5.3 KCl, 1 MgCl₂, 25 sorbitol, 3 CaCl₂ and 10 HEPES, adjusted to pH 7.2 with NaOH) supplemented with 0.5 U/ml Liberase TM (Roche) and 0.6 nM EDTA. Then the tissue was incubated for 15 min at 37°C in CSS containing 0.5 U/ml Liberase TL (Roche), 0.6 mM EDTA and 30 U/ml papain (Worthington Biochemical). DRG neurons were then centrifuged and triturated in 0.5 ml of DRG culture medium containing 1.5 mg/ml bovine serum albumin (low endotoxin) and 1.5 mg/ml trypsin inhibitor (Sigma). The cell suspension was filtered through a 70 μm nylon mesh cell strainer (Becton Dickinson) to remove debris, and the mesh was then washed twice with 2 ml of DRG culture medium. DRG neurons were pelleted by centrifugation and resuspended in 1 ml of DRG culture medium. Aliquots of 100 μl cell suspension were seeded onto poly-D-lysine/laminin-coated coverslips (BD) and incubated at 37°C in a 95% air/5% CO₂ (vol/vol) incubator. After 45 min of incubation, DRG culture medium was added into each well to a final volume of 1 ml. DRG neuron cultures were maintained at 37°C in a 95% air/5% CO₂ (vol/vol) incubator for 18–30 h before current-clamp or voltage-clamp recording. For current-clamp recording, nerve growth factor (50 ng/ml) and glial cell line-derived neurotrophic factor (50 ng/ml) were added to the DRG culture medium.

Electrophysiology

Voltage-clamp

Voltage-clamp recordings were conducted within 24 h after DRG neuron isolation using an EPC 10 amplifier and the Patchmaster program (v 53; HEKA Elektronik) at room temperature (22 ± 1°C). Small (25–30 µm) DRG neurons with no apparent neurites were selected for voltage-clamp recording. Fire-polished electrodes were fabricated from 1.6 mm outer diameter borosilicate glass micropipettes (World Precision Instruments) using a Sutter Instruments P-97 puller and had a resistance of 0.6–1.3 MΩ. Pipette potential was adjusted to zero before seal formation. Liquid junction potential was not corrected. To reduce voltage errors, 60%–85% series resistance compensation was applied. Linear leak currents were subtracted out using the P/6 method. Current traces were sampled at 50 kHz and filtered with a low-pass Bessel setting of 10 kHz. The pipette solution contained (in mM): 140 CsF, 10 NaCl, 1 EGTA, 10 dextrose and 10 HEPES, pH 7.3 with CsOH. The extracellular bath solution contained (in mM): 140 NaCl, 3 KCl, 1 MgCl₂, 1 CaCl₂, 10 HEPES, 5 CsCl, 20 tetraethylammonium chloride (TEA·Cl), pH 7.3 with NaOH. CdCl₂ (0.1 mM) and 4-aminopyridine (4 mM) were added to the bath solution to block endogenous calcium currents and potassium currents, respectively. To record TTX-resistant currents, 1 µM TTX was added to the bath. DRG neurons were held at −100 mV. After a 5-min equilibration period once whole-cell configuration was achieved, total voltage-gated sodium currents were recorded using a series of 100-ms step depolarizations (−80 to +40 mV in 5-mV increments at 5-s intervals) from the holding potential. Current density was calculated by normalizing maximal transient peak currents with cell capacitance. For analysis of gating properties, data were excluded if voltage error exceeded 5 mV for small DRG neurons. Peak inward currents obtained from activation protocols were converted to conductance values using the equation

where G is the conductance, I is the peak inward current, V_m is the membrane potential step used to elicit the response and E_Na is the reversal potential for the sodium channel, which is determined for each cell using the x-axis intercept of a linear fit of the peak inward current responses to the last six voltage steps of the activation protocol.

Conductance data were normalized by the maximum conductance value and fit with a Boltzmann equation of the form

where V_1/2 is the midpoint of activation and k is a slope factor.

Steady-state fast inactivation was assessed with a series of 500-ms prepulses (−140 mV to −10 mV in 10-mV increments); the remaining non-inactivated channels were activated by a 40-ms step depolarization to −10 mV. The steady-state slow inactivation was measured with a series of 30-s prepulses, ranging from −130 to +20 mV in a 10-mV increment, followed by a 100-ms step depolarization to −100 mV and then a 20-ms test pulse at −10 mV to assess the available non-inactivated channels. Peak inward currents obtained from steady-state fast or slow inactivation were normalized by the maximum current amplitude and fit with a Boltzmann equation of the form

where V_m represents the inactivating prepulse membrane potential and V_1/2 represents the midpoint of fast or slow inactivation.

Current-clamp

Current-clamp recordings were conducted 24 h after DRG neuron isolation using an EPC 10 amplifier and the Patchmaster program (v. 53; HEKA Elektronik) at room temperature (22 ± 1°C). Small (25–30 µm) DRG neurons were selected for current-clamp recording. Fire-polished electrodes were fabricated from 1.6 mm outer diameter borosilicate glass micropipettes (World Precision Instruments) using a Sutter Instruments P-97 puller and had a resistance of 0.7–1.5 MΩ when filled with the pipette solution, which contained the following (in mM): 140 KCl, 0.5 EGTA, 5 HEPES, 3 MgATP, pH 7.30 with KOH (adjusted to 310 mOsm with dextrose). The extracellular solution contained the following (in mM): 140 NaCl, 3 KCl, 2 MgCl₂, 2 CaCl₂, 10 HEPES, pH 7.30 with NaOH (adjusted to 320 mOsm with dextrose). Whole-cell configuration was obtained in voltage-clamp mode before proceeding to the current-clamp mode. Cells with stable (<10% variation) resting membrane potentials (RMP) more negative than −35 mV and overshooting action potentials (>85 mV, from RMP to peak) were used for additional data collection. Input resistance was determined by the slope of a linear fit to hyperpolarizing responses to current steps from −5 to −40 pA in −5 pA increments. Threshold was determined by the first action potential elicited by a series of depolarizing current injections (200 ms) that increased in 5 pA increments. Action potential frequency was determined by quantifying the number of action potentials elicited in response to depolarizing current injections (500 ms) of different strengths (25–500 pA in 25 pA increments).

Real-time PCR neuronal ion channels array

Animals (males only) were perfused transcardially with cold PBS on Day 28 after vincristine or vehicle treatment and bilateral L3–L5 DRGs were collected for real-time (RT) PCR array. The DRGs from each animal were pooled. The tissue was frozen with liquid nitrogen and subsequently disrupted and homogenized using the TissueLyser system (Qiagen). The homogenization procedure involved two 1-min runs of the TissueLyser at 50 Hz. Total RNA was extracted and purified using the RNeasy Mini Kit (Qiagen). The concentration and purity of RNA were verified by measuring the absorbance at 260 and 280 nm (NanoDrop 2000, Thermo Scientific). Starting with 1 µg of total RNA, residual genomic DNA removal and reverse transcription into complementary DNA (cDNA) was performed using the RT2 First Strand Kit (Qiagen). A 96-well RT2 Profiler PCR array containing mouse neuronal ion channels (PAMM-036ZD) configured for use in the Bio-Rad CFX96 RT PCR system (Bio-Rad) was used to quantify gene expression. To minimize the variance in sample processing, the samples were handled by the same experimenter. Experiments were performed on tissue from three vincristine-treated animals and three control animals, and the experimenter was blinded.

Statistical analysis

Electrophysiological data were analysed using Fitmaster (HEKA Elektronik) and Origin (Microcal Software) software, and statistical significance was determined by unpaired Student’s t-test when data obeyed normal distribution or non-parametric Mann–Whitney test when data were not normally distributed. Behavioural data were analysed using Prism (GraphPad), and statistical significance was determined (P < 0.05) using two-way ANOVA with Fischer’s least significant difference (LSD) test. Unless otherwise noted, statistical significance for pairwise comparisons was determined (P < 0.05) using an independent t-test. All results are presented as mean ± SEM.

Results

Vincristine induced mechanical allodynia in wild-type mice

A balanced cohort of male (n = 24) and female (n = 22) WT C57/Bl6 mice was used. Baseline mechanical sensitivity measured by von Frey test showed no difference between sexes (males: 1.07 ± 0.09 g; females: 1.13 ± 0.09 g; P = 0.54, unpaired Student’s t-test) (Fig. 1A). Similarly, baseline thermal sensitivity measured by Hargreaves showed no difference between males and females (males: 18.54 ± 0.63 s; females: 16.69 ± 0.78 s; P = 0.07, unpaired Student’s t-test) (Fig. 1B). The vincristine treatment (0.75 mg/kg) twice per week for 4 weeks showed significant effect on the von Frey test [F(1,44) = 21.40; P < 0.0001], reflecting the development of mechanical allodynia at Day 14 (saline: 1.10 ± 0.07 g, vincristine: 0.80 ± 0.06 g; P < 0.05) that persisted through the third (Day 21) (saline: 1.09 ± 0.11 g; vincristine: 0.47 ± 0.08 g; P < 0.001) and the fourth week of treatment (Day 28) (saline: 1.17 ± 0.14 g; vincristine: 0.45 ± 0.06 g; P < 0.001) (Fig. 1C). In addition, when vincristine treatment had the greatest effect (Day 28), we did not detect a significant effect of sex [treatment effect: F(1,40) = 18.15, P = 0.0001; sex effect: F(1,40) = 1.017, P = 0.32, two-way ANOVA]. Hargreaves test revealed a significant interaction between time and treatment [F(41,76) = 3.56, P = 0.0081] with a difference in thermal threshold in the last treatment week (Week 4) (saline: 15.42 ± 0.65 s; vincristine: 12.77 ± 0.61 s; P < 0.05) (Fig. 1D); there was no sex effect: F(1,40) = 1.017, P = 0.32. Vincristine treatment showed an effect on body weight gain. On the last week of treatment, the saline group showed an average weight greater than that of the vincristine group (saline: 25.04 ± 0.84 g; vincristine: 22.35 ± 0.62 g; P < 0.01) (Fig. 1E). Gait parameters [print area (Fig. 1F) and stride length (Fig. 1G)] did not show any abnormality in the vincristine-treated group. Supplementary Fig. 1 illustrates behaviour test data for each sex separately, showing that the vincristine effect was not affected by the sex of the mice.

Vincristine does not cause significant changes in peripheral nerve

We have previously shown that the regimen of vincristine treatment employed here does not produce overt structural changes to the axons in the sciatic nerve.²² We investigated whether the vincristine regimen in this cohort of animals replicated this finding. The morphology of the myelinated axons appeared similar between saline-treated and vincristine-treated groups. The integrity of the sciatic nerve at mid-thigh level was generally not affected by the vincristine treatment. As shown in Fig. 2, the morphology of myelinated axons and integrity of myelin sheaths appeared unaffected in vincristine-treated mice. Bundles of unmyelinated axons with clearly defined structures were visible in cross sections of peripheral nerve in both control and vincristine-treated mice. The morphometry of the transverse sections of sciatic nerve was also analysed. We did not observe a statistically significant difference in the density of myelinated fibres (saline: 29.25 ± 3.54/mm²; vincristine: 19.75 ± 2.39/mm²; P = 0.07) (Fig. 2B), the area of axons (saline: 5.33 ± 0.84 μm²; vincristine: 6.83 ± 0.58 μm²; P = 0.19), myelin sheaths (saline: 8.66 ± 0.99 μm², vincristine: 12.35 ± 1.33 μm²; P = 0.07) (Fig. 2C) or g-ratio (saline: 0.65 ± 0.05; vincristine: 0.60 ± 0.07; P = 0.49) (Fig. 2D) between groups. These data were consistent with our previous observations²² that vincristine does not cause significant axonal loss or degeneration at the peripheral nerve level.

Vincristine causes a significant hyperpolarizing shift in voltage-dependence of Na_v1.8 channel activation

We previously reported a vincristine-induced small 3-mV hyperpolarizing shift in V_1/2 of TTX-S channel activation and no change of TTX-S current amplitude in small DRG neurons.²² We reasoned that the minimal changes in TTX-S current were unlikely to account for the hyperexcitability of small DRG neurons observed in vincristine-treated mice. In this study, we investigated the effect of vincristine treatment on the current density and gating properties of TTX-R channels in small DRG neurons (25–30 µm).

We used voltage-clamp recordings to investigate the effect of vincristine treatment on TTX-R current, primarily that produced by the Na_v1.8 channels, in small-diameter DRG neurons (25–30 µm) isolated from control and vincristine-treated mice (both male and female) that had undergone behavioural testing (Fig. 1). After establishing whole-cell patch, DRG neurons were held at −80 mV to inactivate Na_v1.9 channels, and TTX (1 µM) was applied in the extracellular solution to block all TTX-S channels. Figure 3A and B shows representative Na_v1.8 currents in small DRG neurons from WT mice receiving a 4-week treatment of saline and vincristine, respectively. Vincristine treatment significantly increased the Na_v1.8 current density in small DRG neurons (saline: 253.35 ± 40.24 pA/pF, n = 15; vincristine: 448.35 ± 45.01 pA/pF, n = 17; P = 0.003) (Fig. 3C). We previously reported that vincristine treatment did not alter the TTX-R current density.²² The difference between the previous study and our current findings is due to methodological differences between the two studies. We previously used a reduced NaCl gradient (30 mM) to reduce the amplitude of the total sodium current, which permitted good clamping of the neuron. These conditions may have confounded the measurements of the amplitude of the TTX-R currents. In this study, we used full sodium external solution and blocked TTX-S channels using 1 µM TTX, which permitted a more accurate measurement of the Na_v1.8 current density.

Vincristine treatment also caused a shift in the voltage-dependence of activation (Act) of Na_v1.8 channels in a hyperpolarizing direction (saline: V_1/2, Act = −10.86 ± 1.45 mV, n = 11; vincristine: V_1/2, Act = −17.50 ± 2.14 mV, n = 9; P = 0.022) (Fig. 3D), with no apparent effect on steady-state fast-inactivation (Fast Inact) between control and vincristine-treated groups (saline: V_1/2, Fast Inact = −44.72 ± 1.39 mV, n = 15; vincristine: V_1/2, Fast Inact = −47.76 ± 1.48 mV, n = 10; P = 0.151). No difference was observed in the voltage-dependence of steady-state slow-inactivation (Slow Inact) either (saline: V_1/2, Slow Inact = −61.98 ± 1.38 mV, n = 12; vincristine: V_1/2, Fast Inact = −60.61 ± 2.37 mV, n = 8; P = 0.627) (Fig. 3E and F).

Vincristine treatment causes hyper-excitability of small DRG neurons from wild-type mice

We studied the excitability of small DRG neurons (25–30 µm) from control and vincristine-treated mice that had undergone behavioural testing (Fig. 1). DRG neurons were isolated from mice within the first week after the cessation of treatment, a time frame that we have previously shown to maintain the effect of treatment.²² Since the behavioural testing did not uncover a sex effect, we pooled electrophysiological data from cultures of both sexes. Current-clamp recordings were performed 24 h after isolation of DRG neurons from both male and female mice in each condition. Vincristine treatment appeared to increase the percentage of spontaneously firing neurons to 13.04% (3 of 23 small cells) compared to 5.88% for control (1 of 17 cells); however, this increase did not reach statistical significance. Figure 4 shows representative action potential traces in small DRG neurons from control (Fig. 4A) and vincristine-treated mice (Fig. 4B). Vincristine significantly reduced the current threshold of action potential firing (saline: 118.13 ± 35.78 pA, n = 16; vincristine: 41.75 ± 6.31 pA, n = 20; P = 0.025) (Table 1). In addition, vincristine significantly hyperpolarized the afterhyperpolarization potential (AHP) by 7.3 mV (saline: −18.02 ± 2.25 mV, n = 16; vincristine: −25.31 ± 1.81 mV, n = 20; P = 0.017) (Table 1 and Fig. 4C). There was no significant difference in RMP (saline: −50.52 ± 1.69 mV, n = 16; vincristine: −49.71 ± 1.07 mV, n = 20; P = 0.691), input resistance (saline: 0.99 ± 0.10 GΩ, n = 16; vincristine: 0.92 ± 0.11 GΩ, n = 20; P = 0.622) or action potential amplitude, which was measured from RMP to peak (saline: 110.19 ± 3.33 mV, n = 16; vincristine: 109.39 ± 2.76 mV, n = 20; P = 0.853) (Table 1).

To assess the excitability of small DRG neurons, a series of 500-ms current injections ranging from 25 to 500 pA was applied, and the firing frequency was determined. Compared to the control group, vincristine significantly increased the firing frequency of small DRG neurons (Fig. 4D). The number of animals and cells used for all electrophysiology experiments is listed in Table 1.

Knockout of Na_v1.8 delays the development of vincristine-induced mechanical allodynia

To investigate whether genetic deletion of Na_v1.8 channels in DRG neurons attenuates the development or maintenance of vincristine-induced pain, we examined behavioural phenotypes of Na_v1.8 KO mice and their WT littermates following vincristine treatment (Fig. 5). Na_v1.8 KO and WT littermate mice exhibited similar mechanical (WT: 0.99 ± 0.10 g; Na_v1.8 KO: 1.03 ± 0.11 g) and thermal baseline thresholds (WT: 12.3 ± 0.8 s; Na_v1.8 KO: 12.3 ± 0.6 s). Although vincristine treatment caused mechanical allodynia in mice of both genotypes, this effect was delayed in Na_v1.8 KO mice (Fig. 5). A significant mechanical allodynia developed as early as Day 7 and was maintained to Day 28 in WT mice (Fig. 5A) (P = 0.012 at Day 7, P = 0.025 at Day 14, = 0.01 at Day 21, P < 0.0001 at Day 28; two-way ANOVA with Fischer’s LSD test; statistically significant difference compared to WT baseline). In the Na_v1.8 KO group; however, the reduction in mechanical threshold following vincristine treatment did not reach statistical significance until Day 28 (Fig. 5A) (P < 0.01; two-way ANOVA with Fischer’s LSD test; statistically significant difference compared to Na_v1.8 KO baseline). Compared to the WT control group, KO of Na_v1.8 significantly attenuated vincristine-induced mechanical allodynia at Day 7 (Fig. 5A) (Na_v1.8 KO: 0.98 ± 0.10 g; WT saline: 0.676 ± 0.09 g; P = 0.046, two-way ANOVA with Fischer’s LSD test) but not at other measurement points. These observations suggested that the absence of Na_v1.8 delays the onset but does not prevent the eventual development of vincristine-induced mechanical allodynia. Vincristine treatment did not cause thermal allodynia in either WT control or Na_v1.8 KO mice (Fig. 5B). Weight loss was apparent in WT mice after vincristine treatment (baseline: 25.7 ± 1.16 g; Day 28: 24.1 ± 0.84 g). Na_v1.8 KO mice appeared to have less weight change compared to WT controls (baseline: 24.5 ± 0.9 g; Day 28: 23.7 ± 0.74 g) (Fig. 5C).

Vincristine does not significantly alter the excitability of small DRG neurons from Na_v1.8 KO mice

To further investigate the contribution of Na_v1.8 to the hyper-excitability of small DRG neurons, we examined action potential firings in small DRG neurons from Na_v1.8 KO mice with or without vincristine treatment. The RMP was similar between control and vincristine-treated Na_v1.8-null neurons (saline: −56.3 ± 3.1 mV, n = 7; vincristine: −59.2 ± 2.2 mV, n = 7, P = 0.454). In agreement with our published data,²⁶ in the absence of Na_v1.8 channels, small-diameter DRG neurons were unable to fire all-or-none action potentials at their RMP (Fig. 6A and B), with small-amplitude action potentials that crossed 0 mV being observed in a few cells with relatively hyperpolarized membrane potential (Fig. 6C and D). Hyperpolarizing the cell membrane potential in DRG neurons to −80 mV allowed TTX-S channels to recover from inactivation that occurs at RMP, thus more TTX-S channels became available for action potential generation, which restored the ability of those cells to fire all-or-none action potentials with normal amplitude (Fig. 6E and F). The mean current thresholds at −80 mV were 163 ± 22 pA (n = 4) for control neurons and 184 ± 36 pA (n = 7) for vincristine-treated neurons were not significantly different (P = 0.679). Figure 6G shows repetitive firing when DRG neurons were held at −80 mV, with no apparent difference between control and vincristine groups. This result is consistent with the little or limited effect of vincristine on the TTX-S channels in small-diameter DRG neurons, which we reported previously.²² Taken together, these data demonstrated that Na_v1.8 KO abolished vincristine-induced hyperexcitability in small DRG neurons.

Changes in gene expression of multiple ion channels in DRG neurons after exposure to vincristine

We previously reported that Na_v1.6-mediated increase in TTX-S sodium current density contributes to hyper-excitability of DRG neurons after vincristine treatment,²² and our new data show an increase in the Na_v1.8 TTX-R current density. We therefore hypothesized that vincristine chemotherapy might lead to upregulation of sodium channel expression. Using an RT-PCR array, we systematically examined gene expression changes of neuronal ion channels, including sodium channels, HCN channels, potassium channels and TRP channels. We used DRG neurons from male WT mice for this analysis. Vincristine did not alter sodium channel expression, including expression of Na_v1.6, Na_v1.7, Na_v1.8 or Na_v1.9; however, small but statistically significant downregulation of expression was observed in five potassium channels and TRPA1 (Supplementary Table 1).

Discussion

We have previously shown that Na_v1.6 channels in medium-diameter DRG neurons contribute to the maintenance of mechanical allodynia in mice.²² We have now shown in this model that vincristine-induced mechanical allodynia is not sex-dependent. We also show a significant increase in current density and hyperpolarized activation of Na_v1.8 TTX-R channels in small-diameter DRG neurons from mice of both sexes treated with vincristine. These changes in channel properties would be expected to contribute to the hyperexcitability of these neurons. Notably, deletion of Na_v1.8 in DRG neurons delayed the development of vincristine-induced mechanical allodynia, but while there is a pattern of attenuated pain throughout the treatment period, loss of Na_v1.8 did not abolish allodynia at the end of the 4-week treatment. Together, our data indicate that small-diameter DRG neurons contribute to the development of vincristine-induced mechanical allodynia and that sodium channel Na_v1.8 is critical for this effect.

Our published data²² and evidence that we provide here show that Na_v1.8 and Na_v1.6 channels in small-diameter and medium-diameter DRG neurons, respectively, play an important and possibly complementary role in the induction (Na_v1.8) and maintenance (Na_v1.6) of mechanical allodynia in mice treated with vincristine. A limitation in our study, however, is that the knockout of Na_v1.8 renders the small-diameter DRG neurons hypoexcitable because Na_v1.8 channels contribute most of the current underlying the upstroke of action potential^26,37; thus we cannot formally rule out the possibility that additional ion channels and receptors in small-diameter DRG neurons play a role in inducing mechanical allodynia. However, even after rescuing firing of all-or-none action potential by hyperpolarizing the neuronal membrane to −80 mV, we did not observe a vincristine-induced increase in repetitive firing, which suggested that the effect of vincristine on other targets in small diameter DRG neurons may not be enough to compensate for the loss of Na_v1.8 channels during the induction stage of mechanical allodynia.

We have shown here that small-diameter DRG neurons from WT mice treated with vincristine are hyper-excitable, compared to neurons from control mice. Since we did not observe a difference in the allodynia between males and females, we combined recordings from neuronal cultures of both sexes. The hyperexcitability manifested as a 65% reduction in the action potential current threshold, and an enhancement of repetitive firing in response to graded stimuli. We also observed significant hyperpolarization of the AHP in these neurons, but there was no change in other membrane properties including RMP and input resistance (Table 1). At the channel level, we observed an increase in the current density of the Na_v1.8 TTX-R currents in neurons from vincristine-treated mice but did not see an increase in the expression of the channel, which suggested a post-transcriptional and possibly post-translational modulation of the channel. Additionally, we observed a significant hyperpolarizing shift of −7.3 mV in the activation of Na_v1.8, which would be expected to increase repetitive firing of DRG neurons and may also contribute to the reduction in the threshold for action potential firing. Gain-of-function mutations in Na_v1.8 channels from patients with small fibre neuropathy, which shift activation in a hyperpolarizing direction of a similar magnitude, reduce the threshold for action potential firing and increase repetitive firing of DRG neurons consistent with the pain phenotype in these patients.^36,38 These data support the contribution of Na_v1.8 to the hyper-excitability of small DRG neurons in vincristine treated mice.

Vincristine treatment causes tissue inflammation.³⁹ It is known that the inflammatory mediator prostaglandin E2 causes a hyperpolarizing shift in the activation of Na_v1.8 channels via mechanisms that depend on PKA and PKC.^40,41 Thus, the −7.3 mV hyperpolarizing shift in activation of Na_v1.8 that we observed in this study could have resulted from inflammation-induced signalling cascades. Other inflammatory mediators, for example TNF-α, increase the current density of Na_v1.8.⁴² Thus, the inflammatory milieu that develops during chemotherapy may contribute to the modulation of Na_v1.8 channels in small DRG neurons. Although inflammatory mediators are known to increase the current density of the Na_v1.7 TTX-S channel and increase axonal trafficking of the channel in sensory neurons,⁴³ we did not detect changes in either current density or gating properties of the TTX-S currents in small DRG neurons from mice treated with vincristine.²² Taken together, these data point to possible Na_v1.8 isoform-specific regulation by vincristine-induced inflammatory response.

The majority of small-diameter DRG neurons from Na_v1.8 KO mice produce graded responses to depolarizing stimuli, with only 20% producing the all-or-none action potential.²⁶ Hyperpolarizing the resting potential of these neurons to −80 mV could restore the firing of action potentials because enough TTX-S channels could be rescued from inactivation at hyperpolarized membrane potentials.²⁶ Consistent with these earlier findings, our data showed that small DRG neurons from vincristine-treated Na_v1.8 KO mice only produced graded responses and that the threshold for firing an all-or-none action potential when neurons were held at −80 mV was not significantly different from control mice (Fig. 6). This is in agreement with our previous data showing that vincristine treatment has no detectable effect on TTX-S channels in small DRG neurons.²² Together with our published data showing that knockout of Na_v1.7, which contributes 70% of the TTX-S current in small DRG neurons,^30,31 did not diminish vincristine-induced mechanical allodynia,²² we now provide evidence supporting an important role of Na_v1.8 activation in the induction phase of vincristine-induced allodynia.

Vincristine-induced mechanical allodynia is likely a multi-factorial process, which involves multiple classes of ion channels and receptors in peripheral and central neurons. Transient receptor potential vanilloid 1 (TRPV1) and TRPV4,^19,20 the 5-hydroxytryptamine receptor 2A^20,21 T- and N-type Ca_V channels,^39,44 might be involved. To gain insights into vincristine-mediated effects on the expression levels of ion channels that regulate neuronal excitability in our animal model, we obtained RT-PCR array data showing that vincristine treatment did not cause significant changes in the expression levels of Na_v or HCN channels (Supplementary Table 1), while the expression levels of several K⁺ channels were significantly reduced, including K_Ca1.1, a large conductance calcium-activated potassium channel encoded by KCNMA1 (also termed BK or BK_Ca). The K_v2 and calcium-activated potassium channels are prominent contributors to action potential repolarization and afterhyperpolarization.^45,46 The BK_Ca currents in small DRG neurons were reduced and appeared to activate more slowly 3 days after complete Freund’s adjuvant injection.⁴⁷ Interestingly, Kimm et al.⁴⁵ demonstrated that inhibiting BK_Ca channels paradoxically hyperpolarized the AHP, which was attributed to the compensatory enhancement of slowly deactivating K_v2 current during the spike. This suggests that in addition to downregulation of BK channels at the transcription level (Supplementary Table 1), channel activity may also be suppressed by vincristine-induced inflammation, and the crosstalk with the K_v channels might result in the hyperpolarizing shift of AHP (Table 1) in small-diameter DRG neurons.

Na_v1.8 channels have an unequivocal role in DRG neuron firing²⁶ and pain,^28,48 and our data provide evidence for a role of Na_v1.8 in the development of vincristine-induced mechanical allodynia. However, previous studies have yielded conflicting results on the role of Na_v1.8 channels in vincristine-induced neuropathic pain. Levels of Na_v1.8 were reported to be upregulated using immunohistochemistry,²³ and pharmacological studies reported that vincristine-induced pain can be attenuated by pan Na_v channel blockers lidocaine and mexiletine but not TTX, findings that were interpreted as suggesting the involvement of TTX-R Na⁺ channels.^24,25 There are studies, however, that did not report amelioration of vincristine-induced pain in rats, in which Na_v1.8 expression levels were knocked down using antisense oligonucleotides or after pharmacological block of this channel.^29,49 In the current study, we observed an increase in the current density of Na_v1.8 and a hyperpolarizing shift in its activation, which is predicted to contribute to hyperexcitability of small DRG neurons and demonstrated a delayed induction of vincristine-induced mechanical allodynia in Na_v1.8 KO mice. The difference between our findings and those that did not report an effect of Na_v1.8 in vincristine-induced pain may be related to methodological differences including the dose of the drug delivered, the species (mouse versus rat) used, sex (mixed cohort versus male only) or the stages at which mechanical allodynia was assessed.

We have previously shown that the maintenance of vincristine-induced mechanical allodynia but not its induction is attenuated in conditional KO Na_v1.6^Nav1.8 mice, and, in parallel, an increase in the Na_v1.6 current density in medium-diameter but not small DRG neurons.²² Na_v1.6 channels are the main isoforms at nodes of Ranvier of myelinated fibres,⁵⁰ which could explain their critical role in the transmission of nerve impulses in these fibres that contribute to the maintenance of pain in this model. We now show a delay in the onset of vincristine-induced mechanical allodynia in Na_v1.8 knockout mice, with an increase in current density and a hyperpolarizing shift in activation of Na_v1.8 in small-diameter DRG neurons. It is notable that Na_v1.7 channels do not appear to play an important role in either of the two phases of vincristine-induced allodynia.²² Together, these data suggest a different role of these two classes of sensory neurons in the induction and maintenance of vincristine-induced mechanical allodynia and a possible contribution of Na_v1.6 (medium neurons) and Na_v1.8 (small neurons) in these processes. Recently, the isoform-specific small molecule inhibitor of Na_v1.8 VX-548 has shown promising results in treating acute pain in phase 2 clinical trials in patients with bunionectomy and abdominoplasty.⁵¹ An isoform-specific Na_v1.6 small molecule inhibitor NBI-921352 has shown antiepileptic seizures in rodent models⁵² and is currently being tested in clinical trials in patients with SCN8A developmental and epileptic encephalopathy syndrome (NCT04873869). These isoform-selective inhibitors could potentially be used as a monotherapy or in combination to treat vincristine-induced pain.

Data availability

All information necessary to evaluate the findings of the paper is included in the paper. Additional data can be provided by the corresponding author upon request.

Funding

This work was supported by grants RX003621 (S.D.H.) and Center Grant B9253-C (S.D.H. and S.G.W.) from the US Department of Veterans Affairs Rehabilitation Research and Development Service, and by a grant from the Thomson Family Foundation. The Center for Neuroscience and Regeneration Research is a collaboration of the Paralyzed Veterans of America with Yale University.

Competing interests

The authors report no competing interests.

Supplementary material

Supplementary material is available at Brain online.

References

Author notes