PAR2-mediated upregulation of BDNF contributes to central sensitization in bone cancer pain

Implantation of tumor cells into the tibias significantly decreased the ipsilateral paw withdrawal threshold responsive to the mechanical stimuli (Figure  1C), which was remarkably detectable at day 5, and maintained through at least day 15 (the endpoint of the study) after the procedure. Meanwhile, no significant change of the ipsilateral paw withdrawal threshold was observed in the rats receiving boiled cells. Similarly, the ipsilateral paw withdrawal latencies responsive to the radiant thermal stimuli were also significantly shortened in the rats implanted with tumor cells, which started at day 5 through day 15 (Figure  1C). These results indicated the bone cancer-induced mechanical allodynia and thermal hyperalgesia in the modeled rats.

To determine the adaptation of glutamatergic transmission in the spinal dorsal horn of the rats with bone cancer pain, the whole cell-recording was performed in the ipsilateral laminae II neurons in spinal slices (L4-L5), and the glutamatergic EPSC was recorded in the presence of strychnine (2 μM) and bicuculline (10 μM). The EPSC was completely abolished by the perfusion of CNQX (10 μM) and APV (10 μM), confirm its glutamatergic components. Compared to that in the control rats, the input (intensity of stimuli) –output (amplitude of evoked EPSC) response was significantly left shifted (Figure  1B), indicating a significantly enhanced glutamatergic strength in the dorsal horn neurons in the rats with bone cancer pain.

BDNF, synthesized from central neurons and astrocytes, is critically involved in the glutamatergic synaptogenesis and synaptic plasticity in the brain [8]. Here, we investigated the expression of BDNF in the dorsal horn, and its functional significance in the bone cancer-induced pain. As shown in Figure  1A, the expression of BDNF was significantly increased in the dorsal horn of the rats with bone cancer pain. The further electrophysiological recording studies revealed that intrathecal delivery of TrkB-Fc chimera (1.5 μg, day 3 to day 15), which quenching the endogenous BDNF, significantly ameliorated the increase of glutamatergic strength in the dorsal horn neurons of the rats with cancer-induced pain (Figure  1B). Behavioral studies demonstrated that inhibition of BDNF signaling by TrkB-Fc chimera also significantly attenuated the mechanical allodynia and thermal hyperalgesia in these modeled rats (Figure  1C). Note that blockade of the BDNF signaling failed to remarkably change basal glutamatergic transmission and pain behavior in the control rats (Figure  1B and C). These results suggested the critical involvement of BDNF signaling in the enhanced glutamatergic transmission in the dorsal horn neurons of the rats with bone cancer pain.

Considering the important role of PAR2 in neuroinflammation and the pathogenesis of several kinds of pain [12], we next determine the involvement of spinal PAR2 signaling in the induction of the enhancement of spinal glutamatergic transmission and pain behavior in the rats implanted with tumor cells. Firstly, as shown in Figure  2A, the expression of PAR2 was significantly increased in the dorsal horn of the rats with bone cancer. This indicated the potential role of PAR2 in the pathogenesis of bone cancer-induced pain.

Then, FSLLRY-NH2 (10 μg, from day3 to day 15), the antagonist of PAR2, was daily delivered via the intrathecal catheter to determine its effect on BDNF function, glutamatergic strength and pain behavior in the rats with bone cancer. As shown in Figure  2B, blockade of PAR2 signaling by FSLLRY-NH2 significantly reversed the upregulation of spinal BDNF in the rats implanted with tumor cells, while it did not obviously change the expression of spinal BDNF in the control rats. This suggested a PAR2-mediated spinal BDNF upregulation in the setting of bone cancer pain.

Further recording results demonstrated that, as shown in Figure  2C, intrathecal delivery of FSLLRY-NH2 significantly attenuated the enhanced glutamatergic input–output response in the dorsal horn neurons of the rats with bone cancer, while it did not remarkably change the basal glutamatergic strength in the control rats (Figure  2C). It was also found that intrathecal delivery of FSLLRY-NH2 attenuated the increased expression of glutamate receptor subunits GluR1 and NR2B in the dorsal horn of the rats with bone cancer (Figure  2D). As well, behavioral studies also demonstrated that inhibition of PAR2 signaling by FSLLRY-NH2 significantly attenuated the mechanical allodynia and thermal hyperalgesia in these modeled rats (Figure  2E). Notably, intrathecal injection of scramble peptides in the same dose failed to affect the mechanical and thermal pain behaviors in either control or modeled rats (Figure  2F). These results suggested that upregulation of PAR2 signaling, via modulating the expression of BDNF, critically participated in the induction of enhanced spinal glutamatergic transmission and pain behavior in the rats with bone cancer.

Activation of PAR2 signaling may induce the degradation of NF-κB inhibitor protein IkB, and lead to the activation and nuclear translocation of transcription factor NF-κB [18], which is critically involved in the synaptic plasticity [16] and processing of nociceptive information [19]. Here, as shown in Figure  3A, although the total expression level of NF-κB subunit p65 did not have significant change, the phosphorylated p65 was significantly increased in the ipsilateral dorsal horn of the rats implanted with tumor cells. Then, pyrrolidine dithiocarbamate (PDTC, 1 μg/10 μl from day 3 to day 15), an inhibitor of NF-κB, was daily delivered via intrathecal catheter to suppress the PAR2-NF-κB signaling in the rats with bone cancer pain. As shown in Figure  3B, blockade of NF-κB signaling by PDTC significantly reversed the upregulation of spinal BDNF in the rats with bone cancer-induced pain, while it did not obvious change the expression of spinal BDNF in the control rats. Further electrophysiological studies revealed that, as shown in Figure  3C, intrathecal delivery of PDTC significantly attenuated the enhanced glutamatergic strength in the dorsal horn of the rats implanted with tumor cells, while it did not remarkably change the basal glutamatergic strength in the control rats. Meanwhile, behavioral studies also demonstrated that inhibition of NF-κB signaling by PDTC significantly attenuated the mechanical allodynia and thermal hyperalgesia in these modeled rats (Figure  3D). These results suggested that activation of NF-κB signaling contributed to the PAR2-mediated BDNF upregulation, enhanced spinal glutamatergic transmission and pain behavior in the rats with bone cancer.

Adult female Wistar rats (weighing 200–250 g) were purchased from the Institutional Center of Experiment Animals, and were housed in the standard lab conditions (22 ± 1°C and 12:12 h light cycle) with unrestricted free access to food and water. All animal experiments were approved by the Institutional Animal Care and Use Committee in China Academy of Chinese Medical Sciences, and were in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All surgery was performed under appropriate anesthesia, and all efforts were made to minimize suffering in the animals.

The rat model of bone cancer pain was established by implantation of Walker 256 rat mammary gland carcinoma cells into the right tibias, as previously reported [3]. Briefly, suspensions of tumor cells in PBS was prepared as previously described [39]. Tumor cell implantation was achieved by injecting the tumor cells (1 × 10⁵ cells/μl, 5 μL) into the intramedullary space of the right tibia to induce bone cancer in rats under the anesthesia with pentobarbital (45 mg/kg). Boiled cells were injected with a similar procedure in the control group. The investigators who performed behavioral tests, immunoblotting, and electrophysiological recording were blinded to this procedure.

Intrathecal cannulation was performed as described previously [40]. Intrathecal catheters (PE-10 polyethylene tubing) were implanted 1 week before any procedures. The catheters were advanced 8 cm caudally through an incision in the cisternal membrane and secured to the musculature at the incision site. Only animals with no evidence of neurological deficits after catheter insertion were studied. TrkB-Fc chimera (1.5 μg, R&D systems cat#: 688-TK-100) [41], FSLLRY-NH2 (10 μg, Tocris cat#: 4751) [42], pyrrolidine dithiocarbamate (PDTC, 1 μg, Sigma cat#: P8765) [43], or vehicle was administered in a volume of 10 μl followed by a 10-μl flush with normal saline. The drugs were given daily from day 3 to day 15 after carcinoma cell inoculation.

Mechanical allodynia and thermal hyperalgesia were assessed in rats to evaluate painful behavior induced by tumor cell implantation. Mechanical sensitivity was assessed by using a series of von Frey filaments with logarithmic incremental stiffness (Stoelting Co., Wood Dale, IL), and 50% probability paw withdrawal thresholds were calculated with the up-down method as previously described [44]. In brief, the animals were placed individually beneath an inverted ventilated cage with a metal-mesh floor, and allowed to habituation for 30 minutes before testing. The filaments were applied to the plantar surface of the hindpaws for approximately 6 seconds in an ascending or descending order after a negative or positive withdrawal response, respectively. Six consecutive responses after the first change in response were used to calculate the paw withdrawal threshold (in grams). The maximum pressure was set as 16 g.

Thermal nociceptive thresholds were measured using radiant heat [45]. Briefly, rats were placed into individual plastic cubicles mounted on a glass surface maintained at 30°C and allowed an hour habituation period. A thermal stimulus (a radiant heat source) was then delivered to the plantar surface of the hindpaws. The time for the rat to remove the paw from the thermal stimulus was electronically recorded as the paw withdrawal latency (PWL). The stimulus shut off automatically when the hind paw moved or after 16 seconds to prevent tissue damage. The intensity of the heat stimulus was maintained constant throughout all experiments.

After behavioral testing, the rats were deeply anesthetized with pentobarbital sodium (50 mg/kg), and the ipsilateral L4-L5 segments were quickly removed and homogenized in the lysis buffer (25 mM Tris–HCl, pH 7.6, 150 mM NaCl, 0.1% SDS, 1 mM PMSF, 1 mM NaF, 1 mM NaVO₃, 1 μg/ml leupeptin, 1 μg/ml pepstatin, and 1 μg/ml aprotinin). The lysates were centrifuged at 14,000 rpm for 10 min at 4°C, and the protein concentrations in the supernatant were measured by using the Bio-Rad protein assay kit. The samples (containing 20 μg proteins) were separated on a 15% (for BDNF) or 7.5% (for other proteins) SDS-polyacrylamide gel, and blotted to a nitrocellulose membrane. The blots were incubated overnight at 4°C with a rabbit polyclonal anti-BDNF primary antibody (1:250; Santa Cruz Biotechnology cat#: sc-546), monoclonal anti-PAR2 antibody (1:1000; Millipore cat#: MABF243), polyclonal anti-NF-κB p65 (1:1000, Abcam cat#: ab7970), anti-phosphorylated NF-κB p65 (1:1000, Abcam cat#: ab106129), or monoclonal anti-β-actin antibody (1:2000; Santa Cruz Biotechnology, cat#: sc-81178). The membranes were washed extensively with Tris-buffered saline and incubated with horseradish peroxidase-conjugated anti-mouse and anti-rabbit IgG antibody (1:10,000; Jackson ImmunoResearch Laboratories Inc., West Grove, PA). The immunoreactivity was detected using enhanced chemiluminescence (ECL Advance Kit; Amersham Biosciences). The intensity of the bands was captured digitally and analyzed quantitatively with Image J software. The immunoreactivities of target proteins were normalized to that of β-actin.

The rats were deeply anesthetized with inhalation of halothane, and the lumbar segment of the spinal cord was removed through laminectomy. The spinal tissue was immediately placed in ice-cold artificial cerebrospinal fluid containing (in mM): sucrose 230, KCl 3.5, MgCl₂ 1.5, CaCl₂ 2.0, NaH₂PO₄ 1.2, glucose 12, and NaHCO₃ 25. Transverse spinal cord slices (400 μm) were cut with a vibratome (Technical Products International, St. Louis, MO), and incubated in Krebs’ solution (containing: 117 mM NaCl, 3.6 mM KCl, 1.2 mM MgCl₂, 2.5 mM CaCl₂, 1.2 mM NaH₂PO₄, 11 mM glucose, and 25 mM NaHCO₃, bubbled with 95% O₂ and 5% CO₂) at 35°C for at least 1 h before the recording was performed.

Whole-cell recording on the spinal cord slices (L4-L5) was performed as described previously [40]. The neurons in lamina II of the ipsilateral dorsal horn were visualized using an upright microscope with infrared illumination (BX50WI; Olympus, Tokyo). Whole-cell voltage-clamp recordings were performed using an Axopatch 200B amplifier (Molecular Devices) with 3–5 MΩ glass electrodes containing the following internal solution (in mM): K-gluconate, 126; NaCl, 5; MgCl₂ 1.2; EGTA, 0.5; Mg-ATP, 2; Na₃GTP, 0.1; HEPES, 10; guanosine 5-O-(2-thiodiphosphate) 1; lidocaine N-ethyl bromide (QX314), 10; pH 7.3; 290 – 300 mOsmol. A seal resistance of ≥ 2 GΩ and an access resistance of 15 – 20 MΩ were considered acceptable. The series resistance was optimally compensated by ≥70% and constantly monitored throughout the experiments. The membrane potential was held at -60 mV throughout the experiment. Excitatory postsynaptic currents (EPSCs) in ipsilateral lamina II neurons were evoked by electrical stimulation (0.25 ms, 0.05 – 0.3 mA) of the dorsal root in the presence of strychnine (2 μM) and bicuculline (10 μM). The evoked EPSCs were filtered at 2 kHz, digitized at 10 kHz, and acquired and analyzed using pCLAMP 9.2 software (Molecular Devices).

All data were presented as means ± SEM. For analysis of immunoblotting data, differences between groups were compared by Student’s t-test or ANOVA followed by Fisher’s PLSD post hoc analysis. The electrophysiological and behavioral data were compared with two-way ANOVA with repeated measurements. The criterion for statistical significance was P < 0.05. Statistical tests were performed with SPSS 13.0 (SPSS, USA).