A total of 72 adult male Sprague-Dawley rats weighing 150–180 g and aged 6–8 weeks, purchased from the Experimental Animal Centre of Shenyang Pharmaceutical University (Shenyang, China), were used in all experiments. Rats were maintained under controlled environmental conditions at 23±2°C with a 12-h light/dark cycle and ad libitum access to food and water. All procedures were performed in accordance with the guidelines of the International Association for the Study of Pain (18). The study was approved by the ethics committee of Shenyang Pharmaceutical University.

The CCI model was induced as previously described (19,20). Briefly, under anesthesia with 3.5% chloral hydrate (10 ml/kg; Hebei Gaobeidian Chunguang Chemical Reagent Company, Gaobeidian, China) administered intraperitoneally (i.p.), the right sciatic nerve was exposed and loosely ligated with 4–0 chromic catgut (Shanghai Pudong Medical Supplies Co., Ltd., Shanghai, China) in four regions separated by ~1 mm, then the incision was closed with sutures. For the sham group, the right sciatic nerve was exposed without ligation. Pain-associated behavioral assessments were performed at the time-points of: −1 (prior to CCI surgery), 1, 3, 5, 7, 10, 14 and 21 days after CCI surgery.

The abnormal posture of each animal was evaluated using a subjective pain-associated behavioral grade as described previously (21) by an investigator who was blinded to the experimental conditions. Briefly, grades were determined as: 0, normal; 1, coiling of the toes; 2, valgus deformity of the paw; 3, partially weight bearing; 4, non-weight bearing and 5, avoidance of any contact with the hind paw.

The paw withdrawal response was measured using the Plantar Test meter (IITC Life Science Inc., Woodland Hills, CA, USA). Rats were placed individually in a clear, transparent box (17×11.5×14 cm) and allowed to acclimate for 15 min prior to the assessment. An infrared beam of a radiant heat source was applied to irradiate the plantar surface of the hind paw through the glass plate until the rat withdrew or contracted its paw. The withdrawal thresholds were measured in each hind paw and the ipsilateral hind paw was assessed at 15 min intervals, while the contralateral hind paw was assessed at 5 min intervals. A 20 sec limit was imposed to prevent tissue damage. Each rat was assessed five times and the mean value was expressed as the thermal withdrawal latency (22).

The paw withdrawal response to mechanical stimuli was measured using the Electronic von Frey Anesthesiometer (IITC Life Science Inc.). Rats were placed individually into wire mesh-bottom boxes (20×14×16 cm) and allowed to acclimate for 30 min prior to assessment. The probe was placed beneath the plantar surface of the hind paw and the force was increased until the rat was observed to vellicate its paw. The maximum force was recorded at the time of paw withdrawal. Withdrawal thresholds were measured in each hind paw and the ipsilateral hind paw was assessed at 30 sec intervals, while the contralateral hind paw was assessed at 15 sec intervals. Each rat was assessed five times and the average value was expressed as the mechanical withdrawal threshold (23).

L_4-6 DRGs, quickly dissected from recently sacrificed animals, were fixed with 10% formalin overnight at 4°C. Following embedding the tissue in paraffin, a series of 5-µm sections were cut for immunofluorescence and H&E staining. Paraffin sections were treated with dimethylbenzene solution I for 15 min, dimethylbenzene solution II for 15 min, dimethylbenzene:pure ethanol (1:1) solution for 2 min, 100% ethanol I for 5 min, 100% ethanol II for 5 min, 95% ethanol solution for 3 min, 90% ethanol solution for 1 min, 85% ethanol solution for 1 min, 75% ethanol solution for 1 min, 50% ethanol solution for 1 min, running water for 2 min, haematoxylin solution for 2 min, running water for 1 min, 1% hydrochloric acid ethanol solution for 20 sec, running water for 5 min, Eosin solution for 30 sec, running water for 30 sec, 75% ethanol solution for 30 sec, 85% ethanol solution for 20 sec, 95% ethanol solution I for 1 min, 95% ethanol solution II for 1 min, 100% ethanol for 2 min, 100% ethanol II for 2 min, dimethylbenzene solution I for 2 min, dimethylbenzene solution II for 2 min then dimethylbenzene solution III for 2 min. The primary antibodies polyclonal rabbit Na_v1.8 (ASC-016; 1:200; Alomone Laboratories Ltd., Jerusalem, Israel), monoclonal mouse neurofilament (NF)200 (N0142; 1:200; Sigma-Aldrich, St. Louis, MO, USA) and polyclonal rabbit Na_v1.9 (ASC-017; 1:200; Alomone Laboratories Ltd.) were administered for immunofluorescence staining and used to incubate the sections overnight at 4°C. Following three washes with phosphate-buffered saline containing Tween-20, the sections were incubated with anti-mouse IgG fluorescein isothiocyanate-conjugated antibody produced in goat (F0257) and anti-rabbit IgG (whole molecule)-Cy3 antibody produced in sheep (C2306) (1:100; Sigma-Aldrich) for 1 h at 37°C. Images were captured under an inverted fluorescence microscope (Olympus BX40; Olympus, Tokyo, Japan) and imported into Image pro plus 6.0 software (Media Cybernetics, Silver Spring, MD, USA) for further analysis. H&E staining was performed according to the manufacturer's instructions (Sigma-Aldrich). Nuclear material within the nucleus was stained a deep purple/blue, while the cytoplasmic material, including connective tissue and collagen appeared orange/pink.

At 3, 7, 14 and 21 days after CCI, rats were sacrificed via anesthesia with 3.5% chloral hydrate (10 ml/kg, i.p.). The ipsilateral and contralateral L_4–6 DRGs were rapidly removed and placed into Eppendorf tubes. Total RNA was extracted using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) and the cDNA was reverse transcribed using the PrimeScript^® RT reagent kit (Takara Bio Inc., Otsu, Japan) according to the manufacturer's instructions. PCR was performed in a 25-µl reaction mixture containing 2 µl templates, 12.5 µl SYBR^® Premix Ex Taq™ (2X), 0.5 µl ROX reference dye II (50X) and 0.4 µM primer for each gene. The thermal cycling conditions comprised 30 sec polymerase activation at 95°C, 40 cycles of 15 sec denaturation at 95°C and 1 min at 60°C for annealing and extension. A dissociation curve was used to determine the amplification specificity. The primer sequences (24) were as follows: Na_v1.8 (GenBank accession number, U53833) forward, 5′-GACTCCCGGACAAATCAGAA-3′ and reverse, 5′-AGCAGCGACCTCATCTTCAT-3′; Na_v1.9 (GenBank accession number, AF059030) forward, 5′-TCTCCACCCCTACCTCACTG-3′ and reverse, 5′-CGTTCAGCCAAAAACACAGA-3′; and GAPDH forward, 5′-TGCCAAGTATGATGACATCAAGAAG-3′ and reverse, 5′-AGCCCAGGATGCCCTTTAGT-3′. The threshold cycle values of Na_v1.8 and Na_v1.9 mRNA were measured and normalized to GAPDH, and then expressed as a relative ratio. The M×3000P qPCR system was used (Agilent Technologies, Inc., Santa Clara, CA, USA).

Bilateral L_4–6 DRGs were dissected and total proteins were extracted via homogenization in ice-cold lysate buffer (Thermo Fisher Scientific, Waltham, MA, USA). Samples (30–50 µg) were separated on 8% SDS-PAGE separation gels (Amresco, Boise ID, USA) and subsequently transferred onto polyvinylidene difluoride membranes (Merck Millipore, Boston, MA, USA). The membranes were blocked in 5% skimmed milk solution at room temperature for 2 h and then rabbit polyclonal Na_v1.8 (1:500; Alomone Laboratories Ltd.), Na_v1.9 (1:500; Alomone Laboratories Ltd.) and mouse monoclonal β-actin (1:500; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) primary antibodies were used to incubate the samples overnight at 4°C. The target bands were detected with secondary horseradish peroxidase (HRP)-labeled goat anti-rabbit (1:10,000; ZB-2301) or HRP-labeled goat anti-mouse IgG (1:5,000; ZB2305) (Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China) antibodies for 1 h at room temperature. The band intensity of Na_v1.8 and Na_v1.9 was normalized to that of β-actin and expressed as a relative ratio.

Rat DRG neurons were acutely dissociated as previously described (25). The samples were superfused at a rate of 3 ml/min and all patch clamp recordings were performed using an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA, USA), filtered at 1 kHz and digitally sampled at 10 kHz at room temperature (23±2°C). Clampfit 10.0 software (Molecular Devices) was used for data acquisition and analysis.

The bath solution for the Na_v1.8 currents contained (in mM): 140 NaCl, 1 MgCl₂, 3 CaCl₂, 5 KCl (Tianjin Bodi Chemical Co. Ltd., Tianjin, China), 10 tetraethylammonium (TEA)-Cl, 1 4-aminopyridine, 0.2 CdCl₂, 10 4-(2-hydroxyethyl)-1-piper-azineethanesulfonic acid (HEPES), 10 glucose (Tianjin Bodi Chemical Co. Ltd.), 0.001 TTX (Hebei Fisheries Research Institute, Qinhuangdao, China), pH 7.3 with NaOH (Tianjin Bodi Chemical Co. Ltd.). The bath solution for the Na_v1.9 currents contained 30 NaCl, 20 TEA-Cl, 90 choline chloride, 3 KCl, 1 CaCl₂, 1 MgCl₂, 10 HEPES, 10 glucose, 0.1 CdCl₂, 0.001 TTX, pH 7.3 with Tris base (all purchased from Sigma-Aldrich unless specified). The pipette solution used for recording the DRG neuron Na_v1.8 and Na_v1.9 currents contained (in mM): 140 CsCl, 10 TEA-Cl, 10 ethylene glycol tetraacetic acid (EGTA; Amresco), 10 HEPES, pH 7.2 with CsOH and 135 CsF, 10 NaCl, 10 HEPES, 5 EGTA, 2 adenosinetriphosphate bisodium, pH 7.2 with CsOH, respectively. The pipettes, fabricated with a P-97 puller (Sutter Instruments, Novato, CA, USA), had resistances of 3–5 MΩ when filled with pipette solution.

Data were analysed using SPSS 16.0 (SPSS, Inc., Chicago, IL, USA) with a one-way analysis of variance. All measurements are expressed as the mean ± standard error. P<0.05 was considered to indicate a statistically significant difference.

The CCI model induced the typical features of neuropathic pain, which were assessed via behavioral grade, the Plantar Test meter and the Electronic von Frey Anesthesiometer. Rats with nerve injury (CCI) exhibited curling, eversion, partial weight bearing or non-weight bearing on the right injured side, while they exhibited normal behavior on the left uninjured paw. The score for the right paw increased significantly from the first day after surgery (data not shown).

The sensitivity to heat and mechanical stimulus revealed certain discrepancies among the experimental groups. In the CCI model group, the thermal withdrawal latency and mechanical withdrawal threshold of the right injured paw decreased markedly from day 1 after injury, with the maximal level of hypersensitivity at 7 days after surgery, while observable changes were not observed in the sham surgery group (Fig. 1A and C). By contrast, no differences were observed in the contralateral paw withdrawal in all groups (Fig. 1B and D).

In peripheral nerve injury, the majority of neuronal cell death is observed in peripheral nerve lesions, which are considered to be a major factor in the poor clinical outcome following the injury (26). With H&E staining, ipsilateral and contralateral DRGs to the injury (Fig. 2) were analyzed histologically. In each contralateral DRG group (Fig. 2G–L), the neurons exhibited a normal cellular appearance, including a large, round nucleus, clear cell boundaries and even chromatin distribution. However, neurons on the ipsilateral side to the injury exhibited a range of morphologies at different time-points (Fig. 2C–F), although all observed cells had abnormal nuclei exhibiting pyknosis, membrane irregularities or even vacuolation, with the maximal magnitude of neuronal death at 21 days after surgery.

The neurofilament NF200 is preferentially expressed in large DRG neurons and is a useful marker for this population (7). When co-incubated with antibodies against Na_v1.8 and Na_v1.9 in an immunofluorescence assay, it was observed that Na_v1.8 and Na_v1.9 subunits were primarily located in the small and medium diameter DRG neurons (Fig. 3A).

RT-qPCR and western blot analyses were performed at 3, 7, 14 and 21 days after the CCI surgery instead of at an earlier point, in order to avoid potential error due to the effects of post-surgical pain. CCI induced a marked downregulation of the Na_v1.8 transcript at 7 days (0.76±0.05; P<0.01), 14 days (0.71±0.07; P<0.01) and 21 days (0.76±0.02; P<0.001) and a substantial downregulation of Na_v1.9 transcript at 14 days (0.81±0.04; P<0.01) and 21 days (0.806±0.02; P<0.001) in L_4–6 ipsilateral DRGs (Fig. 3B); however, no significant differences were observed among groups in the contralateral DRGs (Fig. 3C; P>0.05). Representative images of Na_v1.8 and Na_v1.9 protein expression within the ipsilateral and contralateral DRGs are shown in Fig. 3D. Consistent with these findings, RT-qPCR analysis revealed that the relative ratio of band intensity of Na_v1.8 protein in ipsilateral DRGs was significantly reduced at 7 days (0.62±0.04; P<0.001), 14 days (0.51±0.04; P<0.001) and 21 days (0.63±0.04; P<0.001) after CCI treatment, while Na_v1.9 protein expression following CCI exhibited a decrease at 14 days (0.88±0.06; P>0.05) and 21 days (0.89±0.06, P>0.05; Fig. 3E). However, the differences were not statistically significant when compared with the sham group. In addition, there were no detectable differences among contralateral DRG neurons from the rats subjected to CCI (Fig. 3F). These results markedly suggested that Na_v1.8 and Na_v1.9 sodium channels have distinct roles following CCI treatment.

The small- and medium-diameter DRG neurons (12–25 µm) were selected as the main focus of the present study. Differences in the voltage protocols and pharmacological inhibition by TTX were used to evoke Na_v1.8 and Na_v1.9 inward currents (Fig. 4A). Following CCI surgery, the sodium current densities mediated by Na_v1.8 and Na_v1.9 decreased ~50% (Fig. 4B; between −27.10±6.03 and −14.75±4.01 pA/pF; P<0.01) and 18% (Fig. 4B; between −6.90±2.89 and −5.67±2.00 pA/pF; P>0.05), respectively.

Steady-state activation curves were constructed as described from current-voltage curve experiments. The normalized activation curves of Na_v1.8 and Na_v1.9 were fitted with the Boltzmann function expressed as: G/G_max = 1/{1+exp[(V−V_1/2)/K, where V_1/2 is the membrane potential at half-activation and K represents the slope factor. Characterization of activation curves revealed that CCI treatment caused a depolarizing shift of 5.3 mV and 1.44 mV in Na_v1.8 (Fig. 4E; V_1/2 between −15.84±0.21 and −10.51±0.24 mV) and Na_v1.9 (Fig. 4F; between −40.28±1.02 mV and −38.84±2.07 mV) steady-state activation curves, respectively.

The voltage-dependent inactivation of the Na_v1.8 current is shown in Fig. 4E. The solid lines were fitted with the Boltzmann equation, I/I_max = 1/{1+exp[(V−V_1/2)/K, where V_1/2 is the membrane potential when I/I_max = 0.5 and K represents the slope factor. There was a 10-mV depolarizing shift in the midpoint of inactivation (Fig. 4E; between −23.85±0.74 and −33.73±0.91 mV). This change reflected a parallel shift in the availability curve of the current as no change was observed in the slope factor (between −6.82±0.63 mV and −6.64±0.76 mV). However, no significant differences were identified between injured and control DRG neurons with respect to the biophysical properties of Na_v1.9 (Fig. 4F).