XZP is composed of the six aforementioned traditional Chinese medicines. High performance liquid chromatography (HPLC) analysis has shown that XZP contained tetrahydropalmatine (0.0568%), imperatorin (0.0041%), isoimperatorin (0.0122%), coptisine (0.0358%) and palmatine chloride (0.112%) (16).

Animal use and care protocols were reviewed and approved by the Animal Care and use Committee of the China Academy of Chinese Medical Sciences, Beijing, China. All animal studies were carried out in accordance with the guidelines of the International Association for the Study of Pain (35). Female Wistar rats, weighing 150–180 g, were obtained from the Department of Experimental Animal Sciences, Peking University Health Science Center (Beijing, China). Individual rats were housed in specific pathogen-free (SPF) facilities under strict temperature (24±1°C) and humidity (60%) control on a 12/12 h light/dark cycle with free access to standard food and tap water.

Wister rat Walker 256 breast sarcocarcinoma cells were prepared, and a rat bone cancer pain model was established as previously reported (32). Briefly, Wistar rats were intraperitoneally (i.p) injected with Walker 256 cells (2×10⁷ in 0.5 ml of Hank's solution). Seven to 14 days later, cells in the produced ascites were collected. After being anesthetized, 50 rats were injected with 10⁵ Walker 256 cells in 10-µl of Hank's solution into the left tibia of the hind paw. Then, the injection site was covered by bone wax and the wound was closed. Sham control rats (n=10) received the same surgery and volume of vehicle injection. All rats were subjected to the same post-operative care.

Rats with bone cancer were randomly divided into five groups (n=10/group): one placebo control group, topically treated with inert paste; three XZP groups, treated with low (15.75 g/kg), medium (31.5 g/kg) and high (63 g/kg) doses of XZP; one OPG group, subcutaneously injected with 5 mg/kg of OPG twice per day. Sham control rats were also topically treated with inert paste. XZP and placebo pastes were evenly applied on the skin of tumor-bearing tibias; which were then covered with gauze and a layer of plastic film, sealed and fixed with desensitized adhesive plaster. Treatments were performed twice a day at 8:00 am and 20:00 pm for a total of 21 days, beginning at day one after cancer cells were implanted into the tibias (17).

Cancer-related osteolytic lesions in rat tibias were examined by X-ray radiology at day 21 after cell inoculation. Rats were anaesthetized with sodium pentobarbital (45 mg/kg) by i.p injection and exposed to X-ray (Emerald 125) at 40 kVp for 1/20 sec. X-ray film (Henry Schein blue sensitive film; Shanghai Han Ruixiang Trade Co., Ltd.) was developed with a film developer (Konica SRX-101; Suzhou Zhongyi Electronic Technology Co., Ltd.). Tibias were scanned and reconstructed with an isotropic voxel size of 8-µm on a micro-CT system (eXplore Locus SP; GE Medical Systems, Zhongtong Shanghai Automation & Electrics Co., Ltd.). Reconstructed 3D images of femurs were analyzed using Microviewer (GE Medical Systems, Zhongtong Shanghai Automation & Electrics Co., Ltd.) as previously described (36). Bone mineral density (BMD) of left tibias in individual rats were measured by dual energy X-ray absorptiometry using a PIXImus Mouse Densitometer; and bone mineral contents were calculated using small animal analysis software (both from GE Lunar Medical System, Zhongtong Shanghai Automation & Electrics Co., Ltd.) at day 21, after cell inoculation.

The effects of XZP treatment on spontaneous nociceptive behavior were determined using mechanical threshold and paw withdrawal latency (PWL), as previously described (11,16,27,28,30,32). In brief, individual rats were placed in an inverted plastic chamber on the glass surface of a Paw Thermal Stimulator System (UCSD, San Diego, CA, USA) for 30 min before the test. Mechanical hyperalgesia was measured using a single rigid filament attached to a handheld transducer (automatic plantar analgesia tester; Institute of Biomedical Engineering, Chinese Academy of Medical Science, Tianjin, China). Animals were acclimated to their surroundings daily for 10 min for three consecutive days in a plexiglass box on a metal grid surface prior to testing. On testing days, rats were allowed to acclimate for 5–10 min. A rigid filament was pressed perpendicularly against the medial plantar surface of the hind paw with increasing force. Brisk paw withdrawal or paw flinching accompanied by head turning, biting and/or licking upon application of the increasing force was considered as a positive response. Paw withdrawal threshold (PWT) was defined as the minimal force (g) required in evoking the cited positive responses. Each hind paw was tested three times and data were averaged. Interval between consecutive tests of the same paw was 5 min. The same procedure was performed on day 3, 6, 9, 12, 15, 18 and 21 after cell inoculation. Each hind paw was stimulated with a focused beam of radiant heat using an analgesiometer (37360; Ugo Basile, Italy) underneath the glass surface. When the paw was withdrawn from the stimulus, PWL was automatically recorded to the nearest 0.1 sec. Stimuli intensity was adjusted to generate an average baseline PWL of ~10.0 sec in naive animals. Maximum stimulation was controlled to <20 sec to prevent potential tissue damage. Paws were alternated randomly to preclude 'order' effects. Individual rats were subjected to four tests with 5-min intervals before surgery; and 2, 5, 8, 11, 14, 17 and 20 days after cell inoculation. Mean PWL values were calculated for each time point for individual rats.

Rats were deeply anesthetized by pentobarbital and transcardially perfused with saline on day 0 and 21 after inoculation (n=8/group). The left tibia of each rat was dissected, fixed in 10% formalin overnight, decalcified in 15% EDTA-phosphate-buffered saline (PBS) for seven days and paraffin-embedded. Tissue sections (6-µm) were prepared and stained with hematoxylin and eosin (H&E) for pathological analysis.

Orbital venous blood samples from each rat were collected on day 0, 7, 14 and 21 after cell implantation. Inflammatory mediator levels were determined using ELISA kits (Beijing Jia Mei Nuo Si Biotechnology Co., Ltd.; trypsin, ng/ml, catalog no. ABIN1117631, detection range 0.16–10 ng/ml, detection minimum 0.09 ng/ml; interleukin-1β, pg/ml, catalog no. ABIN365189, detection range 62.5–4,000 pg/ml, detection minimum 62.5 pg/ml; tumor necrosis factor-α, pg/ml, catalog no. ABIN365380, detection range 6.25–400 pg/ml, detection minimum 6.25 pg/ml).

To analyze c-Fos, GFAP, IBA1 and CGRP mRNA levels, total RNA was isolated from the L4-L5 spinal cord using the TRizol reagent (Invitrogen, Beijing, Mao Jian united Stars Technology Co., Ltd., Beijing, China) according to the manufacturer's instructions. Total RNA (5 µg) was reverse transcribed to cDNA with a first-strand cDNA synthesis kit (Invitrogen). Quantitative PCR was performed using a LightCycler system (Roche, Beijing Mao Jian United Stars Technology Co., Ltd.) with SYBR-Green. Each PCR was performed in triplicate to a final solution volume of 20-µl: 10 µl of SYBR-Green dye, 1 µl of diluted cDNA products,0.2 µM of each paired primer and 8.6 µl of deionized water. Protocols were as follows: initial denaturation for 5 min at 95°C, followed by 35 cycles denaturation for 15 sec at 95°C, and extension for 30 sec at 56°C. Last cycle for dissociation of SYBR-Green probe was 15 sec at 95°C, 30 sec at 56°C, and 15 sec at 95°C. Primer sequences are shown in Table I. Relative mRNA levels of c-Fos, GFAP, IBA1 and CGRP were normalized to GAPDH.

To analyze PAR2, PKC-γ, PKA and TRPV1 mRNA levels from rat bones with tumor invasion, total RNA was isolated using a TRIzol reagent and RT-PCRs were performed as above. Primer sequences are shown in Table I.

To analyze c-Fos, GFAP, IBA1 and CGRP protein levels, total proteins from the L4-L5 spinal cords in various groups were extracted by sonication in RIPA buffer containing protease inhibitors (Roche). Extracted proteins were dialyzed in PBS. After measuring protein content with a BCA protein assay kit (Thermo Scientific, Rockford, IL, USA), protein samples with equal amounts of total protein were separated on SDS-PAGE gels and transferred to nitrocellulose (NC) membranes (Millipore). Western blot assay was performed, as previously reported (16,27,28,31). Primary antibodies used in the present study were obtained from Abcam (Cambridge, MA, USA) including anti-c-Fos (phospho T325) (ab27793), anti-GFAP (ab4674), anti-IBA1 (ab107159) and anti-CGRP [EPR9670(B)] (ab139264) antibodies. Target protein bands were detected using horseradish peroxidase-conjugated secondary antibodies and visualized by an enhanced chemiluminescence detection system (Beijing Mao Jian United Stars Technology Co., Ltd.). Relative protein expression levels were quantified by ImageQuant™ LAS 4000 (GE Healthcare) and normalized to GAPDH.

To analyze PAR2, PKC-γ, PKA and TRPV1 protein levels in rat bones with tumor invasion, total proteins were extracted, quantified and analyzed by western blotting as previously described. Primary antibodies used in experiments were also obtained from Abcam including anti-PKC γ (phospho T514) (ab109539), anti-PAR2 (ab180953), anti-PKA α + β (catalytic subunits) (phospho T197) (ab5815) and anti-VR1 (ab63083) antibodies.

Behavioral data are expressed as mean ± standard deviation (SD) and analyzed with repeated measures ANOVA, while ELISA data were analyzed by one-way ANOVA followed by Newman-Keuls test. SPSS 13.0 statistical software was used in the present study. P<0.05 was considered to indicate a statistically significant result.

We first determined the effects of XZP treatments on bone cancer-related nociceptive behavior in bone cancer bearing Wister rats. While constant levels of mechanical thresholds were observed in sham control rats, levels of mechanical thresholds were gradually reduced throughout the observation period in rats in the placebo group (Fig. 1A). In contrast, treatment with different doses of XZP significantly mitigated mechanical allodynia in a dose- and time-dependent manner; although mechanical threshold values were lower than OPG-treated positive controls. Similar PWL patterns were observed in different groups of rats (Fig. 1B). These data indicated that XZP treatment reduced mechanical and thermal nociceptive behavior in rats with bone cancer.

Next, we examined the effects of XZP treatment on tibia cancer-induced bone structural damage. X-ray images indicated less severe loss of medullary bone and cortical bone erosion were observed in positive control (OPG) rats and XZP-treated rats at various doses at day 21 of post cell inoculation, while full-thickness bicortical bone loss was accompanied by displaced fractures in rats with bone cancer in the placebo group (Fig. 2A). Quantitative analysis revealed that radiological scores in rats that received different doses of XZP were significantly less than the placebo group, yet remained higher than OPG-treated positive controls (Fig. 2B). Quantitative analyses of BMC and BMD revealed that high-dose XZP treatment, similar to OPG, significantly preserved BMC at day 7, 14 and 21 post cell inoculation (P<0.05, P<0.01, Fig. 2C). Furthermore, medium- or high-dose XZP treatment, similar to OPG treatment, significantly mitigated BMD loss at day 7, 14 and 21 post cell inoculation (Fig. 2D). Histological examinations revealed that very few cancer cells invaded into bone tissues in positive controls and in rats treated with different doses of XZP, while a significant number of cancer cells invaded bone tissues and destroyed bone structure in the placebo group (Fig. 2E). Collectively, XZP treatment mitigated cancer invasion-mediated bone damage and structural changes.

To determine potential mechanisms underlying the action of XZP, we analyzed microglial IBA1 and astrocyte GFAP markers, and neurotransmitters c-Fos and CGRP, in the spinal cord of rats (Fig. 3). Significantly increased levels of c-Fos, GFAP, IBA1 and CGRP mRNA transcripts were detected in rats with bone cancer, compared with the sham controls (Fig. 3A); while relative levels of their gene mRNA transcripts were significantly reduced in rats treated with XZP, similar to rats that received OPG treatment. A similar pattern was detected when protein expressions in the spinal cord were analyzed (Fig. 3B and C). Thus, XZP treatment inhibited the activation of astrocytes and microglial cells, contributing to its therapeutic effects in rats with bone cancer.

ELISA analysis indicated that upstream levels of the PAR2 signaling pathway including trypsin, TNF-α and IL-1β serum in rats treated with XZP were significantly reduced at day 7, 14 or 21 post-inoculation, compared with placebo controls (Fig. 4); and these effects appeared to be dose-dependent. Furthermore, relative downstream levels of PAR2 signaling pathway mediators including PAR2, PKC-γ, PKA and TRPV1 mRNA transcripts in bone lesions were significantly increased in the rats with bone cancer, compared with the sham controls (Fig. 5A); while relative levels of these gene mRNA transcripts were significantly reduced in the rats that received XZP treatment, similar to rats with OPG treatment. A similar pattern was detected when protein expression was analyzed in bone lesions in rats in the different groups (Fig. 5B and C). Collectively, these data indicate that XZP treatment inhibited the PAR2 signaling pathway in rats with bone cancer, which contributes to the observed antinociceptive effects.