TLR4 Inhibits Spinal Gabaergic Activities via Microglial Activation in Chronic Constriction Injury Mice

*Corresponding Author:

Weifeng Yu and Feixiang Wu
Department of Anesthesiology & Intensive Care
Shanghai Eastern Hepatobiliary Surgery Hospital
Second Military Medical University, Shanghai, 200438, China

Abstract

Neuropathic pain (NP) remains a significant clinical problem worldwide. Toll-like rec e ptor  4 (TLR4) expressed on microglia in the spinal cord exert great influence on NP induced hyperalgesia. Impaired γ-aminobutyric acid (GABAergic) neuron activities in the spinal cordare also confirmed to be associated with NP progression. However, the role of TLR4 inducedmicroglial activation on impaired spinal GABAergic neuron activities inneuropathic painconditions remains unclear. Mechanical thr esholds were measured after intrathecal injectionof TLR4 inhibitor (TAK-242) or GABAA receptor agonist (muscimol) in male C57 mice. Microglialactivation, glutamic acid decarboxylase 65 (GAD65), as well as GABA concentrations incerebrospinal fluid (CSF) were further analyzed. Spontaneous inhibitory postsynaptic current(sIPSC) of GABAergic neurons were recorded using electrophysiological test. TLR4 expressionlevel was measured after chronic constriction injury (CCI) surgery and after intrathecaladministration of a TLR4 agonist (lipopolysaccharide, LPS) in naive mice.CCI surgery and TLR4agonist LPS treatment both induced hyperalgesia, while TLR4 inhibitor TAK-242 or GABAAreceptor agonist muscimol increased the mechanical thresholds in both CCI surgery andLPS-treated mice. Furthermore, decreased GAD65 fluorescence, GABA concentrations in CSFand sIPSC in CCI and LPS mice were a lleviated by TAK-242. Moreover, TAK-242 and microg lia  activa tion inhibitor (minocycline) suppressed the enhanced microglial activation in CCI surgeryand LPS-treated mice, and GAD65 fluorescence and GABA concentrations were increased byminocycline intervention. Additionally, TLR4 expression levels were up-regulated followingCCI surgery and intrathecal administration of LPS. TLR4 suppresses spinal GABAergic neuronactivities viamicroglial activation.

Keywords

TLR4, Microglial activation, GABA, Neuropathic pain

Abbreviations

NP: Neuropathic Pain; Tlr4: Toll-Like Receptor 4; Gaba: Γ-Aminobutyric Acid; Lps: Lipopolysaccharide; Tak-242: Tlr4 Inhibitor; Minocycline: Microglia Activation Inhibitor; Muscimol: Gabaa Receptor Agonist; Gad65: Glutamic Acid Decarboxylase 65; Sipsc: Spontaneous Inhibitory Postsynaptic Current; Csf: Cerebrospinal Fluid; Acsf: Artificial Cerebrospinal Fluid; Cci: Chronic Constriction Injury; I.t.: Intrathecal Injection.

Introduction

Neuropathic pain (NP), defined as the pain arising as a direct consequence of a lesion or disease affecting the somatosensory system, severely reduces the life quality of patients [1,2]. Patients with NP usually experience hyperalgesia and allodynia, and less satisfaction with routine drugs due to insufficient analgesia or severe side effects [3]. The mechanism underlying the initiation and progression of NP need to be further explored.

Over recent years, NP has been considered closely related to the imbalance of excitatory and inhibitory functions in the spinal cord caused by infection, inflammation or injury of the pain regulation signaling in the central nervous system [4]. GABA (γ-aminobutyric acid) is the main inhibitory transmitter in the spinal cord. Loss of GABAergic neurons and impairments in GABAergic inhibition contributed to NP [1,5- 7]. The concentration of GABA in the central nervous system is closely related to pain sensation [8]. GABA is synthesized from glutamate by glutamic acid decarboxylase (GAD). Knockout GAD65 in mice impaired GABA function and induced hyperalgesia [9]. Functional loss of GAD65 in the superficial spinal dorsal horn contributed to the development and maintenance of NP [10].

Toll-like receptor 4 (TLR4), a transmembrane receptor protein with extracellular leucine-rich repeated domains and a cytoplasmic signaling domain, is involved in innate immune responses. In vivo and in vitro studies have shown that TLR4 is exclusively expressed by microglia in the spinal cord, and a key receptor in the initiation of microglia activation and neuroinflammation in several neurodegenerative and central trauma diseases [11-13]. TLR4 knockout mice showed relieved hyperalgesia and allodynia after peripheral nerve injury, as well as the decreased release of pain related cytokines [14]. Moreover, treatment with TLR4 antagonist in NP mice resulted in a significant alleviation of hyperalgesia and allodynia, revealing involvement of TLR4 in NP. Consistently, Jurga AM demonstrated that TLR4 in the spinal microglia was essential in the progress of NP. LPS (Lipopolysaccharide), a major component of gram-negative bacteria, TLR4 agonist in the spinal cord [15]. Moreover, activation of TLR4 by intrathecal injection of LPS induced hyperalgesia [16]. Recently TLR4 was proved to attenuate GABA synthesis and postsynaptic GABA receptor activities in the naïve rat spinal dorsal horn [16]. Although mounting evidence has shown the important role of spinal TLR4 in nociceptive processing, whether activation of TLR4 in microglia could modulate spinal GABA activity in chronic constriction injury mice is not explored.

Activation of spinal microglia is a common feature in nerve injury-induced neuropathic pain. Peripheral nerve injury in mice leads to activation of microglia in the spinal dorsal horn, as indicated by morphological changes of microglia and increased expression of microglial markers such as IBA1 or CD-11b [3,17,18]. Minocycline, a semisynthetic second-generation tetracycline, is used as potent inhibitor of microglial activation without exerting direct action on neurons and astrocytes [19] Furthermore, minocycline has a superior blood-brain barrier penetration into the central nervous system [17]. Activation of TLR4 in spinal migroglia is critical for the genesis of pathological pain conditions induced by nerve injury and peripheral tissue inflammations [16,20]. Even though TLR4- dependent microglia activation is pivotal for the maintenance of NP, the way TLR4 acts on the GABAergic pathway in the process of NP is not explored. Therefore, in the present study we used in vitro and in vivo approaches to study the role of TLR4 in the modulation of spinal GABAergic neuron activities in CCI induced NP in mice.,

Materials and Methods

▪ Animals and drugs

Adult male C57 mice were provided from Shanghai Experimental Animal Center, Chinese Academy of Sciences (CAS). Mice (weight: 20-30 g, age: 10-12 weeks) were housed in a pathogen free environment (24°C room temperature and 50% humidity) under a 12/12 hour light/dark cycle, with food and water ad libitum. Maximal effort was made to minimize the number of animals used for this study, and to reduce the animal suffering. Drugs used in the experiment were administrated by intrathecal injection of 10 μL saline mixed with: LPS (Sigma, St. Louis, MO, USA) 30 μg, TAK-242 (a TLR4 inhibitor) (Sigma-Aldrich, St. Louis, MO, USA) 100 μg, minocycline (a potent inhibitor of microglial activation) (Sigma, St. Louis, MO, USA) 5 or 50 μg, muscimol (a GABAA receptor agonist) (Sigma, St. Louis, MO, USA) 0.1 μg.

This study was performed in accordance with the International Association for the Study of Pain guidelines. In addition, all animal studies (including the mice euthanasia procedure) were done in compliance with the regulations and guidelines of Second Military Medical University institutional animal care and conducted according to the AAALAC and the IACUC guidelines.

▪ Grouping

Following chronic constriction injury (CCI) surgery and intrathecal injection of LPS, mice were divided into CCI and LPS groups. In order to further test the changes in mechanical thresholds and GABAergic neuron activities, each animal group was divided into 4 additional subgroups and were intrathecally treated with saline (CCI+saline or LPS+saline groups), TAK- 242 (CCI+TAK-242 or LPS+TAK-242 groups), muscimol (CCI+muscimol or LPS+muscimol groups) and minocycline (CCI+minocycline or LPS+minocycline groups), separately.

▪ Intrathecal catheter implantation

Polyethylene catheter PE-10 (Becton Dickinson and Company, USA) was implanted 4 days before experiments. Under anesthesia, mice were placed on the operating table in the prone position. A 0.8-1.0 cm dorsal midline incision centered on the L4-L5 vertebral segment was then made for lumbar intrathecal catheterization. Next, a 3mm hole was drilled in the lateral aspect of the vertebral body to expose the Dura mater. Consequently, the sterilized catheter was inserted into the intrathecal space and advanced rostrally 0.5-1 cm. Successful implantation was confirmed by observing tailflick reflex and cerebrospinal fluid (CSF) flow from the catheter tip. Then catheter was fixed tightly with the opposite end that led from the cervical region through a subcutaneous tunnel. All wounds were closed in multiple layers and 80 thousand units of penicillin were administrated intramuscularly. After surgery, rats showing severe motor weakness and non-responsiveness to lidocaine were excluded from the study [21].

▪ Preparation of CCI model

CCI mice were prepared by ligating the right sciatic nerve according to a previously described method [22]. After anesthesia, an incision was made in the skin on the lateral surface of the right thigh, then the muscle layers were separated carefully to expose the sciatic nerve at the mid-thigh level. Proximal to the sciatic nerve trifurcation, the nerve was freed of adhering muscle and four loose ligatures of 6-0 chromic gut (W812, Ethicon Inc., Somerville, NJ, USA) were tied 1mm apart. The knots were slowly tightened until a brief twitch in the hind limb was observed. After nerve ligation, the muscle and skin were closed with 6-0 silk. The same procedures were performed in the sham group except for the ligation.

▪ Mechanical thresholds measurement

Mechanical thresholds measurements were performed before and after CCI surgery/or intrathecal injection of LPS in a quiet room between 9:00AM and 11:00AM, using the von Frey filaments [23]. In CCI mice, withdrawal thresholds of the ipsilateral hind paw were tested 0.5, 1, 2 h after drugs intrathecal injection on day 4 post-CCI surgery. On the other hand, LPS mice were pre-treated with TAK-242 and muscimol for one hour before LPS injection, and withdrawal thresholds were performed 1, 3, 5 h post-LPS injection. Testing was carried out in a set of plexiglas cages (6.5 × 7.5 × 8.0 cm) with a wire mesh floor. Animals were allowed to adapt for 60 min before testing. After acclimatization, the von Frey filaments were applied in an order of increasing stiffness through the wire mesh floor to the mid-plantar surface of the right hind paw until the filaments bent slightly. A positive withdrawal was scored when the animal showed a response (brisk withdrawal). The animals were tested using 13 von Frey monofilaments (North Coast Medical Inc., CA, USA) with logarithmically incremental stiffness (0.008, 0.04, 0.07, 0.16, 0.4, 0.6, 1.0, 1.4, 2.0, 4.0, 6.0, 8.0 and 10.0 g). The maximal value obtained during the procedure was recorded as mechanical threshold. The assays were repeated three times; a 5 minute interval was essential between two tests to avoid sensitization.

▪ Analysis of GABA by high-performance liquid chromatography

GABA content in the CSF were analyzed in CCI and LPS groups with or without TAK- 242 and minocycline intervention, using highperformance liquid chromatography with fluorescence detection after derivatization with o-phthaldialdehyde (Sigma-Aldrich, St. Louis, MO, USA). CSF was extracted through the foramen magnum and was double diluted. The o-phthaldialdehyde derivatizing reagent was prepared by dissolving o-phthaldialdehyde (54 mg) in absolute methanol (1ml) and adding 2-mercaptoethanol (40 μl) with sodium carbonate (0.1 M, pH 9.5, 99 ml). 10 ml of derivatizing reagent was then mixed with 20 μl of CSF and allowed to react for 2.5 min. Consequently, 20 ml of this solution was used for further analyzing. The chromatographic conditions were as follows: column, Eicompak (SC-5ODS 2.1 × 150 mm; Eicom); mobile phase, 0.1 M sodium dihydrogenphosphate and 0.1 M disodium hydrogenphosphate (pH 3.5) containing 50.0% methanol and 0.1 mM disodium EDTA; flow rate, 0.23 ml/min. Detector temperature was set at 30°C. Retention time for GABA was 14.83 min. The detection limit for GABA analysis was 0.2 pg/20 μl [24].

▪ Immunofluorescence

Mice were euthanized and perfused with sterile saline and 4% paraformaldehyde in 0.1 M phosphate buffer. L4-5 spinal lumbar of mice was quickly removed and post-fixed in 4% paraformaldehyde for 1 day, and then moved into the 30% sucrose solution for dehydration. Next, tissue transverse frozen sections (20 μm thick) were cut using freezing microtome (Leica, Germany) and were incubated with 5% goat serum for 2 h at room temperature. Samples were then incubated with primary antibodies diluent (contain 0.3% Triton X-100) overnight at 4℃. Before incubation with secondary antibodies, the sections were washed using 0.1M phosphate buffer. The antibodies we used were listed as follows: rabbit IBA-1 antibody (1:500; Wako, Japan), mouse GAD65 antibody (1:200; abcam, America), anti-rabbit IgG (1:1000; Invitrogen, America), and anti-mouse IgG (1:1000; Invitrogen, America). After incubation with secondary antibodies for 2 hours, digital images were captured with a high-resolution CCD Spot camera (fluorescent microscope) and then merged by Adobe Photoshop software.

▪ Electrophysiological test

The lumbar enlargement of spinal cord was quickly removed after anesthesia and transferred into cold artificial cerebrospinal fluid (ACSF). The ionic composition of the ACSF was (in mM): NaCl 117，KCl 3.6，CaCl₂ 2.5, MgCl₂ 1.2, NaH₂PO₄ 1.2, NaHCO₃ 25 and Glucose 11 with 95% O₂ and 5% CO₂ saturated. After removing endorhachis and cleaning the nerve root under the microscope, the spinal cord was cut transversely in the agar groove with a 400 µm thickness for whole-cell recording using a vibratome at a 0.14 mm/s speed and 3500 rpm vibration frequency. The slices were incubated for 1h in a Gibb groove full of ACSF with 95% O₂ and 5% CO₂ saturated at a constant temperature of 33℃. The incubated slices were transferred to the recording chamber and perfused with ACSF at room temperature. Substantia gelatinosa neurons, which are semitransparent, were identified and then visualized using infrared differential interference contrast microscopy. The patch electrode, filled with pipette solution, whose composition was (in mM): Cs₂SO₄ 110, CaCl₂ 0.5, MgCl₂ 2, EGTA 5, HEPES 5, TEA 5, ATP-Mg 5, was advanced onto the surface of the neuron until an increase in electrode resistance appeared, and then sucked the cytomembrane. A well-sealed neuron was confirmed based on action potential induced by a current input generated from pClamp10.2 software (Axon Instruments Inc, USA). Then, spontaneous inhibitory postsynaptic current (sIPSC) was recorded in a voltage-clamp mode using MultiClamp 700B amplifier and pClamp10.2 data acquisition software. All drugs were administered through the Drug Delivery System, and a 2-minute perfusion of ACSF was essential before and after the LPS perfusion [25].

▪ Western blot analysis

L4-5 segments of mice were quickly removed after anesthesia and snap frozen in liquid nitrogen and stored at -80℃. Tissues were sonicated in ice-cold (4℃) RIPA lysis buffer containing protease inhibitor cocktail. The homogenates were split on a rotary device on ice for 1 hour and then centrifuged at 13,000 g for 15 min at 4°C. The supernatants were collected and protein concentrations were measured using a bicinchoninic acid (BCA) Protein Assay Kit (Pierce). Proteins were heated at 99°C for 10 min to denature, followed by separation using 10% SDS-PAGE and transfered onto a Polyvinylidene fluoride (PVDF) membrane (Bio-Rad, Hercules, CA, USA). The membrane was blocked using 5% bull serum albumin for 2 h at room temperature, followed by incubation with primary antibodies diluent at 4°C overnight. Before incubation with secondary antibodies, the sections were washed using TBST. The antibodies we used were listed as follows: mouse TLR4 antibody (1:500; Santa Cruz, America), mouse GAPDH antibody (1:2000; Proteintech, America), and anti-mouse IgG (1:5000; CST, America). The proteins were detected using enhanced chemiluminescence western detection reagents and densitometry analysis was performed using Image J software.

Statistical analysis

Quantitative data are presented as the mean±sem. Two way repeated-measures ANOVA was used to compare withdrawal thresholds among different groups and time points. Relative grey value of western blot and immunofluorescence were performed using one-way ANOVA followed by Dunnett t test. Electrophysiological results were analyzed using the Clampfit software (Axon Instruments Inc, USA). P<0.05 was considered significantly different.

Results

▪ TLR4 is involved in CCI induced neuropathic pain

To evaluate whether TLR4 was involved in NP, we firstly detected TLR4 expression levels in L4-5 lumbar of spinal cord in mice subjected to chronic constriction injury (CCI) or sham surgery. Western blotting experiment showed that TLR4 expression was increased in CCI group compared with sham group (Figure 1A,B). In addition, application of LPS (30 μg, i.t.), the agonist of TLR4, also increased TLR4 expression compared with vehical group. These data suggested that TLR4 signaling may play an important role in NP. Next, we used behavior tests to investigate whether TLR4 contribute to NP (Figure 2A,2B). Von frey test showed that mechanical allodynia was partially reversed in CCI mice after intrathecal injection of TLR4 inhibitor TAK 242 (100 μg, i.t.). Conversely, activation of TLR4 by LPS (30 μg, i.t.) induced mechanical allodynia in naïve mice, which was partially blocked by pretreatment of TAK- 242 for one hour (100 μg, i.t., Figure 2B). Collectively, these data indicate that TLR4 participated in CCI induced NP in mice.

Figure 1: Increased TLR4 protein levels in spinal dorsal horn in mice with CCI-induced neuropathic pain or LPS treatment.
A and B, GAPDH served as a loading control. TLR4 protein expression levels in CCI group or LPS treatment group were standardized with that in the sham group. TLR4 proteins increased by CCI surgery and LPS treatment compared with those in the sham mice. Data were represented as mean ± sem, n=12 in the Western blot experiment; *P<0.05 vs. sham group.

Figure 2: CCI surgery and LPS treatment both induce behavioral hypersensitivity.
A, CCI surgery induces significant mechanical allodynia in the ipsilateral hind paw, which is partially reversed by TLR4 antagonist TAK-242 and GABAA receptor agonist muscimol. Data were represented as mean ± sem, n=12; *P<0.05 vs. CCI group B, An intrathecal injection of TLR4 agonist LPS acutely induced mechanical pain in naive mice. Proadministration (intrathecal) of TAK-242 and muscimol before LPS completely prevents LPS induced mechanical pain. n = 12 *P<0.05 vs. LPS treatment.

▪ TLR4 regulate GABAergic neuron activities in NP mice

In neuropathic pain conditions, spinal GABAergic inhibitory control is significantly reduced and contributes to increased excitation and central sensitization of pain transmission [26]. We hypothezed that TLR4 modulate CCI induced NP via GABA in the spinal cord. GABAergic neuron activities were assessed by measuring GAD65 fluorescence intensity (marker of GABAergic neuron activities) and GABA concentrations in CSF. Briefly, a square located in Figure 3A indicated the typical position acquired in enlarged photomicrographs of the superficial dorsal horn. Weaker density of GAD65 immunoreactivity was detected on the ipsilateral spinal dorsal horn of CCI mice compared with contralateral side and ipsilateral side of sham mice. Interestingly, TAK-242 treatment (100 μg, i.t.) reversed the decrease in GAD65 fluorescence in the ipsilateral side of spinal cord rather than the contralateral side (Figure 3B). Figure 3B shows the relative fluorescence of GAD65 acquired in the spinal dorsal horn. GABA concentrations in CSF were significantly decreased in both CCI group and LPS group Figure 3C, 3D, which was blocked by TAK-242 intervention. Taken together, these results suggested that GABAergic neuron activities in NP mice were regulated by TLR4in the spinal dorsal horn

Figure 3: Expression of GAD65 in the spinal dorsal horn.
A and B, Immunohistochemistry reveals that the fluorescence intensity of GAD65-positive cells increases in spinal dorsal horn on day post CCI surgery, which is reversed by application of TAK-242. n = 12 *P<0.05 vs. Sham; #P<0.05 vs. CCI. C and D, CCI surgery reduces GABA concentration in cerebrospinal fluid, which is mimicked by LPS treatment and partially reversed by TAK-242. n = 12 *P<0.05 vs. Sham; #P<0.05 vs. CCI.

Electrophysiology tests were further performed to evaluate LPS modulation of GABAergic activity by recording Sipsc Figure 4 A1,A2 represented the currents before and during LPS perfusion, respectively. The frequency of sIPSC was significantly decreased during LPS perfusion, indicating that LPS suppressed the GABAergic currents. Moreover, we also observed that pretreatment with TLR4 antagonist TAK 242 reversed the LPS-induced decrease on the frequency of sIPSC (Figure 4B).

Figure 4: LPS modulates sIPSCs in the superficial dorsal horn neurons.
A and B, Patch-clamp recording of sIPSCs shows an increase in the frequency of sIPSCs after perfusion of LPS (A) which is blocked by pretreatment of TAK-242 (B). n = 12 *P<n = 12 *P<0.05 vs ACSF.

▪ TLR4 promotes microglia activation in the spinal dorsal horn.

How does TLR4 on the microglia modulate GABAnergic neurons activities in CCI induced NP? Given the important role of microglia in NP [27], we first studied whether TLR4 promote microglia activation in NP. Microglia activation in spinal dorsal horn was judged by density of microglial marker IBA-1 immunoreactivity after CCI surgery (Figure 5A) and TLR4 activation by LPS (Figure 5B) both enhanced microglia activation, as indicated by upregulation of IBA-1 expression, while TLR4 antagonist TAK-242 (100 μg, i.t.) and the microglia activation inhibitor minocycline (5-50 μg, i.t.) both reversed the microglia activation on the ipsilateral dorsal horn of CCI group (Figure 5A). The immunofluorescence intensity of IBA-1- positive cells was increased on the ipsilateral side of CCI group in a square of 1×105μm2 in size of the superficial dorsal horns, while the trend was partially reversed after TAK-242 or minocycline intervention. Moreover, LPS treatment also significantly enhanced IBA-1 expression and the immunofluorescence intensity of IBA-1- positive cells when compared with saline control group. Consistently, the LPS-induced microglia activation was largely attenuated by TAK-242 or minocycline intervention (Figure 5C, 5D). These data suggested that TLR4 promoted microglia activation in the spinal dorsal horn.

Figure 5: Microglial activation on the lumbar spinal cord following CCI surgery
A and B. Immunohistochemistry data showed obvious microglial activation, as indicated by intense IBA1 immunoreactivity following CCI surgery, which is reversed by application of TAK-242 or minocycline. n = 12 *P<0.05 vs. Sham; #P<0.05 vs. CCI. C and D, LPS treatment ((pre? Post? How long? Dose?) Also induces microglial activation, which is blocked after application of TAK-242 or minocycline as mentioned above. n = 12 *P<0.05 vs. Sham; #P<0.05 vs. CCI.

▪ TLR4 regulates GABAergic neuron activities through microglia activation in spinal cord of NP mice

Given the importance of microglia and GABAergic neurons activities in NP, we examined whether TLR4 activation on microglia could impair GABAergic neuron activities. A square located in Figure 6A indicates the typical position acquired in enlarged photomicrographs of the superficial spinal dorsal horn. Minocycline (5-50 μg, i.t.), the inhibitor of microglia, reversed the decrease in GAD65 fluorescence on the ipsilateral dorsal horn without any significant changes on the contralateral side in CCI mice Figure 6A, 6B. Similar result was found in GABA concentration test, as minocyclinetreated group reversed the decrease of GABA in CSF in CCI mice compared with sham mice Figure 6C. Combined with previous result that TLR4 contributed to microglial activation, we hypothesize that TLR4 impair GABAgergic neuron activities through microglial activation in CCI induced NP. Figure 6B showed the relative fluorescence of GAD65 acquired in the spinal dorsal horn; To test this hypothesis, we detected inhibition TLR4 by LPS treatment on GABA concentrations in CSF. and found that LPS treatment significantly decreased GABA concentrations in CSF of naïve mice, which was blocked by minocycline treatment Figure 6D. Taken together, these data indicated that TLR4 activation impaired GABAergic neuron activities through microglia activation.

Figure 6: The effect of minocycline treatment on the expression of GAD65 in the spinal dorsal horn.
A and B, Immunohistochemistry reveals that the fluorescence intensity of GAD65-positive cells increases in spinal dorsal horn post CCI surgery, which is reversed by application of minocycline. n = 12 *P<0.05 vs. Sham; #P<0.05 vs CCI. C and D, CCI surgery reduces GABA concentration in cerebrospinal fluid, which is mimicked by LPS treatment and partially reversed by minocycline. n = 12 *P<0.05 vs. Sham; #P<0.05 vs. CCI.

Discussion

The mechanism of how TLR4 regulates the inhibitory GABA pathway in NP is not well understood. The present study explored the role of TLR4 on spinal dorsal horn GABAergic activities in CCI induced NP in mice. CCI surgery and spinal TLR4 activation by LPS both induced mechanical pain, while application of TLR4 inhibitor TAK-242 increased the mechanical thresholds in both CCI and LPS LPStreated naïve mice, indicating the involvement of TLR4 in NP. Next, we confirmed the spinal GABA signaling play a role in NP. Increased mechanical thresholds were observed following administration of GABAA receptor agonist muscimol. In the spinal cord, the decreased GAD65 fluorescence and GABA concentrations in CSF caused by CCI surgery were reversed by TLR4 inhibitor TAK-242, which means the involvement of TLR4 in GABA modulation of CCI induced NP. Moreover, sIPSC of GABAergic neurons was suppressed by LPS perfusion, which was antagonized by TLR4 inhibitor TAK-242. These results indicate that TLR4 suppress the spinal GABAergic neuron activities in CCI induced NP.

As the main expresser of TLR4 in the spinal cord, microglial activation was enhanced following CCI surgery and LPS intrathecal injection. Minocycline, a well-recognized inhibitor of microglia [19], was used to investigate the role of microglia on GABAergic signaling in spinal cord. GAD65 fluorescence and GABA concentrations were increased by minocycline intervention, suggesting the involvement of microglia in spinal GABA signaling. Single application of TAK-242 or minocycline is sufficient to inhibit microglial activation in CCI and LPS -treated mice, which indicated that TLR4 suppressed GABAergic neuron activities via microglia activation in NP.

▪ TLR4 activation suppresses GABAergic neuron activities in the spinal dorsal horn in mice

Electrophysiological results implied that TLR4 activation by LPS significantly suppressed sIPSC of GABAergic neurons in the spinal cord, which antagonized by TLR4 inhibitor TAK- 242, indicating TLR4 may suppress GABAergic neuron activities. GABAergic synapses are the major inhibitory synapses in the central nervous system. GABA content from GABAergic interneurons can directly act on GABA receptors at presynaptic terminals reducing the release of excitatory neurotransmitters, which target postsynaptic GABA receptors in neurons to produce inhibitory postsynaptic currents [28]. In the present study, TLR4 activation induced mechanical pain, decreased GAD65 and GABA concentrations, which were reversed by TAK- 242. These findings revealed the modulation of TLR4 on GABAergic neuron activities in CCI induced NP.

▪ TLR4 on microglia mediates microglia activation

In vivo and in vitro studies have demonstrated that TLR4 is exclusively expressed by microglia in the central nervous system, and that it has an essential role during neuroinflammation in several neurodegenerative and CNS trauma diseases [29, 30]. However, during different types of neuronal damage, TLR4 activates microglia through various factors. For example, TLR4 induced microglia activation in hypoxic condition is transcription factor hypoxia inducible factor- 1α (HIF-1α) dependent; transcription factor NF-κB pathway may also mediate the release of inflammatory mediators induced by TLR4, thus contributing to microglia activation [31]. Moreover, LPS (exogenous ligand for TLR4) can induce microglia activation based on two pathways: MyD88 dependent and MyD88 independent. The former pathway results in the rapid activation of NF-κB, whereas the latter one mediates the activation of transcription factor interferon regulatory factor 3 (IRF-3) and delayed NF-κB activation. Then various genes encoding pro-inflammatory cytokines (TNF-a, IL-1, IL-2 and IL-6), cyclooxygenase-2, inducible nitric oxide synthase (iNOS) were initiated in microglia cells, which consequently promote the production of inflammatory mediators and the activation of microglia in nociceptive pathway [27]. This progress also contributes to the central sensitization and changes the synaptic plasticity. Furthermore, our data suggested that microglial activation is closely associated with TLR4 upregulation and activation induced by nerve injury and LPS treatment in the spinal cord. Activation of TLR4 by LPS induced mechanical pain in a dose-dependent manner, which confirmed the role of microglial TLR4 activation in pain regulation at the spinal level.

▪ Microglial activation supresses GABAergic neuron activities

Microglia is one of the most susceptible sensors of brain pathology which can be rapidly activated upon any detection of signs of nerve injury, infection and inflammation [32,33]. Upon activation, microglia alters functional and structural properties of Schwann cells and axons, leading to the release of neurogenetic cytokines, including interleukin-1β (IL-1β), IL-6, leukemia inhibitory factor, TNF-α and transforming growth factor-β. These proinflammatory cytokines exhibit pleiotropic effects on homeostasis of glia in the spinal cord [34].

GABA system mainly exerts the inhibitory function in the central nervous system including spinal cord. In our study the decline of GABA concentrations in CSF in spinal cord of NP model was observed, while GABAA receptor agonist muscimol increased pain thresholds in CCI group. Microglia activation may play an important role in GABA synthesis and the function of GABAergic neurons, resulting in the development and maintenance of NP. Liang [35] and Bak [36] reported that GABA synthesis through the glutamate-glutamine cycle might be a crucial mechanism in maintaining the GABA homeostasis in the spinal cord. Moreover, microglia activation could also promote the production of cytokine such as IL-1β, which could reduce GABAergic sIPSC evoked by GABA. Through activating protein kinase C (PKC), IL-1β can suppress GABA receptor activities at the postsynaptic site in neurons, while the glial glutamate transporter activities can be suppressed by IL-1β and PKC activation. The integrated effect leads to a deficiency of glutamine supply, which results in reduction of the glutamate-glutamine cycle-dependent GABA synthesis and GABA receptor activities [16]Thus, TLR4, microglia or IL-1β may be used as therapeutic targets to alleviate NP. Mounting evidence has proved that peripheral nerve injury can induce spinal microglia or astrocytes activation in several chronic NP models [37, 38]. Our study confirmed microglial activation in CCI induced NP. Various algetic factors such as proinflammatory cytokines that facilitate pain transmission are released by activated microglia, resulting in hyperalgesia and allodynia [39]. Therefore, more efforts need to be done to reveal the mechanism of the spinal GABA signaling modulation by TLR4 induced microglial activation.

Conclusion

The present study has demonstrated that TLR4 suppress GABAergic neuron activities in the spinal cord via microglial activation in CCI induced NP. The present study highlights the possibility of targeting TLR4 on microglia to alleviate NP.

References

1. Ming-Tatt L, Khalivulla SI, Akhtar MN, et al. Anti-hyperalgesic effect of a benzilidine-cyclohexanone analogue on a mouse model of chronic constriction injury-induced neuropathic pain: Participation of the kappa-opioid receptor and KATP. Pharmacol. Biochem. Behav 114(1), 58-63 (2013).
2. Treede RD, Jensen TS, Campbell JN, et al. Neuropathic pain: redefinition and a grading system for clinical and research purposes. Neurology 70(1), 1630-1635 (2008).
3. Yan YY, Li CY, Zhou L, et al. Research progress of mechanisms and drug therapy for neuropathic pain. Life. Sci 190(1), 68-77 (2017).
4. Medrano MC, Dhanasobhon D, Yalcin I, et al. Loss of inhibitory tone on spinal cord dorsal horn spontaneously and nonspontaneously active neurons in a mouse model of neuropathic pain. Pain 157(1), 1432-1442 (2016).
5. Moon HC, Park YS. Reduced GABAergic neuronal activity in zona incerta causes neuropathic pain in a rat sciatic nerve chronic constriction injury model. J. Pain. Res 10(1), 1125-1134 (2017).
6. Li K, Cai B, Li C, et al. Presynaptic inhibition of nociceptive neurotransmission by somatosensory neuron-secreted suppressors. Sci. China. Life. Sci (2017).
7. Du X, Hao H, Yang Y, et al. Local GABAergic signaling within sensory ganglia controls peripheral nociceptive transmission. J. Clin. Investig 127(1), 1741-1756 (2017).
8. Crosby ND, Weisshaar CL, Smith JR, et al. Burst and Tonic Spinal Cord Stimulation Differentially Activate GABAergic Mechanisms to Attenuate Pain in a Rat Model of Cervical Radiculopathy. IEEE Trans. Biomed. Eng 62(1), 1604-1613 (2015).
9. Mackie M, Hughes DI, Maxwell DJ, et al. Distribution and colocalisation of glutamate decarboxylase isoforms in the rat spinal cord. Neuroscience 119(1), 461-472 (2003).
10. Kami K, Taguchi Ms S, Tajima F, et al. Improvements in impaired GABA and GAD65/67 production in the spinal dorsal horn contribute to exercise-induced hypoalgesia in a mouse model of neuropathic pain. Mol. Pain 12 (2016).
11. Lv YN, Ou-Yang AJ, Fu LS. MicroRNA-27a Negatively Modulates the Inflammatory Response in Lipopolysaccharide-Stimulated Microglia by Targeting TLR4 and IRAK4. Cell. Mol. Neurobiol 37(1), 195-210 (2017).
12. Gaikwad S, Patel D, Agrawal-Rajput R. CD40 Negatively Regulates ATP-TLR4-Activated Inflammasome in Microglia. Cell. Mol. Neurobiol 37(1), 351-359 (2017).
13. Han B, Lu Y, Zhao H, et al. Electroacupuncture modulated the inflammatory reaction in MCAO rats via inhibiting the TLR4/NF-kappaB signaling pathway in microglia. Int. J. Clin. Exp. Pathol 8(1), 11199-11205 (2015).
14. Sorge RE, LaCroix-Fralish ML, Tuttle AH, et al. Spinal cord Toll-like receptor 4 mediates inflammatory and neuropathic hypersensitivity in male but not female mice. J. Neurosci 31(1),15450-15454 (2011).
15. Palsson-McDermott EM, Curtis AM, Goel G, et al. Pyruvate kinase M2 regulates Hif-1alpha activity and IL-1beta induction and is a critical determinant of the warburg effect in LPS-activated macrophages. Cell. Metab 21(1), 65-80 (2015).
16. Yan X, Jiang E, Weng HR. Activation of toll like receptor 4 attenuates GABA synthesis and postsynaptic GABA receptor activities in the spinal dorsal horn via releasing interleukin-1 beta. J. Neuroinflamm 12(1), 222 (2015).
17. 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 306(1), 624-630 (2003).
18. Wen YR, Tan PH, Cheng JK, et al. Microglia: a promising target for treating neuropathic and postoperative pain, and morphine tolerance. J. Formos. Med. Assoc 110(1), 487-494 (2011).
19. Mika J. Modulation of microglia can attenuate neuropathic pain symptoms and enhance morphine effectiveness. Pharmacol. Rep 60(1), 297-307 (2008).
20. Lehnardt S, Lachance C, Patrizi S, et al. The toll-like receptor TLR4 is necessary for lipopolysaccharide-induced oligodendrocyte injury in the CNS. J. Neurosci 22(1), 2478-2486 (2002).
21. Njoo C, Heinl C, Kuner R. In vivo SiRNA transfection and gene knockdown in spinal cord via rapid noninvasive lumbar intrathecal injections in mice. J. Vis. Exp JoVE (2014).
22. Han HJ, Lee SW, Kim GT, et al. Enhanced Expression of TREK-1 Is Related with Chronic Constriction Injury of Neuropathic Pain Mouse Model in Dorsal Root Ganglion. Biomol. Ther 24(1), 252-259 (2016).
23. Guo JR, Wang H, Jin XJ, et al. Effect and mechanism of inhibition of PI3K/Akt/mTOR signal pathway on chronic neuropathic pain and spinal microglia in a rat model of chronic constriction injury. Oncotarget (2017).
24. Kawamata T, Omote K, Toriyabe M, et al. Intracerebroventricular morphine produces antinociception by evoking gamma-aminobutyric acid release through activation of 5-hydroxytryptamine 3 receptors in the spinal cord. Anesthesiology 96(1), 1175-1182 (2002).
25. Yamanaka M, Taniguchi W, Nishio N, et al. In vivo patch-clamp analysis of the antinociceptive actions of TRPA1 activation in the spinal dorsal horn. Mol. Pain 11(1), 20 (2015).
26. Malcangio M. GABAB receptors and pain. Neuropharmacology (2017)
27. Bettoni I, Comelli F, Rossini C, et al. Glial TLR4 receptor as new target to treat neuropathic pain: efficacy of a new receptor antagonist in a model of peripheral nerve injury in mice. Glia 56(1), 1312-1319 (2008).
28. Bardoni R, Takazawa T, Tong CK, et al. Pre- and postsynaptic inhibitory control in the spinal cord dorsal horn. Ann. N. Y. Acad. Sci 1279(1), 90-96 (2013).
29. Song M, Jin J, Lim JE, et al. TLR4 mutation reduces microglial activation, increases Abeta deposits and exacerbates cognitive deficits in a mouse model of Alzheimer's disease. J. Neuroinflamm 8(1), 92 (2011).
30. Fernandez-Lizarbe S, Pascual M, Guerri C. Critical role of TLR4 response in the activation of microglia induced by ethanol. J. Immunol 183(1), 4733-4744 (2009).
31. Yao L, Kan EM, Lu J, et al. Toll-like receptor 4 mediates microglial activation and production of inflammatory mediators in neonatal rat brain following hypoxia: role of TLR4 in hypoxic microglia. J. Neuroinflammation 10(1), 23 (2013).
32. Tsuda M, Beggs S, Salter MW, et al. Microglia and intractable chronic pain. Glia 61(1), 55-61 (2013).
33. Milligan ED, Watkins LR. Pathological and protective roles of glia in chronic pain. Nat. Rev Neurosci 10(1), 23-36 (2009).
34. Wang D, Couture R, Hong Y. Activated microglia in the spinal cord underlies diabetic neuropathic pain. Eur. J. Pharmacol 728(1), 59-66 (2014).
35. Liang SL, Carlson GC, Coulter DA. Dynamic regulation of synaptic GABA release by the glutamate-glutamine cycle in hippocampal area CA1. J. Neurosci 26(1), 8537-8548 (2006).
36. Bak LK, Schousboe A, Waagepetersen HS .The glutamate/GABA-glutamine cycle: aspects of transport, neurotransmitter homeostasis and ammonia transfer. J. Neurochem 98(1), 641-653 (2006).
37. Tsuda M, Inoue K, Salter MW. Neuropathic pain and spinal microglia: a big problem from molecules in "small" glia. Trends. Neurosci 28(1), 101-107 (2005).
38. Tsuda M, Koga K, Chen T, et al. Neuronal and microglial mechanisms for neuropathic pain in the spinal dorsal horn and anterior cingulate cortex. J. Neurochem 141(1), 486-498 (2017).
39. Sommer C. Painful neuropathies. Curr. Opin. Neurol 16(1), 623-628 (2003)