Animals were obtained from the Model Animal Research Center of Nanjing University. Sixty 4–6-weeks old adult male C3H/HeJ mice, weighing 20–25 g, were used in the present study. The mice were maintained in a temperature-controlled (21±1°C) room with 12-h alternating dark/light cycles. The experimental protocols were approved by the Animal Care and Use Committee at the Medical School of Nanjing University. All efforts were made to minimize animal suffering and to reduce the number of animals used.

The sequences of the sense and antisense KIF17 ODNs were designed as previously reported (15). The sequences used were: sense, 5′-TTCGGTGGTGAGCCTCTG-3′, and antisense, 5′-CAGAGGCTCACCACCGAA-3′. The ODNs were modified with phosphorothioate to improve stability and were purchased from Sangon Biotechnology Co. (Shanghai, China). The ODNs were dissolved in saline, and 5 μl were injected intrathecally (i.t.) via lumbar puncture at the intervertebral space of L4–5 or L5–6. Injections occurred once per day for six consecutive days starting from day 14 after tumor cell implantation (TCI). KIF17 sense ODN (5 μg), antisense ODN (5 μg) and control saline (5 μl) were injected i.t.

TCI was achieved by injecting the osteosarcoma NCTC 2472, purchased from the American Type Culture Collection (Manassas, VA, USA; no. 2087787) into the intramedullary space of the right femur to induce bone cancer in mice. According to the ATCC recommendations, the tumor cells were cultured in NCTC 135 medium (Sigma-Aldrich, St. Louis, MO, USA) with 10% horse serum (Gibco, Grand Island, NY, USA). The cell incubator (Thermo Forma, Inc., Marietta, OH, USA) was kept at 37°C and had an atmosphere of 5% CO₂ and 95% air humidity. The cells were passaged twice per week.

The protocol described by Schwei et al was used to induce the murine model of bone cancer pain (18). The surgery was performed under anesthesia with pentobarbital sodium (50 μg/kg, intraperitoneally, i.p.). Gonarthrotomy was performed, exposing the femur condyles. A needle was employed to perforate the cortex and a hole was drilled using a dental bur. Twenty microliters of α-minimal essence medium (α-MEM), containing 2×10⁵ NCTC 2472 cells, were injected into the intramedullary space of the right femur to induce bone cancer pain. α-MEM (20μl) with no cancer cells was injected in the sham surgery. The injection hole was subsequently plugged with a dental amalgam to tightly seal the injection hole in the bone and prevent the escape of tumor cells from the bone. The wound was irrigated with saline and then closed. The surgical operations were conducted under aseptic conditions.

The mice were randomly divided into the sham (n=20) and TCI (n=40) groups. On post-operation day 14, after the pain-associated behaviors were tested, the TCI mice were randomly divided into the TCI + saline group (n=10), TCI + KIF17 sense ODN group (n=10) and TCI + KIF17 antisense ODN group (n=10). The number of spontaneous lifting actions and mechanical allodynia events was assessed in mice with bone cancer to evaluate bone cancer-associated pain. The tests were performed during the light phase; prior to each test, the mice were allowed to acclimatize for a minimum of 30 min.

The mice were placed in individual plexiglass compartments (10×10×15 cm). The behaviors of the mice were evaluated by measuring the number of spontaneous flinches of the right hind limb over a 2-min period of observation (19). Five periods were recorded for each mouse. Positive responses of the right hind limb involved a lifting movement and were not related to walking or grooming.

Mechanical allodynia was measured using von Frey filaments (Stoelting, Wood Dale, IL, USA) with a protocol similar to that previously described (20). Briefly, the animals were individually placed beneath transparent plexiglass compartments (10×10×15 cm) on a metal-mesh floor (graticule: 0.5×0.5 cm). The minimum pressure was set to 0.16 g, and the maximum pressure was set to 2.0 g. The duration of each stimulus was 6–8 sec, and the interstimulus interval was a minimum of 15 sec. Brisk withdrawal of the paw and paw flinching were considered positive responses. The paw withdrawal mechanical threshold (PWMT) was determined by sequentially increasing and decreasing the stimulus strength (the ‘up- and -down’ method), which is the lowest von Frey filament that had three or more positive responses. Every mouse was tested five times per stimulus strength.

The spinal cord L3–5 segments were immediately removed from the deeply anesthetized mice (pentobarbital sodium, 50 μg/kg, i.p.) and stored in liquid nitrogen. The tissue samples were homogenized in lysis buffer containing protease and phosphatase inhibitor. The samples were centrifuged at 12,000 rpm for 10 min at 4°C, and the supernatant was removed. The BCA method was used to determine the protein concentration (Kaiji Biotechnology, Nanjing, China). The samples were boiled at 100°C for 10 min with 1X loading buffer. Protein lysates (50 μg) were separated using 10% SDS-PAGE and run for 30 min at 80 V and 90 min at 120 V. The lysates were then transferred into polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA) at 200 mA for 2 h (depending on the molecular weight of the target protein). The membranes were blocked in 5% non-fat milk in Tris-buffered saline with Tween-20 (TBST) for 2 h at room temperature and then incubated with primary rabbit antibodies [anti-KIF17 (1:1,000), anti-NR2B (1:1,000), anti-mLin-10 antibody (1:1,500) all from Abcam, Cambridge, UK, and anti-β-actin (1:5,000) Cell Signaling Technology, MA, USA] diluted in blocking buffer. The following day, after washing six times with TBST for 1 h, the membranes were incubated with the secondary goat anti-rabbit antibody conjugated with horseradish peroxidase (1:5,000; Cell Signaling Technology) for 2 h at room temperature and then washed again for 1 h. Immunoblots were detected using the ECL system (Millipore). The images of western blot analysis products were collected and analyzed using Quantity One V4.40 (Bio-Rad, Hercules, CA, USA). Each of the control and treatment groups was first standardized with β-actin.

Data are presented as the means ± SD (standard deviation). The data were analyzed using repeated measures analysis of variance. The LSD post-hoc tests were performed to determine the sources of the differences when significant main effects were observed. P values of ≤0.05 was considered significant.

Prior to the operation, there was no significant difference between groups in the number of spontaneous flinches or in the PWMT (P>0.05). Compared with the baseline before the operation, on day 4 after the operation, the spontaneous flinches increased and the PWMT decreased in the TCI and sham groups (P<0.05). However, the pain behaviors in the sham group recovered to the baseline level at day 7. By contrast, TCI induced an increase of the number of spontaneous flinches at day 7 (4.40±0.58), which gradually increased over time until day 14, when the P-value was 11.40±0.96 (Fig. 1A). The PWMT decreased by day 7 (1.00±0.21 g), and on day 14, the P-value was 0.40±0.12 g (P<0.05) (Fig. 1B).

The results confirmed that in the sham and TCI groups, when compared with the baseline prior to operation, the expression of KIF17, mLin-10 and NR2B increased beginning on day 4 after the operation (P<0.05). However, the sham group recovered to the baseline level on days 7–14 (P>0.05). Compared with the sham group, the expression of KIF17, mLin-10, and NR2B in the TCI group was increased at day 7 and peaked on day 14 after the operation (P<0.05). At day 14, the P-values were 1.77±0.10, 1.64±0.27 and 2.16±0.27, respectively (Fig. 2).

In the TCI group injected with saline and KIF17 sense ODN and the sham group injected with saline, the pain behaviors did not change, but remained at a regular level (P>0.05). Compared with the pre-lesion P-value prior to dosing and the TCI group injected with saline or KIF17 sense ODN, from 16 to 28 days after the operation, the spontaneous flinches were decreased and the PWMT was significantly increased in the TCI + KIF17 antisense ODN group (P<0.05). At day 16, the P-values were 3.38±0.47 and 1.28±0.35 g. At day 28, the P-values were 8.98±1.15 and 0.70±0.19 g, respectively. However, at day 31, there was no significant difference between TCI groups in terms of the number of spontaneous flinches or the PWMT (P>0.05). Although injected KIF17 antisense ODN produced a significant transient inhibition of the spontaneous flinches and PWMT, compared with the sham + saline group, the difference remained significant (P<0.05) (Fig. 3).

We examined whether the effect of KIF17 antisense ODN during the development of bone cancer pain caused the reduction of NR2B protein within the spinal cord. Western blot analysis demonstrated that compared with the other TCI groups, the expression levels of KIF17, mLin-10 and NR2B in the spinal cord were significantly reduced following the application of KIF17 antisense ODN on day 19 and for the 6 days after intrathecal administration (P<0.05). Compared with the sham + saline group, the expression of KIF17, mLin-10 and NR2B remained upregulated (P<0.05). The P-values were 0.86±0.05, 0.82±0.07 and 1.07±0.17, respectively. No significant difference was identified between the saline group and the KIF17 sense ODN group for TCI (P>0.05) (Fig. 4). However, at 31 days after the operation, there was no significant difference among the three TCI groups in terms of the expression levels of KIF17, mLin-10 and NR2B in the spinal cord (P>0.05). The P-values were 1.17±0.19, 1.21±0.11 and 1.12±0.09, respectively (Fig. 5).