Male inbred Sprague-Dawley rats (weight 250–300 g) were purchased from Central South University and were housed at the Xiangya Hospital BioResources Centre. Animals were placed in a quiet, temperature (22±2°C) and humidity (60±6%) controlled room with a 12:12 h light-dark cycle (light beginning at 8 a.m.) and all tests were performed during the light phase of the cycle. Pharmaceutical-grade etanercept was purchased from Amgen (Thousand Oaks, CA, USA). Anti-HMGB1 (sc-12523) antibody was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Anti-p38 mitogen-activated protein kinase (p38 MAPK; #8690), anti-phospho-p38-MAPK (Thr180/Tyr182; #9211) and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH; #2118) antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA). All secondary antibodies were from Jackson ImmunoResearch Laboratories (West Grove, PA, USA). All chemicals of reagent grade were purchased from Sigma (St. Louis, MO, USA).

CCI was induced according to the method of Bennett and Xie (10). Briefly, each animal was anesthetized by an intraperitoneal injection of sodium pentobarbital at a dose of 60 mg/kg. The common sciatic nerve was exposed and freed from adherent tissue at the mid-thigh by separating the biceps femoris muscles by blunt dissection. Four loose ligatures were placed 1 mm apart using chromic gut suture (4-0 absorbable suture; Jorgensen Laboratories, Inc., Loveland, CO, USA). The animals were randomly assigned to seven groups (n=20/group): untreated, sham only (animals subjected to sham surgery only), sham/saline (animals subjected to sham surgery plus intrathecal injection of 20 μl saline every two days from two days before surgery), sham/etanercept [animals subjected to sham surgery plus intrathecal injection of 20 μl (100 μg) etanercept every two days from two days before surgery], CCI only (animals subjected to CCI surgery only), CCI/saline (animals subjected to CCI surgery plus intrathecal injection of 20 μl saline every two days from two days before surgery) and CCI/etanercept group [animals subjected to CCI surgery plus intrathecal injection of 20 μl (100 μg) etanercept every two days from two days before surgery]. This study was conducted in accordance with our institutional guidelines on the use of live animals for research and the experimental protocol was approved by the Laboratory Animal Users Committee at Xiangya Hospital, Central South University.

Thermal hyperalgesia was measured according to the Hargreaves test (1) using a plantar analgesia instrument (Stoelting, Wood Dale, IL, USA) every day from one day before surgery to 13 days after surgery. The radiant infrared heat source stimulus intensity was set to IR50 and the cut-off time was set at 15 sec. The rats were placed on a glass platform and allowed to habituate to the testing chambers for a minimum of 15 min prior to each testing session. The thermal stimulus was applied to the plantar surface of the paw. Thermal thresholds were defined as the latency in seconds at the first pain behavior, which includes paw withdrawal, flinching, biting and/or licking of the stimulated paw. The readings for all animals were averaged and the mean and standard error of the mean were determined for each treatment group. Mechanical allodynia was measured using von Frey monofilaments (Stoelting) with varying stiffness (2.0–15.0 g) every day from one day before surgery to 13 days after surgery. The rats were placed on a perforated metallic platform and allowed to habituate to their surroundings for a minimum of 15 min before testing. The paw withdrawal threshold response was determined by a sequential increasing and/or decreasing of the stimulus strength.

On days 3, 7 and 13 after surgery, four randomly selected rats were sacrificed at each time point and DRG neuron cells were isolated from the enlarged part of the lumbar spinal cord as previously described (11). Cells were used for experiments 24–48 h after isolation. DRG neurons used for mRNA extraction were stored at −80°C immediately after isolation. RNA samples were prepared using TRIzol reagent followed by purification with Turbo DNA-free system (Ambion, Austin, TX, USA). The cDNAs were synthesized using SuperScript II reverse transcriptase (Invitrogen, Carlsbad, CA, USA). Real-time quantitative PCR was performed on the LightCycler thermal cycler system (Roche Diagnostics, Indianapolis, IN, USA) using a SYBR-Green I kit (Roche Diagnostics) as described by the manufacturer. Each result was normalized against that of the housekeeping gene GAPDH in the same sample. The primers used were as follows: for rat HMGB1, 5′-GTACGGTACCAAGTGCATTT TGGAGGAATT-3′ (forward) and 5′-GTACAAGCTTGTACT GCAATGGCTGTGAGA-3′ (reverse) and for rat GAPDH, 5′-AAGCCCATCACCATCTTCCA-3′ (forward) and 5′-CCTGCTTCACCACCTTCTTG -3′ (reverse). Each experiment was repeated twice in triplicate.

Immunoblotting was performed as described previously with respective antibodies (12). Briefly, DRG neuron cells were lysed in 0.1% Nonidet P-40 lysis buffer [(0.1% Nonidet P-40, 50 mM Tris-HCl (pH 7.4), 150 mM NaCl and 1 mM ethylenediamine tetraacetic acid (EDTA)]. Equal amounts of lysates were loaded onto 10% sodium dodecyl sulphate (SDS)-polyacrylamide gels and the proteins were blotted onto a polyvinylidene difluoride microporous membrane (Millipore, Billerica, MA, USA). The membranes were incubated for 1 h with a 1/1,000 dilution of anti-HMGB1, anti-p38 MAPK, anti-phospho-p38 MAPK (Thr180/Tyr182) or anti-GAPDH antibodies and then washed and revealed using secondary antibodies with a horseradish peroxidase conjugate (1/5,000, 1 h). Peroxidase was revealed with an ECL kit (GE Healthcare Life Sciences, Piscataway, NJ, USA). The proteins were quantified before being loaded onto the gel and equal loading of extracts was verified by Ponceau coloration.

Statistical analyses were performed with SPSS (SPSS Inc., Chicago, IL, USA) for Windows 10.0. Data values were expressed as the mean ± standard deviation. Comparisons of means among multiple groups were performed with one-way analysis of variance (ANOVA) followed by post hoc pairwise comparisons using the least significant difference method. P<0.05 was considered to indicate a statistically significant difference.