Background
Pain is the first clinical symptom of cancer in a large population of cancer patients, particularly in advanced or terminal cancer patients [
1], which strongly impacts the patients' quality of life. Tumor-derived, inflammatory, and neuropathic factors may simultaneously contribute to cancer pain such as bone cancer pain [
2].
Lysophosphatidic acid (LPA) is a lipid metabolite released after tissue injury, which induces diverse cellular responses including proliferation, adhesion, migration, morphogenesis, differentiation and survival [
3]. Increasing evidence shows that LPA is a key mediator in cancer development including cancer cell proliferation, survival and migration [
4‐
7]. There is a high concentration of LPA in ascitic fluid and plasma of cancer patients. Released by activated blood platelets [
8], LPA promotes progression of bone metastases by inducing secretion of tumor-derived cytokine (IL-6 and IL-8) in breast and ovarian cancer cells [
9,
10]. Additionally, lines of study have revealed that LPA may also be a crucial factor in the initiation of neuropathic pain mediated by demyelination of peripheral nerves via activation of LPA receptor [
11,
12]. Six subtypes of LPA receptor, LPA
1-6, are all G protein-coupled receptors. Three endothelial differentiation gene (EDG) family of G-protein-coupled receptors, EDG-2, EDG-4 and EDG-7 were identified as LPA receptors successively, and were named LPA
1-3 respectively. Then p2y9 or GPR23, GPR92 and GPR87 were also identified as LPA receptors, named as LPA
4-6 respectively [
3,
13,
14]. Among the six subtypes, LPA
1 receptor is the main subtype expressed in dorsal root ganglion (DRG) [
11]. LPA
1 is capable of interacting with three major G protein families, the G
i, G
q, and G
12 family, resulting in the activation of their downstream cascades: mitogen-activated protein kinase (MAPK), protein kinase C (PKC) and Rho (a small GTP-binding protein)-Rho kinase, while inhibiting protein kinase A (PKA) pathway [
3]. Several studies have demonstrated that LPA
1 participates in the development of neuropathic pain through the Rho pathway [
11,
15,
16].
In patients with advanced cancers such as breast, lung, prostate or myeloma cancer, bone pain is the most frequent cancer-induced chronic pain. However, the mechanisms underlying the development of bone pain are not completely understood. Bone cancer induces mechanical bone deformation and local tissue acidosis which may activate nociceptors via multiple molecular mechanisms, particularly by activation of the capsaicin receptor, transient receptor potential vanilloid (TRPV1). Compelling evidence has testified that TRPV1 is a critical signal molecule in the development of physiological and pathological pain [
17]. It has been shown that TRPV1 is involved in the induction of bone cancer pain [
18‐
22] and is activated by direct phosphorylation via PKC pathway [
23‐
26], particularly via PKCε [
27].
In the last decade, bone cancer pain models have been successfully established and used to explore associated mechanisms [
28‐
30]. Given expression of LPA
1 and TRPV1 in the DRG neurons, the present study focused on whether LPA
1 is involved in bone cancer pain via cross-talking with TRPV1 and the possible signal pathways in the peripheral mechanism underlying bone cancer pain.
Discussion
The transient receptor potential vanilloid subtype 1 (TPRV1) is predominantly expressed in C-fiber nociceptors and responds to capsaicin, noxious heat, protons, and various endogenous ligands [
34,
35]. Recent studies have revealed that TRPV1 is involved in the development of cancer-induced pain [
18‐
22]. The present study provided further evidence that capsaicin-induced currents and expression of TRPV1 in DRG neurons were up-regulated in bone cancer rats. A line of studies have testified that capsaicin-induced TRPV1 currents in DRG nociceptive neurons are enhanced by different inflammatory mediators such as bradykinin, chemokine and substance P etc. [
27,
36‐
39]. LPA was released from diverse sources such as blood platelets, cancer cells and the surrounding tissues. This inflammatory mediator contributed to the initiation of neuropathic pain associated with demyelination in the dorsal root [
11]. Our recent studies exhibited that following the development of bone cancer by injection of cancer cells into the tibia, the posterior articular nerve directly innervating the region of bone cancer was severely degraded and the activating transcription factor 3 (ATF3, a marker of nerve injury and stress) expression was up-regulated in DRG neurons at spinal L4-5 segments [
40]. Moreover, there was almost no spontaneous firing in the ipsilateral posterior articular nerve of bone cancer (unpublished). Noticeably, the sural nerve innervating the adjacent region of the bone cancer was intensely sensitized [
40]. Therefore, capsaicin-induced responses in this study were more likely recorded from uninjured, but not injured, DRG neurons. Considerable evidence has shown that the injured nerve fibers could increase excitability of adjacent uninjured fibers which contributes to the induction of neuropathic pain [
41‐
43]. Given that bone cancer pain characterizes neuropathic pain, it is conceivable that LPA-induced potentiation of TRPV1 current reflects an increase in excitability of uninjured fibers innervating the adjacent region of osteocarcinoma. The facilitation of adjacent uninjured fibers may play a crucial role in the induction of bone cancer pain.
Amongst six subtypes LPA
1-6, LPA
1 receptors is the main subtype expressed in DRG neurons and activated under the neuropathic pain state [
11,
15,
44]. Consistent with alleviation of neuropathic pain [
11], VPC32183, a LPA
1 and LPA
3 receptors antagonist, blocked LPA-induced potentiation of TRPV1 currents in DRG neurons, mechanical allodynia and thermal hyperalgesia in bone cancer rats, suggesting that activation of LPA
1 receptors mediates bone cancer-induced sensitization of DRG nociceptive neurons and pain hypersensitization. A study reported that LPA
3 receptors were expressed in some DRG neurons and knockout of LPA
3 could relieve pain [
45]. Therefore, in addition to LPA
1, LPA
3 may also contribute to bone cancer pain. The finding that LPA increases intracellular calcium concentration in DRG neurons supports the idea that LPA directly activates DRG neurons via LPA receptors by cross-talking with TRPV1 [
46]. The findings that co-localization of TRPV1 and LPA
1 receptor EDG-2 and bone cancer induced an increase in LPA
1 receptor expression in DRG [
40] provide further support for these phenomena.
Compelling evidence demonstrates that PKC regulates sensitivity of nociceptors through phosphorylation of many protein substrates [
35,
37]. Potentiation of TRPV1 currents in DRG neurons by different inflammatory mediators are mediated, at least in part, by PKC [
47‐
51]. It has been known that five isoforms of PKC are expressed in DRG neurons, but only PKCε isoform is translocated to the cell membrane after bradykinin stimulation [
52]. Our previous studies showed co-expression of TRPV1 and PKCε and inflammation-induced up-regulation of PKCε in DRG nociceptive neurons [
53], and potentiation of TRPV1 currents by activation of Neurokinin 1 (NK-1) receptor via PKCε [
27]. As with other G-protein-coupled receptors (GPCR) such as NK-1, LPA
1 receptor activation-induced potentiation of TRPV1 in DRG neurons was also blocked by PKC inhibitor or PKCε inhibitor. In addition to PKC, previous studies demonstrated that PKA inhibitor blocked the potentiation of TRPV1 currents by pro-inflammatory factors such as: NGF, prostaglandins, anandamide, 5-HT, and glutamate [
39,
46,
54‐
58]. However, the present study revealed that LPA-induced potentiation of TRPV1 failed to be blocked by the PKA inhibitor. Taken together, the interaction of LPA
1 and TRPV1 is mediated by PKC, particularly by PKCε, but not by PKA. The identical process occurs in modulation of TRPV1 by activation of NK-1 in DRG neurons [
27]. Therefore, it seems that PKCε, but not PKA, phosphorylation may be a common mechanism in which activation of LPA
1 and NK-1 potentiates TRPV1 in DRG nociceptive neurons contributing to bone cancer pain and neuropathic pain. Interestingly, LPA can increase tetrodotoxin-resistant (TTX-R) sodium current [
59] and induce substance P release from nociceptor endings [
60]. Our recent finding showed that activation of NK-1 receptor potentiated TTX-R sodium channel Nav1.8 currents, mediated by PKCε in DRG neurons [
61]. Collectively, two most important pain-related signal molecules, TRPV1 and Nav1.8, in nociceptors can be modulated by activation of LPA
1 mainly via PKCε phosphorylation.
LPA
1 receptors are reported to interact with members of three major G protein families, G
i, G
q, and G
12 family, to regulate the activity of intracellular messenger molecules [
3,
62‐
64]. LPA
1 couples through G
12/13 to Rho activation, and inhibition of Rho-associated kinase, a downstream effector of LPA
1, abolished LPA-induced action in osteoblasts [
65]. An elegant study from Inoue
et al., (2004) reported that Rho-Rho kinase signal pathway is implicated in the induction of neuropathic pain by activation of LPA
1 receptor in DRG neurons. However, the present study showed that inhibition of Rho failed to block LPA-induced potentiation of TRPV1 currents in whole-cell recording of DRG neurons. The different results between the behavioural test and single cell recording suggest that LPA might be associated with multiple intracellular signals to modulate different targets in DRG neurons. LPA-induced potentiation of TRPV1 currents in DRG neurons is mediated by PKCε rather than Rho pathway. In addition to TRPV1, LPA could modulate other targets in the induction of bone cancer pain, which might be mediated by Rho pathway.
Methods
Animals
Animals used for bone cancer model and all experiments were female Sprague-Dawley (SD) rats weighting 160-200 g. Animals for subculture were female SD rats weighting 80 g. All of them were purchased from the Fudan experimental animal center. All experiments conformed to local and international guidelines on ethical use of animals and all efforts were made to minimize the number of animals used and their sufferings.
Cancer cell inoculation surgery
Walker 256 rat mammary gland carcinoma cells were provided by Fudan University, School of medicine, department of integrative medicine and neurobiology, and were injected into the abdominal cavity of rats weighting 80 g to obtain ascitic fluid for cancer cell culture, continuously cultured from generation to generation weekly for the establishment of a bone cancer model.
Animals were anesthetized with Chloral hydrate (intra-peritoneal (i.p.) 400 mg/kg). The left tibia was chosen as the operated side. 10
7 carcinoma cells in 4 μl phosphate buffered saline (PBS) or 4 μl PBS alone (shame group) were injected into the tibia cavity through the knee joint. The inject site was closed by 1 μl absorbable gelatin sponge. Each animal was injected with 100,000 units of penicillin and had a two days recovery before any experiment. The detailed procedure for cancer cell culture and inoculation has been presented in previous reports [
28].
Intrathecal Injection
Rats received intrathecal injection of drugs by lumber puncture as described previously [
66]. Briefly, rats were anaesthetized with isofluorane in a transparent plastic box and then placed on a roller so that the L
4-6 vertebrae were curved. A lumbar puncture needle was introduced into the intrathecal space. The needle had been introduced intrathecally when a short flicking of the tail was observed. Then drugs were slowly injected into the intrathecal space. The needle was immediately pulled out after the injection.
Mechanical allodynia
After 30 min acclimation in an individual testing cage, the rat's paw withdrawal threshold (PWT) was measured as the hind paw withdrawal response to von Frey hair (Stoelting, IL, USA) stimulation. In detail, an ascending series of von Frey hairs stimuli (2.0, 4.0, 6.0, 8.0, 15.0 and 26.0 g) was applied for 3 s where positive response was defined as a withdrawal of hind paw upon the stimulus, each stimulus was repeated for 5 times with a 10 min interval, and the lowest force to induce at least 3 positive responses was defined as PWT. The tester was blinded to the condition of the animals.
Thermal hyperalgesia
As previously described [
67], thermal hyperalgesia was assessed by measuring rat's paw withdrawal latency (PWL) to a radiant heat (model 336 combination unit, IITC/life Science Instruments, Woodland Hill, CA, USA). Rats were placed individually in plastic cages on an elevated glass platform and allowed for 30 min acclimation. Each hind paw received three stimuli with a 10 min interval, and the mean of the three withdrawal latencies was defined as PWL. The heat was maintained at a constant intensity. To prevent tissue damage, the cut-off latency was set at 20 s. The tester was blinded to the condition of the animals.
Western-blotting
Animals were anesthetized with Chloral hydrate (i.p. 400 mg/kg). L4-6 DRGs were collected from sham and cancer animals at the 14th day after inoculation surgery, homogenized in a lysis buffer containing protease inhibitor (Sigma, St. Louis, MO, USA). The protein concentrations of each group were measured using a BCA assay (Pierce Biotechnology Inc., Rockford, IL, USA). The extracts were separated using SDS-PAGEs (10%) with 3 mg protein in each lane and transferred to PVDF membranes. The membranes were blocked in 10% non-fat dry milk for 2 h at room temperature (RT). After having been incubated overnight with goat anti-TRPV1 primary antibody (R-18, C-terminal, 1:1,000, Santa Cruz Biotechnology Inc., CA, USA) or mouse anti-Tubulin primary antibody (1:5,000, Sigma), the membranes were incubated with horseradish peroxidase (HRP)-conjugated donkey anti-goat or goat anti-mouse secondary antibody (1:3000, Santa Cruz Biotechnology) for 2 h at RT. Secondary antibody reactive bands were visualized in ECL solution (Pierce) for 1 min and exposed onto X-films for 1-5 min.
Immunohistochemistry
Sensory neurons in L4-6 DRGs of normal rats were labelled by antibodies against TRPV1 and EDG-2 (endothelial differentiation gene-2, LPA1 receptor). Rats were given an overdose of urethane (i.p. 1.5 g/kg) and perfused intracardially with saline, followed by perfusion of 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). L4-6 DRGs were then removed, post-fixed in the same fixative for 4 h at 4°c, and immersed from 10% to 30% gradient sucrose in PB for 24-48 h at 4°c for cryoprotection. Frozen 14 μm DRG sections were cut using a cryostat microtome and thaw-mounted onto gelatin-coated slides for processing. The sections were blocked with 10% donkey serum in 0.01 M PBS (pH 7.4) with 0.3% Triton X-100 for 2 h at RT and incubated for 48 h at 4°c with goat anti-TRPV1 (1:200, Santa Cruz Biotechnology) and rabbit anti-EDG-2 (1:50, Novus Biologicals Inc., CO, USA) primary antibodies in PBS with 1% normal donkey serum and 0.3% Triton X-100. Following three 10 min rinses in 0.01 M PBS, the sections were incubated in rhodamine red-X (RRX)-conjugated donkey anti-goat IgG (1:200, Jackson Immunolab Inc., West Grove, PA, US) and fluorescein isothiocyanate (FITC)-conjugated donkey anti-rabbit IgG (1:200, Jackson Immunolab) for 90 min at 4°c, and then washed in PBS. All sections were coverslipped with a mixture of 50% glycerin in 0.01 M PBS, and then observed under a Leica fluorescence microscope, and images were captured with a CCD spot camera.
DRG neurons preparation
Acutely dissociated DRG neurons were prepared from 180 g rats as described previously [
53]. L
4-6 DRGs were removed and incubated in Dulbecco's Modified Eagle Media (DMEM), which was added with 3 mg/ml collagenase (type IA, Sigma) and 1 mg/ml trypsin (type I, Sigma), for 25 min at 36.8°C. After enzyme treatment, ganglia were rinsed three times with standard external solution, and then single cells were dissociated by trituration using fine fire-polished Pasteur pipettes. DRG neurons were placed onto coverslips (10 mm diameter) in the 3.5 cm culture dishes and incubated in Standard external solution for recordings at RT.
Electrophysiological recordings
Whole-cell voltage-clamp recordings were made at RT (20-22°C) with an EPC-9 amplifier (HEKA Elektronik, Lambrecht/Pfalz, Germany). All recordings were performed within 2-8 h after plating. Neurons were prepared as above. Only small-diameter (15-25 μm) DRG neurons were recorded in all of the experiments. Microelectrodes were pulled by a P97 puller (Sutter Instruments, Novato, CA, USA) and those with the resistance of 2-6 MΩ were used. Standard external solution contained (in mM): 150 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 HEPES, and 10 glucose, pH 7.4. The pipette solution contained (in mM): 140 KCl, 5 NaCl, 1 MgCl2, 0.5 CaCl2, 5 EGTA, 3 Na2-ATP, and 10 HEPES, pH 7.2. The whole cell capacitance was cancelled and series resistance was compensated (> 80%) after gigaohm seal formation and membrane disruption, and data were sampled at 5 kHz and low-passed at 1 kHz. Data acquisition was controlled by the software Pulse and Pulsefit 8.5 (HEKA Elektronik).
Drugs
All the drugs for patch-clamp recording and intrathecal injection were purchased from Sigma, except that the LPA1 inhibitor VPC32183 was from Avanti Polar Lipids and PKCϵ inhibitor ϵV1-2 was from Biomol (Plymouth Meeting, PA). All the drugs were stocked at -20°C at a concentration at least 1000-fold the working concentration and prepared on the day of the experiment. LPA and capsaicin were applied close to the neurons by an ALA-VM8 perfusion system (ALA Scientific Instruments, Westbury, NY, USA). Capsaicin was applied once every minute and LPA was applied for 1 min. After cells were washed by standard external solution for 10 seconds, capsaicin was applied for 3 seconds, followed by standard external solution washing for 47 seconds. The control current was averaged from three continuous recordings. In the fourth minute, standard external solution was changed by LPA. Inhibitors were applied (where appropriate) to the chamber 30 min before LPA perfusion, except that BoTXC3 and ϵV1-2 were delivered intracellularly via a recording electrode. Inhibitors existed during the whole recording course.
Statistical analysis
The western bolt and electrophysiology data were analyzed using student's t test. Two Way Repeated Measures ANOVA were used to testing differences of PWT and PWL values between groups. The criterion of significance was set at *, # p < 0.05, **, ## p < 0.01, ***, ### p < 0.001, all results are expressed as means ± standard error of the mean (SEM).
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
HLP performed all of the experiments. YQZ was partially involved in experimental design and guiding. ZQZ is the corresponding author. All authors read and approved the final manuscript.