Secondary Logo

Share this article on:

Involvement of lysophosphatidic acid–induced astrocyte activation underlying the maintenance of partial sciatic nerve injury–induced neuropathic pain

Ueda, Hiroshi*; Neyama, Hiroyuki; Nagai, Jun; Matsushita, Yosuke; Tsukahara, Tamotsu; Tsukahara, Ryoko

doi: 10.1097/j.pain.0000000000001316
Research Paper
Editor's Choice

We have previously demonstrated that lysophosphatidic acid (LPA) plays key roles in the initial mechanisms for neuropathic pain (NeuP) development. Here, we examined whether LPA receptor mechanisms and LPA production are related to the glial activation at a late stage after partial sciatic nerve ligation (pSNL) by use of microglial inhibitor, Mac1-saporin or astrocyte inhibitor, and L-α-aminoadipate (L-AA). Although single intrathecal injection of LPA1/3 antagonist, Ki-16425 did not affect the pain threshold at day 7 after the spinal cord injury, repeated treatments of each compound gradually reversed the basal pain threshold to the control level. The intrathecal administration of a microglia inhibitor, Mac-1-saporin reversed the late hyperalgesia and LPA production at day 14 after the pSNL, whereas L-AA inhibited the hyperalgesia, but had no effect on LPA production. The involvement of LPA receptors in astrocyte activation in vivo was evidenced by the findings that Ki-16425 treatments abolished the upregulation of CXCL1 in activated astrocytes in the spinal dorsal horn of mice at day 14 after the pSNL, and that Ki-16425 reversed the LPA-induced upregulation of several chemokine gene expressions in primary cultured astrocytes. Finally, we found that significant hyperalgesia was observed with intrathecal administration of primary cultured astrocytes, which had been stimulated by LPA in a Ki-16425–reversible manner. All these findings suggest that LPA production and LPA1/3 receptor activation through differential glial mechanisms play key roles in the maintenance as well as initiation mechanisms in NeuP.

This study describes a mechanism underlying the maintenance of neuropathic pain through lysophosphatidic acid production and actions, which are mediated differentially by microglia and astrocytes.

Department of Pharmacology and Therapeutic Innovation, Nagasaki University, Graduate School of Biomedical Sciences, Nagasaki, Japan

Corresponding author. Address: Department of Pharmacology and Therapeutic Innovation, Nagasaki University Graduate School of Biomedical Sciences, 1-14 Bunkyo-machi, Nagasaki 852-8521, Japan. Tel.: +81-95-819-2421; fax: +81-05-819-2420. E-mail address: ueda@nagasaki-u.ac.jp (H. Ueda).

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

Received January 18, 2018

Received in revised form May 28, 2018

Accepted May 29, 2018

Back to Top | Article Outline

1. Introduction

Neuropathic pain (NeuP) is often characterized by abnormally hypersensitive sensory perception, called hyperalgesia or allodynia, in which innocuous (tactile) stimuli cause intense pain. Accumulating evidence suggests that spinal cord mechanisms underlying NeuP are characterized by temporal activation of microglia (within days) and late activation of astrocytes (within days to weeks).6,10,11 Under nerve injury conditions, microglia produce multiple inflammatory mediators such as IL-1β and TNF-α, subsequently activate c-Jun N-terminal Kinase (JNK) and induce the production of chemokines such as C–C motif chemokine ligand 2 (CCL2) and C-X-C motif chemokine ligand 1 (CXCL1) in astrocytes, which are implicated in pain sensitization.44,45 Accordingly, the astrocyte activation at the late stage after nerve injury would be a good target for the treatment of established NeuP, although the detailed mechanisms remain elusive.

We have reported that lysophosphatidic acid (LPA) produced by partial sciatic nerve ligation (pSNL) initiates abnormal hyperalgesia and allodynia through LPA1 receptor mechanisms.9 LPA1 signals also cause demyelination of the dorsal root (DR) fibers, which may cause functional cross-talk between noxious and innocuous sensory fibers underlying allodynia, and upregulation of CaVα2δ19 and ephrin B130 genes in the DR ganglion, which may cause an enhanced spinal pain transmission underlying hyperalgesia. The roles of LPA signals in NeuP also have been supported by many other studies.1,7,13,19,38,43 Pharmacological findings4,14,16,18,31 suggest that LPA signals play key roles in the initiation of NeuP.9,32–35 As the LPA1-mediated demyelination is only found in the DR, but not in the sciatic nerve, which was given partial ligation,21 it is presumed that LPA is produced in the spinal cord and goes back to the DR. Recent studies have revealed that LPA amplifies the LPA production in in vivo and in vitro studies. This feed-forward system is mediated by LPA1, LPA3, and microglia-derived IL-1β.15–17,22,41 In this study, we investigated the roles of LPA signaling in the maintenance of NeuP and underlying mechanisms, by studying the pharmacological blockade of established NeuP by LPA receptor antagonist and glial actions affecting the LPA signaling at the late stage after the pSNL.

Back to Top | Article Outline

2. Methods

2.1. Experimental animals

Male C57BL/6J mice (6-8 weeks old) purchased from TEXAM corporation (Nagasaki, Japan) were used. They were kept in a room maintained at 22 ± 3°C and 55 ± 10% relative humidity for a 12-hour period (lights on at 8:00 AM and off at 8:00 PM), and had free access to a standard laboratory diet and tap water. We used 179 mice for the entire set of experiments. All procedures used in this work were approved by the Nagasaki University Animal Care Committee, and complied with fundamental guidelines for the proper conduct of animal experiments and related activities in academic research institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Back to Top | Article Outline

2.2. Drugs

18:1-LPA was purchased from Cayman Chemicals (Ann Arbor, MI). Ki-16425, an LPA1 and LPA3 antagonist,23 was generously provided by Kirin Brewery Co (Takasaki, Japan), and was dissolved in sesame oil (Sigma, St. Louis, MO) just before administration. For subcutaneous (s.c.) injection, these drugs were dissolved in physiological saline or artificial cerebrospinal fluid (aCSF: 125 mM NaCl, 3.8 mM KCl, 1.2 mM KH2PO4, 26 mM NaHCO3, and 10 mM glucose), respectively. L-α-Aminoadipate (L-AA), an astroglial toxin,39 was purchased from Sigma (St. Louis, MO) and administered intrathecally (i.t.) in an amount of 30 nmol per 5 μL of aCSF. Mac1-saporin (a chemical conjugate of mouse monoclonal antibody to CD11b and the ribosome-inactivating protein, saporin) was purchased from Advanced Targeting Systems (San Diego, CA) and injected i.t. in an amount of 30 nmol per 5 μL of aCSF to deplete microglia, as reported2,42

Back to Top | Article Outline

2.3. Nerve injury

For the preparation of peripheral neuropathic pain model, the pSNL was performed as previously described.14 Briefly, the sciatic nerve of the right hind limb was exposed at the high thigh level through a small incision, and the dorsal half of nerve thickness was tightly ligated with a silk suture. We used 50 mg/kg of sodium pentobarbital intraperitoneally (Nacalai Tesque, Kyoto, Japan), which anesthetized and suppressed nociceptive responses at least for 3 hours after the surgery.

Back to Top | Article Outline

2.4. Nociceptive tests

Thermal paw withdrawal tests were used to evaluate the hyperalgesia. In the thermal paw withdrawal test, latency of paw withdrawal on a thermal stimulus was assessed as previously reported.29 Unanesthetized animals were placed in Plexiglas cages on top of a glass sheet, and an adaptation period of 1 hour was allowed. The thermal stimulator (IITC Inc, Woodland Hills, CA) was positioned under the glass sheet, and the focus of the projection bulb was aimed exactly at the middle of right hind paw plantar surface of mouse. A mirror attached to the stimulator permitted visualization of the plantar surface. A cutoff time of 20 seconds was set to prevent tissue damage. During the thermal test, the thresholds were determined from 3 repeated tests at 10-minute intervals. In the mechanical paw pressure test, mice were placed in a Plexiglass chamber on a 6 × 6-mm wire mesh grid floor and allowed to acclimatize for a period of 1 hour, as reported.29,36 A mechanical stimulus was then delivered to the middle of the right hind paw plantar surface using Electronic digital von Frey Anesthesiometer and Rigid Tip (Model 2390, 90 g probe, 0.8 mm in diameter; IITC Inc, Woodland Hills, CA). The pressure needed to induce a flexor response was defined as the pain threshold. A cutoff pressure of 20 g was set to avoid tissue damage.

Back to Top | Article Outline

2.5. Quantification of lysophosphatidic acid by LC-MS/MS

The quantification of LPA was performed as previously described.8 Briefly, the acutely isolated mouse spinal dorsal horn was mixed and sonicated in a total volume of 200 µL MeOH. After centrifugation, 50 µL of the supernatant was diluted 4-fold in MeOH supplemented with an internal standard (2.5 pmol of LPA 17:0). After mixing and centrifugation, the supernatant was filtrated with Duo-filter (0.2 μm pore size, 4 mm inner diameter; YMC, Kyoto, Japan) and used for the analysis. An aliquot (40 µL) of each methanol-extracted sample was injected into the LC-MS/MS system. LC-MS/MS was performed using a TSQ Quantum Ultra triple quadrupole MS (Thermo Fisher Scientific, Waltham, MA) equipped with a heated-electrospray ionization-II (HESI-II) source, with a NANOSPACE SI-II HPLC (Shiseido, Tokyo, Japan). Lysophosphatidic acid was separated on a Capcell Pak ACR C18–reversed phase column (1.5 × 250 mm; Shiseido) using a gradient of 2 solvents, solvent A (5 mM ammonium formate in water, pH4.0) and solvent B (5 mM ammonium formate in 95% [vol/vol] acetonitrile, pH4.0), at a flow rate of 150 µL/minute. The initial condition was set at 60% of solvent B. The following solvent gradient was applied: maintained 60% of solvent B and then followed by a linear gradient to 95% of solvent B from 0.2 to 14 minutes, and held at 95% of B for 3 minutes. Subsequently, the mobile phase was returned to the initial conditions and maintained for 5 minutes. The column eluate was introduced into the MS between 8 and 23 minutes after injection. The column was washed after each sample was applied, the following solvent gradient was used (flow rate: 150 µL/minute): started at 60% of solvent B, a linear gradient to 95% B over 2 minutes, and held at 95% B for 6 minutes, then return to the initial condition and kept for 7 minutes. For MS/MS analysis, the negative HESI-II spray voltage was set at 2500 V, the heated capillary temperature was 350°C, the sheath gas pressure was 65 psi, the auxiliary gas setting was 20 psi, and the heated vaporizer temperature was set at 350°C. Nitrogen was used for both the sheath and the auxiliary gases, and the collision gas (argon) was set at 1.5 mTorr. All the data were acquired using Xcalibur 2.2 operating software (Thermo Fisher Scientific). 18:1-LPA was analyzed by multiple reaction monitoring in negative ion mode. Q1 was set for the deprotonated molecular ion for (m/z 435.47 for 18:1-LPA 18:1). The relative amount of each LPA was calculated from the ratios of their peak areas using the same software as described above.

Back to Top | Article Outline

2.6. Flow cytometry

Freshly isolated lumbar spinal cords (L4-L6) from 5 mice were cut into small fragments (2-3 mm) and transferred into C-tube of gentleMACS to dissociate single-cell suspension, by the manufacture's protocol (Miltenyi Biotec, Bergisch Gladbach, Germany). The tissue was further dissociated using Neural Tissue Dissociation Kit (T) in combination with the gentleMACS Dissociator. Briefly, enzyme Mix1 (solution 1: 50 µL and solution 2: 1900 µL) and 1.3 µL of GolgiStop (BD Biosciences, San Jose, CA) were added to the tissue suspension, followed by homogenization using gentleMACS Dissociator (m_brain_01) and incubation at 37°C for 15 minutes under continuous agitation with MACSMix Tube Rotator. The tube was then resuspended in a gentleMACS Dissociator (m_brain_02) and added enzyme mix 2 (solution 3: 20 mL and solution 4: 10 mL) and incubated with inversion for another 10 minutes at 37°C. After rehomogenization (m_brain_03) and incubation at 37°C for 10 minutes, Hank's balanced salt solution was added to the dissociated cells and centrifuged at 300g for 10 minutes. The supernatant was removed and the cells were resuspended in 30% Percol. After centrifugation at 500g for 25 minutes at 20°C, the cell pellet was washed with Hank's balanced salt solution by use of centrifugation at 1700 rpm for 5 minutes. The cells were then suspended with phosphate-buffered saline (PBS) and treated with filtration with 70-µm cell strainer (BD Biosciences) to obtain single-cell suspension. The dissociated spinal cord cells were stained with anti-mouse CD11b-PE-Cyanine7 (1:400; eBioscience, San Diego, CA) and ACSA-2-APC (1:40; Miltenyi Biotec) for 30 minutes at 4°C under the condition of protection from light. The anti-mouse CD16/CD32 (FcγIII/II Receptor, clone 2.4G2, 1:800; BD Science) was used to block nonspecific staining of antibodies. Mouse IgG1 κ Isotype Control, PE-Cyanine7 and mouse IgG1 Isotype Control, and APC (eBioscience) were used as negative control. After staining, the cells were washed with PBS containing 2% fetal calf serum twice and filtered through 70-µm nylon mesh. BD FACSVerse flow cytometer (BD biosciences) and FLOWJO (Tomy Digital Biology Co, Ltd, Tokyo, Japan) were used for the fluorescent intensity of ACSA-2- or CD11b-positive population of astrocytes or microglia, respectively.

Back to Top | Article Outline

2.7. Experiments using cultured astrocytes

For the primary culture of astrocytes, whole brains of 3 neonatal mice were placed in DMEM (Nacalai Tesque). Cerebral hemispheres without the olfactory bulb and cerebellum from 12 neonates were cut into small pieces using sharp blades and grinded by a Dounce homogenizer 10 times in 5 mL of DMEM with antibiotics penicillin and streptomycin (Sigma-Aldrich, P4333, MO). The pellet after the centrifugation of cell suspension (0.5 mL) at 1000g for 10 minutes was plated on T175 culture flask (Corning, #431466, NY) and incubated at 37°C in a CO2 incubator. When cells become confluent (7 or 8 days in culture), overlaying microglia on the astrocyte layer were removed by shaking, followed by aspiration. Astrocytes were harvested by adding 1 mL of trypsin-EDTA (Sigma-Aldrich, T4049, MO), followed by incubation at 37°C for 2 to 3 minutes until cells detach from the flask. Harvested astrocytes were added by 5 mL DMEM with 10% fetal bovine serum to inactivate the trypsin-EDTA, followed by centrifugation at 1000g for 10 minutes, and plated in 2 T175 culture flasks at 37°C in a CO2 incubator. By changing the medium every 2 to 3 days, astrocytes 12 to 14 days in culture were split in an appropriate cell concentration, 24 to 48 hours before use of further experiments. Astrocytes (107 cells) were treated with 1 or 10 μM LPA with or without 5 μM Ki-16425 in Nunclon Delta surface (#150350; Thermo Fisher Scientific) at 37°C for 3 hours in a CO2 incubator, and cells were harvested by scraper gently and suspended in PBS. After twice washing with PBS, cells were counted using a TC-20 automated cell counter (Bio-Rad, Hercules, CA). The cells were resuspended in PBS to make a final concentration of astrocytes 104 cells/5 μL in PBS and used for i.t. injection to assess the nociceptive activity.

Back to Top | Article Outline

2.8. Real-time quantitative reverse transcription polymerase chain reaction

Total RNA was prepared from LPA-treated astrocytes by use of TRizol Reagent (Thermo Fisher Scientific). After purification, the amount of RNA was measured spectrophotometrically using OD260, and the quality of RNA was checked by spectrophotometric analysis using OD260/280. Total RNA (500 ng) was converted into first-strand cDNA by PrimeScript RT Reagent Kit (Takara Bio Inc, Otsu, Japan). The cycling parameters were 10 minutes at 95°C followed by 45 cycles of 15 seconds at 95°C and 1 minute at 60°C. Primer sets used for real-time PCR are shown in Table 1. The obtained data were normalized to the expression of GAPDH and analyzed by the Comparative CT Method.

Table 1

Table 1

Back to Top | Article Outline

2.9. Immunohistochemistry

Anesthetized mice were transcardially perfused with cold potassium-free PBS (K+-free PBS, pH7.4), followed by cold 4% paraformaldehyde solution. The fixed spinal cord was immersed overnight in 25% sucrose solution. The tissues were fast-frozen in cryoembedding compound on a mixture of ethanol and dry ice and stored at −80°C until use. The spinal cord was cut at 30 μm with a cryostat. For immunolabeling, spinal cord sections were washed 3 times with 0.1% Triton X-100 in K+-free PBS (PBST), followed by permeabilization with 50% and 100% methanol for 10 minutes each at room temperature (RT). The sections were then washed 3 times with PBST and preincubated in blocking buffer (2% bovine serum albumin [BSA] with 0.1% Triton X-100 in K+-free PBS) for 1 hour at RT. Next, the sections were then incubated with rabbit anti-CXCL1 polyclonal antibody (1:200; Boster Biological Technology Co, Ltd, Pleasanton, CA) and mouse anti-glial fibrillary acidic protein (anti-GFAP) monoclonal antibody (clone GA5, 1:500; Millipore Corporation, Burlington, MA) overnight at 4°C, washed with PBST, and incubated with Alexa 488-conjugated donkey anti-rabbit IgG antibody (1:300; Thermo Fisher) and anti-mouse IgG antibody (1:300; Thermo Fisher) for 2 hours at RT. The specificity of anti-CXCL1 and anti-GFAP antibodies has been previously reported.20,45 The immunolabeled sections were mounted with Fluoromount (Diagnostic BioSystems, Pleasanton, CA) on silane-coated glass slides and analyzed using a fluorescence microscope (Biorevo; Keyence, Tokyo, Japan). For the quantification of CXCL1 intensity, the imaging cytometry was performed using an image analyzer, IN Cell Analyzer 2000, and Developer toolbox software (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom).28

Back to Top | Article Outline

2.10. Statistical analysis

All data are expressed as the mean ± SE. Data were analyzed with Graphpad Prism 7.0 software (Graphpad Software, San diago, CA) using the unpaired t test, 1-way ANOVA with the Tukey multiple comparisons test or the Dunnett multiple comparisons test, and 2-way ANOVA with the Tukey multiple comparisons test or the Bonferroni multiple comparisons test. Differences with a P value of less than 0.05 were considered statistically significant.

Back to Top | Article Outline

3. Results

3.1. Extinction of established NeuP by repeated LPA1/3 antagonist treatments

As shown in Figure 1A, the single injection of Ki-16425 at a dose of 30 mg/kg s.c. did not affect the pain threshold in the thermal nociception test at day 7 after the pSNL. However, the repeated treatments with this compound twice daily for 8 days staring from day 7 showed a gradual recovery of the basal pain threshold to the normal pain threshold, as shown in Figure 1B. It should be noted that the recovery of pain threshold remains evident at day 18, even after the cessation of treatments, suggesting that the pain memory may disappear by repeated blockade of LPA1/3 signaling. The Ki-16425–reversible abnormal pain in the mechanical nociception test was also observed at day 19 (Fig. 1C).

Figure 1

Figure 1

Back to Top | Article Outline

3.2. Selective astrocyte depletion reverses hyperalgesia, but not lysophosphatidic acid production

The dissociated spinal cord cells, which had been stained with anti-mouse CD11b-PE-Cyanine7 and ACSA-2-APC, were applied to flow cytometer. As shown in Figure 2A, the cell population of microglia (high CD11b-PE-Cyanine7 and low ACSA-2-APC) of aCSF-treated control mice was found in the Q1 compartment, whereas that of astrocytes (low CD11b-PE-Cyanine7 and high ACSA-2-APC) was found in Q3 compartment. When control mice were treated with 30 nmol (i.t.) of L-AA, an astrocyte toxin twice at days 1 and 2, the cell population of astrocytes harvested at day 4 was significantly decreased, whereas there was no significant change in the population of microglia (Figs. 2B–D). L-AA treatments also significantly reversed the thermal hyperalgesia, as well as Mac-1-saporin, a known microglial toxin (Fig. 2E). In the mechanical nociception test, mice at day 14 after the pSNL showed a significant hyperalgesia, which was reversed by the treatment with L-AA at 30 nmol (i.t.) (Fig. 2F). On the other hand, pSNL-induced increase in 18:1-LPA levels in the spinal dorsal horn at day 14 was reversed by Mac1-saporin treatments, but not by L-AA treatments (Fig. 2G).

Figure 2

Figure 2

Back to Top | Article Outline

3.3. Blockade of upregulated chemokine expression in astrocytes by LPA1/3 antagonist

Previous reports have demonstrated that astrocytes are activated and chemokine levels are upregulated at the late stage after nerve injury.5,45 In this study, we confirmed the upregulation of a representative chemokine, CXCL1-like immunoreactivity in the activated astrocytes showing intense GFAP-like immunoreactivity in the dorsal horn of spinal cord at day 14 after the pSNL (Figs. 3A, E, I, C, G, and K). The upregulation of CXCL1 was abolished by treatments with Ki-16425 at 30 mg/kg (s.c.) twice daily from day 7 to day 13 and once 4 hours before the spinal cord isolation (Figs. 3B, F, J, D, H, and L). When the intensity of CXCL1 signals in whole astrocytes was evaluated by Imaging cytometry using In Cell Analyser software, Developer Toolbox v.1.92 software (GE Healthcare), as reported elsewhere,28 the injury-induced significant increase in CXCL1 signals and its blockade by Ki-16425 were confirmed (Fig. 4).

Figure 3

Figure 3

Figure 4

Figure 4

Back to Top | Article Outline

3.4. Lysophosphatidic acid–induced upregulation of chemokine genes in cultured astrocytes

To see LPA actions on chemokine gene expression in cultured astrocytes, the culture medium was first replaced with DMEM (0.1% BSA) containing vehicle or Ki-16425 (5 μM) for 1 hour before LPA addition. Six hours after the addition of 10 μM LPA, they were washed twice by PBS without calcium and magnesium, and resuspended in PBS. An aliquot of astrocyte suspension was used for the study of chemokine gene expression. Real-time PCR analysis revealed that cultured astrocytes express LPA6 > LPA1 > LPA4 > LPA2 in the order of expression, but do not express LPA3 and LPA5. The incubation with 10 μM LPA (with 0.1% BSA) for 6 hours did not affect the expression of LPA receptors (Fig. 4A). The study of chemokine gene expression revealed that LPA addition upregulated the transcription of CXCL1, CCL7, CCL2, and CCL5, as elsewhere reported in the case with nerve injury5 (Figs. 4B–E). Lysophosphatidic acid–induced upregulation of gene expression of CXCL1 and CCL7, but not CCL2 or CCL5 was significantly reversed by co-incubation with Ki-16425 (Fig. 5).

Figure 5

Figure 5

Back to Top | Article Outline

3.5. Hyperalgesia by intrathecal injection of lysophosphatidic acid–primed mouse astrocytes

To see in vivo effects of LPA-primed astrocytes, the medium of astrocyte culture was replaced with DMEM (0.1% BSA) with or without Ki-16425 (5 μM) for 1 hour before the addition of LPA. Six hours after the addition of 10 μM LPA, they were washed twice by PBS without calcium and magnesium and used for the i.t. injection of astrocytes (106 cells/mouse). As seen in Figure 6A, a single injection of 1 nmol of LPA (i.t.) showed potent hyperalgesic effects at day 1 through day 10 after the injection. The LPA-primed astrocytes also showed significant hyperalgesic effects for 7 days with peak effects at day 1 and 2, whereas astrocytes without LPA treatment did not. When the LPA treatment was performed in the presence of 5 μM Ki-16425, the LPA priming–induced hyperalgesia at day 1 was significantly inhibited (Fig. 6B).

Figure 6

Figure 6

Back to Top | Article Outline

4. Discussion

Accumulating evidence demonstrates that LPA plays key roles in the development of pSNL-induced NeuP.34 Underlying experimental findings in terms of peripheral nerves are as follows; LPA1-mediated demyelination causes functional cross-talk between noxious (C or Aδ) fibers and innocuous (Aβ) fibers,40 which explains a mechanism for allodynia, and the upregulation of pain-related molecules, including Cavα2δ1,9 which is a pharmacological target for pregabalin, a representative medicine for NeuP.3 Experimental findings in the dorsal horn of the spinal cord, on the other hand, are as follows: LPA1- and LPA3-mediated amplification of LPA production,18 involvement of microglia-derived IL-1β in the LPA production,41 LPA3-mediated activation of cultured microglia causes ATP release, P2X4 activation, and production of brain-derived neurotrophic factor,4 which is reported to switch the hyperpolarizing GABAA function to depolarizing one through a downregulation of potassium-chloride transporter member 5 (KCC2).25 All these LPA-mediated mechanisms may form a feed-forward system underlying the development of NeuP.

One important question to be answered is whether LPA receptor signaling also may be involved in the maintenance of NeuP. As a first attempt, we tested the effects of LPA1/3 antagonist, Ki-16425 on NeuP at day 7 after the pSNL, although this inhibitor failed to show any acute antihyperalgesic effects. Repeated treatments with Ki-16425 for 8 days starting at day 7 increased the pain threshold to nearly normal levels over this time course. Of most interest is the finding that largely completes recovery of normal pain threshold lasted at least for 4 or 5 more days after the cessation of Ki-16425 treatments. All these findings indicate that the blockade of LPA1/3 signaling may erase the neural feed-forward plasticity underlying so-called pain memory. Detailed molecular mechanisms are still in progress, but we obtained the findings that microglia, but not astrocytes are involved in the LPA production at the late stage as well as the initial/early stage, whereas LPA-activated astrocytes inhibit the maintenance of NeuP, possibly through chemokines.5 In this study, we found that LPA production measured by LC-MS/MS and hyperalgesia in the thermal nociception test were blocked by the late treatments of Mac1-saporin. This is in contrast with previous findings, which demonstrated that the treatments with minocycline, a microglial inhibitor for 4 consecutive days starting at 1 day after the pSNL did not affect the NeuP, as reported,24 whereas the treatments of minocycline before the pSNL blocked NeuP.16 One of the mechanisms for this discrepancy could be the different time points at which microglia inhibitor was used (days 2-5 vs days 8, 10, and 12). Alternatively, the difference in findings may be related to the unique potency of Mac-1-saporin, which is reported to deplete microglia.12,27 If it is true that the inhibition of microglia activities by Mac-1-saporin is more potent than the inhibition by minocycline, unidentified but more intense microglial activation, possibly due to neurons–astrocytesmicroglia tripartite interactions may occur at the later stage after the pSNL.

As the activation of astrocytes after peripheral nerve injury occurs later than the microglial activation, it is possible that astrocytes play a role in the maintenance rather than the development of NeuP.6,10,11 This speculation is supported by studies showing that several chemokines are expressed in the late stage of astrocytes after nerve injury, and that NeuP was blocked by neutralizing antibodies against chemokines.5 In this study, we confirmed the roles of astrocytes by pretreatment with astrocyte toxin L-AA, which selectively reduced the population of astrocytes without a change in the microglia population in the analysis using flow cytometry. The astrocyte toxin pretreatment significantly inhibited the NeuP, whereas it had no effect on LPA production. We confirmed astrocyte activation in the late stage at day 14 after the pSNL in terms of GFAP expression in Western blot and morphology in the immunohistochemistry. In addition, the expression of CXCL1, a representative chemokine is also upregulated in the activated astrocytes. As Ki-16425 significantly abolished the increased expression of both GFAP and CXCL1 in the imaging cytometry, it is evident that LPA1/3 is involved in astrocyte activation and chemokine expression in the late stage after the pSNL. To examine whether LPA receptor signaling has direct effects on chemokine expression, we tested the effects of LPA on chemokine expressions in primary cultured astrocytes prepared from the brain. The real-time PCR analysis confirmed the LPA-induced upregulated transcription of several chemokine genes. Among these chemokine genes, the LPA-induced upregulation of CXCL1 and CCL7 gene expressions was significantly abolished by Ki-16425, which has selective antagonistic activities to LPA1 and LPA3. Furthermore, the LPA-primed astrocytes caused Ki-16425–reversible hyperalgesia. As astrocytes have no expression of LPA3, the LPA-induced chemokine gene expression and LPA-activated hyperalgesia are presumably mediated by LPA1. Thus, it is presumed that LPA1-mediated astrocyte activation may play roles in the neurons–astrocytesmicroglia tripartite interaction responsible for pain memory. Although microglia was found to play roles in LPA production even at the late stage, the mechanisms underlying microglia activation at the late stage remain elusive. Not only spinal, but peripheral mechanisms including upregulated or downregulated transcription of key molecules regulating the pain transmission26,33,37 may continuously activate microglia and following the maintenance of LPA production.

In conclusion, here we found that repeated treatments with LPA1/3 antagonist significantly reduced the established NeuP at least for several days even after the treatments. Although microglia-depleting toxin, but not astrocyte-depleting toxin inhibited the LPA production at the late stage, LPA-activated astrocytes at the late stage were found to play roles in the maintenance of NeuP.

Back to Top | Article Outline

Conflict of interest statement

The authors have no conflict of interest to declare.

This work was supported in part by Grants-in-Aid for the Platform for Drug Discovery, Informatics, and Structural Life Science [16am0101012j0005] (H.U.) from the Japan Agency for Medical Research and Development (AMED), Japan; and JSPS KAKENHI Grant Numbers JP17H01586 (H.U.) and JP26253077 (H.U.).

Back to Top | Article Outline

Acknowledgements

The authors are grateful to Takehiro Mukae and Shuhei Yamagishi for the technical help of evaluation in the pain and real-time PCR, respectively.

Back to Top | Article Outline

References

[1]. Ahn DK, Lee SY, Han SR, Ju JS, Yang GY, Lee MK, Youn DH, Bae YC. Intratrigeminal ganglionic injection of LPA causes neuropathic pain-like behavior and demyelination in rats. PAIN 2009;146:114–20.
[2]. Echeverry S, Shi XQ, Yang M, Huang H, Wu Y, Lorenzo LE, Perez-Sanchez J, Bonin RP, De Koninck Y, Zhang J. Spinal microglia are required for long-term maintenance of neuropathic pain. PAIN 2017;158:1792–801.
[3]. Field MJ, Cox PJ, Stott E, Melrose H, Offord J, Su TZ, Bramwell S, Corradini L, England S, Winks J, Kinloch RA, Hendrich J, Dolphin AC, Webb T, Williams D. Identification of the alpha2-delta-1 subunit of voltage-dependent calcium channels as a molecular target for pain mediating the analgesic actions of pregabalin. Proc Natl Acad Sci U S A 2006;103:17537–42.
[4]. Fujita R, Ma Y, Ueda H. Lysophosphatidic acid-induced membrane ruffling and brain-derived neurotrophic factor gene expression are mediated by ATP release in primary microglia. J Neurochem 2008;107:152–60.
[5]. Gao YJ, Ji RR. Chemokines, neuronal-glial interactions, and central processing of neuropathic pain. Pharmacol Ther 2010;126:56–68.
[6]. Grace PM, Hutchinson MR, Maier SF, Watkins LR. Pathological pain and the neuroimmune interface. Nat Rev Immunol 2014;14:217–31.
[7]. Hayashi Y, Kawaji K, Sun L, Zhang X, Koyano K, Yokoyama T, Kohsaka S, Inoue K, Nakanishi H. Microglial Ca(2+)-activated K(+) channels are possible molecular targets for the analgesic effects of S-ketamine on neuropathic pain. J Neurosci 2011;31:17370–82.
[8]. Inoue A, Arima N, Ishiguro J, Prestwich GD, Arai H, Aoki J. LPA-producing enzyme PA-PLA(1)alpha regulates hair follicle development by modulating EGFR signalling. EMBO J 2011;30:4248–60.
[9]. Inoue M, Rashid MH, Fujita R, Contos JJ, Chun J, Ueda H. Initiation of neuropathic pain requires lysophosphatidic acid receptor signaling. Nat Med 2004;10:712–18.
[10]. Ji RR, Chamessian A, Zhang YQ. Pain regulation by non-neuronal cells and inflammation. Science 2016;354:572–7.
[11]. Ji RR, Xu ZZ, Gao YJ. Emerging targets in neuroinflammation-driven chronic pain. Nat Rev Drug Discov 2014;13:533–48.
[12]. Kanai T, Watanabe M, Okazawa A, Sato T, Yamazaki M, Okamoto S, Ishii H, Totsuka T, Iiyama R, Okamoto R, Ikeda M, Kurimoto M, Takeda K, Akira S, Hibi T. Macrophage-derived IL-18-mediated intestinal inflammation in the murine model of Crohn's disease. Gastroenterology 2001;121:875–88.
[13]. Lin ME, Rivera RR, Chun J. Targeted deletion of LPA5 identifies novel roles for lysophosphatidic acid signaling in development of neuropathic pain. J Biol Chem 2012;287:17608–17.
[14]. Ma L, Matsumoto M, Xie W, Inoue M, Ueda H. Evidence for lysophosphatidic acid 1 receptor signaling in the early phase of neuropathic pain mechanisms in experiments using Ki-16425, a lysophosphatidic acid 1 receptor antagonist. J Neurochem 2009;109:603–10.
[15]. Ma L, Nagai J, Chun J, Ueda H. An LPA species (18:1 LPA) plays key roles in the self-amplification of spinal LPA production in the peripheral neuropathic pain model. Mol Pain 2013;9:29.
[16]. Ma L, Nagai J, Ueda H. Microglial activation mediates de novo lysophosphatidic acid production in a model of neuropathic pain. J Neurochem 2010;115:643–53.
[17]. Ma L, Uchida H, Nagai J, Inoue M, Aoki J, Ueda H. Evidence for de novo synthesis of lysophosphatidic acid in the spinal cord through phospholipase A2 and autotaxin in nerve injury-induced neuropathic pain. J Pharmacol Exp Ther 2010;333:540–6.
[18]. Ma L, Uchida H, Nagai J, Inoue M, Chun J, Aoki J, Ueda H. Lysophosphatidic acid-3 receptor-mediated feed-forward production of lysophosphatidic acid: an initiator of nerve injury-induced neuropathic pain. Mol Pain 2009;5:64.
[19]. McDougall JJ, Albacete S, Schuelert N, Mitchell PG, Lin C, Oskins JL, Bui HH, Chambers MG. Lysophosphatidic acid provides a missing link between osteoarthritis and joint neuropathic pain. Osteoarthritis Cartilage 2017;25:926–34.
[20]. Mudo G, Belluardo N, Mauro A, Fuxe K. Acute intermittent nicotine treatment induces fibroblast growth factor-2 in the subventricular zone of the adult rat brain and enhances neuronal precursor cell proliferation. Neuroscience 2007;145:470–83.
[21]. Nagai J, Uchida H, Matsushita Y, Yano R, Ueda M, Niwa M, Aoki J, Chun J, Ueda H. Autotaxin and lysophosphatidic acid 1 receptor-mediated demyelination of dorsal root fibers by sciatic nerve injury and intrathecal lysophosphatidylcholine. Mol Pain 2010;6:78.
[22]. Nagai J, Ueda H. Pre-emptive morphine treatment abolishes nerve injury-induced lysophospholipid synthesis in mass spectrometrical analysis. J Neurochem 2011;118:256–65.
[23]. Ohta H, Sato K, Murata N, Damirin A, Malchinkhuu E, Kon J, Kimura T, Tobo M, Yamazaki Y, Watanabe T, Yagi M, Sato M, Suzuki R, Murooka H, Sakai T, Nishitoba T, Im DS, Nochi H, Tamoto K, Tomura H, Okajima F. Ki16425, a subtype-selective antagonist for EDG-family lysophosphatidic acid receptors. Mol Pharmacol 2003;64:994–1005.
[24]. Raghavendra V, Tanga F, DeLeo JA. Inhibition of microglial activation attenuates the development but not existing hypersensitivity in a rat model of neuropathy. J Pharmacol Exp Ther 2003;306:624–30.
[25]. Rivera C, Li H, Thomas-Crusells J, Lahtinen H, Viitanen T, Nanobashvili A, Kokaia Z, Airaksinen MS, Voipio J, Kaila K, Saarma M. BDNF-induced TrkB activation down-regulates the K+-Cl− cotransporter KCC2 and impairs neuronal Cl− extrusion. J Cell Biol 2002;159:747–52.
[26]. Sakai A, Saitow F, Maruyama M, Miyake N, Miyake K, Shimada T, Okada T, Suzuki H. MicroRNA cluster miR-17-92 regulates multiple functionally related voltage-gated potassium channels in chronic neuropathic pain. Nat Commun 2017;8:16079.
[27]. Sorge RE, Mapplebeck JC, Rosen S, Beggs S, Taves S, Alexander JK, Martin LJ, Austin JS, Sotocinal SG, Chen D, Yang M, Shi XQ, Huang H, Pillon NJ, Bilan PJ, Tu Y, Klip A, Ji RR, Zhang J, Salter MW, Mogil JS. Different immune cells mediate mechanical pain hypersensitivity in male and female mice. Nat Neurosci 2015;18:1081–3.
[28]. Suga M, Tachikawa S, Tateyama D, Ohnuma K, Furue MK. Imaging-cytometry revealed spatial heterogeneities of marker expression in undifferentiated human pluripotent stem cells. In Vitro Cell Dev Biol Anim 2017;53:83–91.
[29]. Uchida H, Ma L, Ueda H. Epigenetic gene silencing underlies C-fiber dysfunctions in neuropathic pain. J Neurosci 2010;30:4806–14.
[30]. Uchida H, Matsumoto M, Ueda H. Profiling of BoNT/C3-reversible gene expression induced by lysophosphatidic acid: ephrinB1 gene up-regulation underlying neuropathic hyperalgesia and allodynia. Neurochem Int 2009;54:215–21.
[31]. Uchida H, Nagai J, Ueda H. Lysophosphatidic acid and its receptors LPA1 and LPA3 mediate paclitaxel-induced neuropathic pain in mice. Mol Pain 2014;10:71.
[32]. Ueda H. Molecular mechanisms of neuropathic pain-phenotypic switch and initiation mechanisms. Pharmacol Ther 2006;109:57–77.
[33]. Ueda H. Peripheral mechanisms of neuropathic pain—involvement of lysophosphatidic acid receptor-mediated demyelination. Mol Pain 2008;4:11.
[34]. Ueda H. Lysophosphatidic acid signaling is the definitive mechanism underlying neuropathic pain. PAIN 2017;158(suppl 1):S55–65.
[35]. Ueda H, Matsunaga H, Olaposi OI, Nagai J. Lysophosphatidic acid: chemical signature of neuropathic pain. Biochim Biophys Acta 2013;1831:61–73.
[36]. Ueda H, Neyama H. LPA1 receptor involvement in fibromyalgia-like pain induced by intermittent psychological stress, empathy. Neurobiol Pain 2017;1:16–25.
[37]. Ueda H, Uchida H. Epigenetic modification in neuropathic pain. Curr Pharm Des 2015;21:849–67.
[38]. Velasco M, O'Sullivan C, Sheridan GK. Lysophosphatidic acid receptors (LPARs): potential targets for the treatment of neuropathic pain. Neuropharmacology 2017;113:608–17.
[39]. Wang W, Wang W, Mei X, Huang J, Wei Y, Wang Y, Wu S, Li Y. Crosstalk between spinal astrocytes and neurons in nerve injury-induced neuropathic pain. PLoS One 2009;4:e6973.
[40]. Xie W, Matsumoto M, Chun J, Ueda H. Involvement of LPA1 receptor signaling in the reorganization of spinal input through Abeta-fibers in mice with partial sciatic nerve injury. Mol Pain 2008;4:46.
[41]. Yano R, Ma L, Nagai J, Ueda H. Interleukin-1beta plays key roles in LPA-induced amplification of LPA production in neuropathic pain model. Cell Mol Neurobiol 2013;33:1033–41.
[42]. Yao Y, Echeverry S, Shi XQ, Yang M, Yang QZ, Wang GY, Chambon J, Wu YC, Fu KY, De Koninck Y, Zhang J. Dynamics of spinal microglia repopulation following an acute depletion. Sci Rep 2016;6:22839.
[43]. Yung YC, Stoddard NC, Mirendil H, Chun J. Lysophosphatidic acid signaling in the nervous system. Neuron 2015;85:669–82.
[44]. Zhang J, Shi XQ, Echeverry S, Mogil JS, De Koninck Y, Rivest S. Expression of CCR2 in both resident and bone marrow-derived microglia plays a critical role in neuropathic pain. J Neurosci 2007;27:12396–406.
[45]. Zhang ZJ, Cao DL, Zhang X, Ji RR, Gao YJ. Chemokine contribution to neuropathic pain: respective induction of CXCL1 and CXCR2 in spinal cord astrocytes and neurons. PAIN 2013;154:2185–97.
Keywords:

Neuropathic pain; L-α-aminoadipate; Mac-1-saporin; Chemokine; Astrocytes; Microglia

© 2018 International Association for the Study of Pain