Blockade of spinal dopamine D1/D2 receptor heteromers by levo-Corydalmine suppressed calcium signaling cascade in spinal neurons to alleviate bone cancer pain in rats

Background: Dopamine receptors have been reported to be involved in pain, while the exact effects and mechanism in bone cancer pain have not been fully explored. Methods: Bone cancer pain model was created by implanting walker 256 mammary gland carcinoma into right tibia bone cavity. Primary cultured spinal neurons were used for in vitro evaluation. FLIPR, western-blot, immunofluorescence, and Co-IP were used to detect cell signaling pathway. Results: Our results indicated that spinal dopamine D1 receptor (D1DR) and spinal dopamine D2 receptor (D2DR) could form heteromers in TCI rats, and antagonizing spinal D1DR and D2DR reduced heteromers formation and alleviated TCI-induced bone cancer pain. Further results indicated that D1DR or D2DR antagonist induced antinociception in TCI rats could be reversed by D1DR, D2DR, and D1/D2DR heteromer agonists. And Gq, IP3, and PLC inhibitors also attenuated TCI-induced bone cancer pain. In vitro results indicated that D1DR or D2DR antagonist decreased the Ca2+ oscillations upregulated by D1DR, D2DR, and D1/D2DR heteromer agonists in activated primary cultured spinal neurons. Moreover, inhibition of D1/D2DR heteromers induced antinociception in TCI rats was partially mediated by the CaMKII and MAPKs pathway. In addition, a natural compound levo-Corydalmine (l-CDL), could inhibit D1/D2DR heteromers and attenuate bone cancer pain. Results: Inhibition of spinal D1/D2DR heteromers via l-CDL decreases excitability in spinal neurons, which might present new therapeutic strategy for bone cancer pain.


Introduction
Common tumors, such as breast, lung and prostate cancer frequently metastasize to multiple bones in the body and induce significant bone pain.The mechanisms of bone cancer pain are highly complicated.Cancer cells metastasize to the bone where they release algogenic substances, protons, and acidosis acting on the receptors of peripheral nociceptors [1,2] and thereby inducing peripheral sensitization.The spinal cord receives the input from the primary afferent neurons.The activated spinal neurons can in turn release various excitatory neurotransmitters including glutamate, adenosine Ivyspring International Publisher triphosphate (ATP) and calcitonin gene-related peptide (CGRP) to act on their receptors on postsynaptic neurons, which can increase intracellar Ca 2+ levels [3,4].Upregulation of Ca 2+ facilitates signal transmission by activating Ca 2+ -sensitive proteins, such as calcium/ calmodulin-dependent protein kinase (CaMKII) and mitogen-activated protein kinase (MAPK) to further enhance excitability in spinal cord neurons.
The detailed role and mechanisms of D1DR or D2DR in chronic bone cancer pain largely remains unexplored.D1DR and D2DR were reported to be mainly expressed in neurons of the spinal cord [5,6].Activated spinal neurons plays an important role in the development and maintenance of chronic pain [7].Inhibiting the activation and excitability of spinal neurons could markedly attenuate bone cancer pain [8].It has been reported that D1DR couples to Gs/olf proteins to activate cyclic adenosine monophosphate (cAMP) and D2DR couples to Gi/o proteins to inhibit adenylcyclase (AC) [9].It has been widely accepted that activated cAMP signaling can phosphorylate cAMP response element-binding protein (CREB) to increase neuronal excitability in rodent hippocampal neurons [10] and striatal neurons [11], while inhibiting cAMP decreases neuronal excitability [12].The dopamine D1/D2DR heteromers was first identified in rat striatum [13], and was reported to be coupled to the Gq/11 protein, a finding that suggested a direct link between dopamine and calcium signaling [14][15][16][17][18]. Increased intracellular calcium has been implicated in the increased excitability of neurons [19] and the development of chronic pain.Our previous research indicated that spinal dopamine D1 and D2 receptor form heteromers in spinal cord in neuropathic pain [20].Herein, we further tested the hypothesis that D1DR and D2DR might form heteromers to induce the activation of spinal neurons thereby promoting the development of bone cancer pain.This study aimed to investigate the role and mechanisms of spinal D1DR, D2DR, and their heteromers in bone cancer pain.
Corydalis yanhusuo W.T. Wang is one of the most famous analgesic is China, and tetrahydroproberberines were the main active ingredients and has been shown to act as dopamine agonist or antagonist [21,22].l-CDL, one of the trace tetrahydroproberberines of Corydalis yanhusuo W.T. Wang has been reported to exert strong anti-nociception without notable side effects [23][24][25][26].Our previous researches showed that l-CDL could inhibit the formation of D1/D2DR complex to alleviate neuropathic pain [20], so the present study further investigated whether it could inhibit the D1/D2DR complex in bone cancer pain relief.

Ethics Statement
All experimental protocols were approved by the Animal Experimentation Ethics Committee of China Pharmaceutical University and adhered to the guidelines of the International Association for the Study of Pain (IASP).Meanwhile, the experiments we did were designed to minimize suffering and the number of animals used.

Experimental Animals
Sprague-Dawley rats weighing 180-220 g and 60-80 g were purchased from the Experimental Animal Center at Yangzhou University (Jiangsu Province, China, SCXK-SU-2016-0011).Rats were housed three per cage in a temperature and humiditycontrolled environment on a 12 h light/dark cycle for 3-7 days to allow acclimatization.Rats were anesthetized with pentobarbital (50 mg/kg, i.p.) and euthanized with carbon dioxide.Subsequently, the rats were randomly allocated to the following groups: ).Behavioral testing was performed during the light cycle (between 9:00 a.m. and 5:00 p.m.).Six animals were assigned to each group for behavioral test and four animals were assigned to each group for molecular testing.
Nonspecific oligonucleotide controls consisted in randomly scrambled sequences of siRNA groups (conRNA).33 μg siRNA and 49.5 μg polyethyleneimine (PEI) was diluted in 165 μL of 5 % glucose solutions respectively and were mixed and incubated for 15 min at RT before use [27,28].For the siRNA group, each rat received multiple daily intrathecal injections of D1DR and D2DR siRNA mixed solution (1 μg/20 μL) for 8 consecutive days, and the control RNA group receive conRNA (1 μg/20 μL) for 8 days.Antinociception was measured at 0.5 h after siRNA treatment for 1-7 day and at 0.5 h, 2 h, 4 h, and 8 h on the 8 th day after siRNA treatment.

Model of Bone Cancer Pain induced by Intra-Tibia Inoculation of Walker 256 Mammary Gland Carcinoma Cells
TCI-induced bone cancer pain was established according to our previous research [29].Walker 256 mammary gland carcinoma cells (5×10 6 cells/mL, 0.5 mL) were intraperitoneally injected into rats weighing 60-80 g.The ascites were extracted and centrifuged at 400 g for 6 min to get the cells 5-7 days later.The cells were washed with iced 0.01 M phosphate-buffered saline (PBS) and then diluted to a density of 1×10 5 cells/μL with 0.01 M PBS.Rats were anesthetized and the tibia head of the left leg was exposed with minimal damage.Next, 5 μL Walker 256 ascites tumor cells were slowly injected into the medullary cavity, and 5 μL PBS were injected as a control.To stop the cells from coming out, the syringe was held still for 1 min and bone wax was subsequently applied for 3 min.The injection hole was closed with dental materials.

Behavioral Assays for Bone Cancer Related Pain
Before the test, rats were placed in individual transparent plastic mesh cage to accommodate the environment, then a series of Von Frey hairs (1.4-15.0g) were used to stimulate the hind paw of rats with logarithmically incremental stiffness for about 6 s each.A positive response was defined as a quick withdrawal or licking of hind paw upon the stimulus.Whenever a positive response to a stimulus occurred, the next lower Von Frey hair was applied, and vise verse.Each rat was tested three more times and the applied force (g) was recorded.Then the average of the threshold was measured as mechanical withdrawal threshold (MWT) [29].

Intrathecal Injection Procedure
The rat was placed in a prone position and the midpoint between the tips of the iliac crest was located.Using a stainless steel needle (30 gauge) by means of lumbar puncture at the intervertebral space of L4-5 or L5-6.The injection did not affect the baseline pain threshold of the rats and a proper injection would be accompanied by a tail flick [30].

Primary Cultures of Spinal Neurons
The spinal cords of the embryos were removed aseptically on day 13 of gestation [31] and digested in 0.15 % typsin at 37 °C for 25 min.The cell suspension was centrifuged at 200 g for 4 min and then resuspended in solution containing DNase and soybean trypsin inhibitor.The solution containing MgCl2 and CaCl 2 was added to the cells for 15 min, and the supernatant was collected and centrifuged at 200 g for 4 min.Neurobasal plating medium containing 10 % fetal bovine serum (FBS) supplement and 1 % l-glutamine was added.Cells were planted onto poly-l-lysine-pretreated 96-well (9 mm) clear-bottomed black plates with a density of 2.5×10 6 cells/well [19,32].

Measurement of Intracellular Ca 2+ Concentration
On day 9, the dye loading buffer containing 4 μM fluo-8 was added (100 μL/well) and incubated for 1 h.Then the cells were subsequently washed 5 times with Locke's buffer (the vehicle), leaving a final volume of 150 μL in each well.The plate was then transferred to a FLIPR (Molecular Devices, Sunnyvale, CA) chamber.Fluorescence reading was taken for 5 min to establish the baseline, and then the first test compound solutions (8 ×) (25 μL) were added to the corresponding well.5 min after the fluorescence readings were taken for, the second compounds (25 μL) were added to the cells and the fluorescence readings were taken for another 10 min.The supernatant of primary culture spinal astrocytes stimulated with LPS for 12 h was added to the spinal neurons at the day 9 for 0.5 h before the measurement of Ca 2+ .

Western Blotting
In brief, spinal cord segments at L4-L6 were collected at 2 h after drug treatment and lysed in RIPA.The supernatant was collected and separated on sodium dodecyl sulfate-polyacrylamide gels, and transferred onto polyvinylidene difluoride membranes.The membranes were blocked with 5 % bovine serum albumin (BSA) for 2 h at room temperature (RT) and incubated with primary antibodies for 3 h at RT and overnight in 4 ℃.Subsequently, the membranes were washed with 0.1 % tris buffered saline tween (TBST) and incubated with secondary antibodies (1:3000) at RT for 2 h.The immunoreactivity was detected using enhanced chemiluminescence (ECL) regents (PerkinElmer, Waltham, MA).Data were analyzed with the associated software Quantity one-4.6.5 (Bio-Rad Laboratories).

Immunofluorescence
The L4-L6 spinal cords were collected after the rats were perfused with 0.01 M PBS followed by 4 % paraformaldehyde (PFA) on day 14 after the model was established.The spinal cords were post-fixed with the same 4 % (PFA) for 1 day and then transferred to 30 % sucrose for 3-5 days.The spinal cords were cut into 25 μm thick segments and blocked with 10 % normal donkey serum containing 0.3 % Triton-X-100 (Sigma-Aldrich, St. Louis, MO).Subsequently, the sections were incubated with the primary antibodies for 16-19 h at 4 ℃ and then incubated with secondary antibodies for 2 h after washed with 0.01 M PBS.The tissue sections were washed with PBS and mounted to be observed under a laser-scanning microscopy (Carl Zeiss LSM700, Germany).To obtain quantitative measurements, 8 images were evaluated for each group and photographed at the same exposure time to generate the raw data.Fluorescence intensities in the different groups were analyzed using Image Pro Plus 6.0 (Media Cybernetics, Silver Spring, MD, USA).

Co-immunoprecipitation
In brief, 10 μg primary antibody was diluted with 500 μL coupling buffer (rat anti-D1DR or mouse anti-D2DR) and added to the NHS magnetic beads.After incubated sufficiently for 4 h at 4 ℃, the supernatant was removed and 500 μL blocking buffer was added to incubated for 1 h at 4 ℃.Tissues (spinal cord segments at L4-L6 of rats) were lysed in ice-cold RIPA buffer, and incubated with beads-Ab heteromers overnight at 4 ℃.After that, the immunoprecipitates were incubated with 100 μL elution buffers for 5 min at RT to dissociate the heteromers.The supernatant was transferred and incubated in SDS sample buffer for 10 min at 100 °C.

Statistical Analysis
All values are depicted as mean ± SEM and the statistical analyses were performed using SPSS Rel 15 (SPSS Inc., Chicago, IL, USA).Data of western blot, immunofluorescence and behavioral tests were statistically analyzed by one-way analysis of variance (ANOVA) and two-way ANOVA followed by Bonferroni's post-hoc tests with significance at P < 0.05.

Results
Blockade of spinal dopamine D1DR and D2DR attenuated TCI-induced bone cancer pain and the expression of D1/D2DR heteromers were significantly increased in TCI-induced bone cancer pain Our results indicated that intrathecal administration of D1DR and D2DR antagonists (5, 10, and 20 μg/20 μL) significantly attenuated TCIinduced bone cancer (Figure 1A, B).Intrathecal administration of D1DR and D2DR siRNA (1 μg/20 μL) for 7 and 8 days respectively also alleviated TCI-induced chronic pain, while control RNA (conRNA) did not affect the mechanical threshold of rats (Figure 1C).Western blot results showed that in the siRNA group, the expression of D1DR and D2DR was decreased in the spinal cord, compared to that in the control (Figure 1D).Furthermore, the immunofluorescence (Figure 1E) and Co-IP (Figure 1F) results showed that D1DR and D2DR co-expressed in the spinal cord of rats.The co-expression of D1DR and D2DR was significantly upregulated in TCI-induced bone cancer pain.Intrathecal administration of D1DR and D2DR antagonists decreased D1DR and D2DR co-expression.

Intrathecal administration of D1DR and D2DR antagonists/siRNA inhibited the expression of p-CaMKII, p-ERK, p-JNK, and p-p38 in the spinal cord
D1/D2DR heteromers couple to Gq to increase intracellular Ca 2+ , which in turn activate several kinases including CaMKII and MAPKs [34].Our results herein indicated that D1DR and D2DR antagonists could decrease the upregulated expression of p-CaMKII, p-ERK, p-JNK, and p-p38 in the spinal cord (Figure 4A-D).And D1DR and D2DR siRNA also significantly decreased the expression of p-CaMKII, p-ERK, p-JNK, and p-p38 (Figure 4E-H).

Discussion
In this study the principal findings are as follows: (1) Intrathecal administration of D1DR/ D2DR antagonists or siRNA could significantly alleviate TCI-induced bone cancer pain; (2) D1DR and D2DR form heteromers in spinal neurons that promote bone cancer pain by activating Gq protein and thereby increasing neuronal excitability, leading to the activation of CaMKII and MAPK signaling; (3) l-CDL, a natural compound could attenuate TCIinduced chronic bone cancer pain through inhibiting D1/D2DR heteromers.
Activating or antagonizing spinal D1DR and D2DR have been reported could inhibit the development of pain [36][37][38][39].And D1DR and D2DR have been reported to form heteromers in rat or monkey brain and have a potentially considerable influence in disorders such as drug addiction, schizophrenia and depression [15][16][17][18].Our previous research also confirmed that D1DR and D2DR form complex to promote neuropathic pain [20].Herein, further results indicated that D1DR and D2DR could form heteromers in the spinal cord, and that intrathecal administration of both D1DR and D2DR antagonists could inhibit the heteromers in TCI rats which was consistent with the previous findings that D1/D2DR heteromers mediated signaling could be attenuated by D1DR and D2DR antagonists, respectively [16,33].D1DR and D2DR antagonistsinduced antinociception could be alleviated by D1DR, D2DR, and D1/D2DR heteromer agonists, which indicated that D1DR and D2DR antagonists attenuate TCI-induced bone cancer through inhibiting D1/D2DR heteromers.
D1/D2DR heteromers were reported to couple to Gq, which might lead to intracellular calcium mobilization from IP3 receptor-sensitive stores through a cascade of events involving rapid translocation of Gq to plasma membrane and activation of PLC.Gq protein, IP3-mediated calcium signaling, and PLC [40,41] have all been reported to mediate nociceptor sensitization.Their effects in bone cancer pain have not been explored, however.To further confirm whether D1DR and D2DR antagonists attenuate TCI-induced bone cancer pain through inhibiting D1/D2DR heteromers with the subsequent suppression of the activation of the Gq-IP3-PLC pathway, the effects of Gq inhibitor YM254890, IP3 inhibitor 2-APB, and PLC inhibitor U73122 on bone cancer pain were explored.Intrathecal administration of YM 254890, 2-APB, and U73122 could all attenuate TCI-induced bone cancer pain.We also wondered whether D1DR and D2DR attenuated chronic bone cancer pain through Gi/o or Gs proteins.It has been reported that Gi/o protein inhibitor Pertussis toxin (PTX) produced hyperalgesia and allodynia [42].Herein, intrathecal administration of AC inhibitor SQ22536 could also attenuate TCI-induced bone cancer pain.In order to verify whether D1DR antagonist can attenuate TCI-induced bone cancer pain through the activation of AC through Gs protein, SKF 83822, which exclusively activates the cAMP [16,43] was used.D1DR and D2DR antagonists-induced antinociception could not be reversed by SKF 83822 indicating that D1DR and D2DR antagonists-induced antinociception was not mediated through the inhibition of the AC pathway.
Further in vitro studies were conducted to confirm whether D1DR and D2DR form heteromers that increase the excitatory state of neurons.Spontaneous Ca 2+ transients have been implicated in regulating plasticity in developing neurons [44].Primary cultured spinal neurons display synchronized spontaneous Ca 2+ oscillations [19].Our study showed that administration of D1DR and D2DR antagonists could reduce the Ca 2+ oscillations in spinal neurons.Administration of D1DR, D2DR, and D1/D2DR heteromer agonists increased Ca 2+ oscillations, which in turn could be reduced by D1DR and D2DR antagonists.D1/D2DR heteromers couple to Gq, which activates CaMKII that in turn also promote the development of chronic pain through activating MAPK [34].Expression of p-CaMKII, p-ERK, p-JNK, and p-p38 was observed.Both D1DR, D2DR antagonists, and siRNA could also inhibit the expression of p-CaMKII, p-ERK, p-JNK, and p-p38.
The important role of D1/D2DR heteromers in TCI-induced bone cancer pain make them an attractive target to attenuate bone cancer pain.l-CDL, a trace ingredient from traditional Chinese medicine could significantly attenuate chronic bone cancer pain and other models of neuropathic pain in our previous studies without notable side effects [25,26,29].l-CDL belongs to the tetrahydroprotoberberines (THPBs), which have been reported to show affinity to dopamine receptors and possess a variety of beneficial effects without notable side effects [21,22].l-CDL showed high affinities to both D1DR (see Additional file 1: Figure s1) and D2DR [35] with IC50 of 0.20 μM and 0.86 μM respectively.Herein, our results indicated that l-CDL-induced antinociception could be reversed by D1DR, D2DR, and D1/D2DR heteromer agonists but not SKF 83822, which exclusively activates the cAMP.l-CDL-induced inhibition of p-CaMKII, p-ERK, p-JNK, and p-p38 could be alleviated by D1DR, D2DR, and D1/D2DR heteromer agonists.l-CDL has been reported in our previous study to alleviate TCI-induced chronic bone cancer pain through inhibiting NMDA and mGlu1/5 receptors [29].Synergistic effects in such a multi-targets approach might largely explain the strong observed effects in attenuating bone cancer pain.
We provided the first experimental evidence that spinal D1DR and D2DR might promote chronic bone cancer pain through forming D1/D2DR heteromers, thereby leading to the activation of Gq proteins and the downstream CaMKII and MAPK signaling to increase excitability in spinal neurons.l-CDL, a natural compound, was found to attenuate TCIinduced chronic bone cancer pain by antagonizing spinal D1DR and D2DR to inhibit D1/D2DR heteromers, and in turn downstream CaMKII and MAPKs signaling.Altogether, the findings may provide new avenues to find more effective and safer targets for chronic bone cancer pain.