The animals were randomly divided into two experimental groups (14 in each group). One group was subjected to spinal nerve ligation injury (SNL group), and the other group was subjected to no operation (naïve group). SNL injury was performed as previously described [1]. Briefly, a midline incision above the lumbar spine and deep dissection through the paraspinal muscles exposed the left L6 transverse process, and the process was then removed. The left L5 and L6 spinal nerves were isolated, and both nerves were ligated tightly with 6-0 black silk distal to the DRG and proximal to the formation of the sciatic nerve.

Animals were placed in individual plastic cages with wire mesh bottoms and allowed to acclimate for 20 min, and behavioral testing was done.

To test for mechanical allodynia, a von Frey filament was applied vertically to the plantar surface of the hind paw ipsilateral to the SNL injury. A series of eight calibrated von Frey filaments (0.41, 0.70, 1.20, 2.00, 3.63, 5.50, 8.50, and 15.10 g) was applied serially to the paw in ascending order of strength with sufficient force to cause gentle bending against the paw, and held for six seconds. A sharp withdrawal or paw flinching was considered a positive response. The mechanical stimulus producing a 50% likelihood of withdrawal was determined by using the up-down method as previously described [9].

To test for cold allodynia, acetone was applied to the plantar surface of the hind paw ipsilateral to the SNL injury, five times to each paw at intervals of 5 min. A prompt foot withdrawal response to the acetone application was interpreted as a sign of cold allodynia. The frequency of paw withdrawal was expressed as a percentage (the number of paw withdrawals divided by the total number of trials, times 100).

Baseline testing was done before surgery to verify that the animals had normal responses. All animals were tested on postsurgical days 4 and 7. All testing was done by a single examiner blinded to the experimental groups.

After behavioral testing, 7 rats of both SNL and naïve groups were subjected to histologic findings. The animals were anesthetized with zoletil 12.5 mg and xylazine 3 mg intraperitoneally and perfused transcardially with 4% buffered paraformaldehyde on postoperative day 7. The left L5 DRG was removed, postfixed in 4% paraformaldehyde, and subsequently embedded in paraffin.

DRG samples embedded in paraffin were sectioned (4 µm), and placed on slides. Sections were deparaffinized and rehydrated in a graded ethanol series. After blocking with 5% goat serum, 0.3% Triton X-100, and 1% BSA in PBS-T, double-label immunofluorescence was performed by incubating the paraffin sections with cleaved caspase-3 (Asp 175) antibody (1:200, #9661, Cell Signaling Technology Inc., Danvers, MA, USA) and GFAP antibody (1:200, #3670, Cell Signaling) or NeuN antibody (1:200, MAB377, Chemicon, Billerica, MA, USA), which were applied overnight at 4℃, followed by incubation with Alexa Fluor 555 goat anti-rabbit IgG (red; 1:500; Invitrogen) and Alexa Fluor 488 goat anti-mouse IgG (green; 1:500; Invitrogen, Carlsbad, CA, USA) for 2 hours at room temperature. Sections were rinsed in PBS-T and mounted with DAPI. The immunoreactions in the DRG were observed using a confocal microscope. Two slices of the DRG for each evaluation were used to determine the number of NeuN-immunoreactive (IR) or GFAP-IR cells in the DRG. Co-labeled cells with a clearly definable greenish cell border and reddish nucleus were counted as a marker of apoptotic cells at 400× magnification for each section. The number of co-labeled cells was divided by the total number of NeuN-IR or GFAP-IR cells within the field and multiplied by 100.

ELISA immunoassay findings were analyzed using 7 rats of both SNL and naïve groups that were not used for behavioral testing. The animals were decapitated rapidly under anesthesia on the 7^th day of experiment, and the DRGs of the left L5 and L6 segments were removed and frozen in liquid nitrogen. Samples were homogenized in ice-cold RIPA buffer (R0278, Sigma, St. Louis, MO, USA) and protease inhibitor cocktail (P8340, Sigma, St. Louis, MO, USA), and centrifuged (13,000 rpm for 20 min) to obtain protein extracts. Commercially available ELISA kits were used for assessing TNF-α (R&D Systems, MN, USA, sensitivity: 5 pg/ml). The Bradford protein assay was used to measure total protein concentration in the tissue supernatant. Microplates were measured using a VICTOR3 V Multilabel Counter (1420, Perkin Elmer), and data were standardized as picograms of TNF-α to 200 µg of total protein in the supernatant.

All statistical analyses were performed using SPSS for Windows Version 12.0 (SPSS Inc., Chicago, IL, USA). Data are presented as mean ± SD or SEM. The differences between the groups were compared using the t-test. P values less than 0.05 were considered statistically significant.

The mechanical withdrawal threshold and prompt paw withdrawal response to acetone application are shown in Fig. 1. SNL injury produced significant mechanical and cold allodynia after 4 and 7 days (P < 0.05) compared to baseline.

As shown in Fig. 2, the NeuN-IR cells (DRG neurons, green) and GFAP-IR cells (satellite cells, green) were observed in the DRG (A, D, G, J). The GFAP-IR satellite cells were localized around the DRG neurons. Caspase-3 immunoreactivity (red) was observed in neurons (B, E), astrocytes (H, K). After SNL, activated caspase-3 co-localized with NeuN (C, F) and GFAP (I, L). Co-localized caspase-3 expression was significantly increased in DRG astrocytes and neurons at postoperative day 7 compared with naïve DRG. (P < 0.05, Fig. 3).

Apoptotic cells were observed in the left L5 DRGs in the SNL group. The number of apoptotic cells in the SNL group (13.7 ± 1.1) was significantly higher than in the naïve group (3.5 ± 0.8, P < 0.001) on postoperative day 7. TNF-α protein levels were also increased in the DRG following SNL injury (12.7 ± 3.2 pg/100 µg, P < 0.001), when compared with naïve rats (4.1 ± 1.4 pg/100 µg).