Spinal microglial β‐endorphin signaling mediates IL‐10 and exenatide‐induced inhibition of synaptic plasticity in neuropathic pain

Abstract Aim This study aimed to investigate the regulation of pain hypersensitivity induced by the spinal synaptic transmission mechanisms underlying interleukin (IL)‐10 and glucagon‐like peptide 1 receptor (GLP‐1R) agonist exenatide‐induced pain anti‐hypersensitivity in neuropathic rats through spinal nerve ligations. Methods Neuropathic pain model was established by spinal nerve ligation of L5/L6 and verified by electrophysiological recording and immunofluorescence staining. Microglial expression of β‐endorphin through autocrine IL‐10‐ and exenatide‐induced inhibition of glutamatergic transmission were performed by behavioral tests coupled with whole‐cell recording of miniature excitatory postsynaptic currents (mEPSCs) and miniature inhibitory postsynaptic currents (mIPSCs) through application of endogenous and exogenous IL‐10 and β‐endorphin. Results Intrathecal injections of IL‐10, exenatide, and the μ‐opioid receptor (MOR) agonists β‐endorphin and DAMGO inhibited thermal hyperalgesia and mechanical allodynia in neuropathic rats. Whole‐cell recordings of bath application of exenatide, IL‐10, and β‐endorphin showed similarly suppressed enhanced frequency and amplitude of the mEPSCs in the spinal dorsal horn neurons of laminae II, but did not reduce the frequency and amplitude of mIPSCs in neuropathic rats. The inhibitory effects of IL‐10 and exenatide on pain hypersensitive behaviors and spinal synaptic plasticity were totally blocked by pretreatment of IL‐10 antibody, β‐endorphin antiserum, and MOR antagonist CTAP. In addition, the microglial metabolic inhibitor minocycline blocked the inhibitory effects of IL‐10 and exenatide but not β‐endorphin on spinal synaptic plasticity. Conclusion This suggests that spinal microglial expression of β‐endorphin mediates IL‐10‐ and exenatide‐induced inhibition of glutamatergic transmission and pain hypersensitivity via presynaptic and postsynaptic MORs in spinal dorsal horn.


| INTRODUC TI ON
Neuropathic pain is a major chronic pain disorder that arises from peripheral nerve injury typically characterized by abnormalities in sensation or reactions to stimuli, associated with neuronal loss, glial inflammation, and maladaptive nociceptive circuits. 1 The rodent model of spinal nerve ligation (SNL) is popularly used to assay a variety of drugs for their therapeutic effects in neuropathic pain management and to determine the underlying cellular mechanisms.
The unique axonal branches of the dorsal root ganglia could easily detect and immediately respond to injurious stimuli by transmitting information from the periphery to the brain through the spinal cord. 2 Unlike neural circuits in the brain, spinal dorsal neurons, which are second-order neurons in the process of nociception, receive dominated glutamate from primary afferent fibers and the descending inhibitory pathway from the higher brain regions, contributed to mediating and influencing nociceptive transmission. 3 Nerve injury including rodent model of SNL may generate maladaptive dorsal horn plasticity and lead to pain states associated with alterations in N-methyl-D-aspartate (NMDA) receptor-mediated hypersensitivity, disinhibition of descending γ-aminobutyric acid (GABAergic) /glycinergic inhibitory neurotransmission, and activation of glial cells, especially microglia. 2,[4][5][6] The coordinative activity of microglia and neurons in brain diseases has been intensively studied, 7-9 while microglial-derived factors sensitize sensory processing through tumor necrosis factor (TNF)α, IL-6, IL-1β, and P2X inotropic receptors and leading to restore spinally mediated nocifensive reflexes mediated proinflammatory cytokines. [10][11][12][13] The importance of the interactions between microglia and the neural system has been increasingly recognized in the initiation and development of spinal plasticity in neuropathic pain, and targeting neuron-microglia interactions have exerted the potential to effectively treat neuropathic pain. 7,[14][15][16] Derived from microglia and other monocytes, IL-10 is probably the most prominent anti-inflammatory and immunosuppressive cytokine. 17,18 It also prevents damage to nerve injury and maintains neuronal homeostasis by modulating synaptic functions. 19 For example, IL-10 endowed anti-inflammatory responses in STAT3 generated proinflammatory conditions, and dampen pathogenic Th17 cell responses and colitis. 20 IL-10 deficiency has preserved synaptic integrity and alleviated cognitive impairment in Alzheimer's disease models. 21 Unconjugated bilirubin led to the overexpression of glutamate receptors and nitric oxide after neuronal damage accompanied by neurite outgrowth deficits. In primary neuronal cultures, IL-10 exhibited profound neuroprotection correlated with modulation of neuronal morphogenesis and neuritic arborization. 22 Our behavioral study indicated that stimulating exogenous and endogenous IL-10 expression have displayed marked pain anti-hypersensitive activity in animal models of neuropathic pain, peripheral diabetic pain, bone cancer pain, and complete Freund's adjuvant (CFA)-induced inflammatory pain. [23][24][25][26][27][28] We also demonstrated that IL-10's antineuropathic pain activity via the spinal microglia-derived expression of β-endorphin which depends on cAMP/PKA/p38β/CREB signaling separated from its spinal anti-neuroinflammatory activity. [29][30][31] In addition, recent study from our laboratory identified that activation of microglial expressed GLP-1 receptors, α7 nicotinic acetylcholine (α7-nACh) receptors, and G protein-coupled receptor 40 (GPR40) effectively relieved pain states in a variety of rodent models of chronic pain through stimulating spinal microglial autocrine expression of IL-10, IL-10 receptors, and subsequent expression of β-endorphin. 30,[32][33][34][35] Our data revealed that the microglial-derived IL- The aim of this study was to assess the effects and underlying mechanism of IL-10 and the GLP-1 receptor agonist exenatide on excitatory synaptic and descending inhibitory transmission in the spinal dorsal horn of neuropathic rats. We firstly characterized the L5/L6 spinal nerve ligated rat model of neuropathic pain by measuring long-lastingly pain hypersensitive behaviors, and performed whole-cell patch recordings of spinal excitatory synaptic and descending inhibitory neurotransmission in laminae Ⅱ neurons of spinal dorsal horn. We then tested the effects of IL-10, exenatide, and exogenous β-endorphin on pain hypersensitivity and excitatory and inhibitory synaptic transmissions of laminae Ⅱ neurons in neuropathic rats. Furthermore, the intervening agents such as agonists, antiserums, and antagonists were applied to determine whether β-endorphin expression mediated exenatide-and IL-10-induced inhibition of spinal excitatory synaptic transmission. Lastly, using the microglial metabolic inhibitor minocycline, it exerted fully blockade effects of IL-10-and exenatide-induced inhibition of the frequency and amplitude of mEPSCs, without affecting β-endorphin-induced inhibitory effects. Our results indicate that IL-10 and exenatide inhibit spinal synaptic plasticity through microglial expression of β-endorphin, which contributes to their inhibition of hypersensitivity activity in neuropathic pain.
Consequently, we further investigated the histological distribution of MOR in spinal dorsal horn through double immunofluorescent staining with presynaptic and postsynaptic markers. Together, these findings supported that IL-10/β-endorphin signaling regulates spinal excitatory synaptic transmission that abolish neuronal plasticity in neuropathic pain.

| Animals
To excluded the biorhythm cycle in female rats, male Wistar rats were used in this study. 38

| Spinal nerve ligation and behavioral testing
Spinal nerve ligation (SNL) was performed as described previously. 39,40 Briefly, rats were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg), and the left spinal nerves (L5/ L6) were carefully displayed and ligated tightly with 6-0 silk sutures distally. Wounds were sutured strictly with 4-0 silk sutures. After 14 days of recovery, the rats with significant hypersensitivity to mechanical and thermal stimuli in operated side (mechanical paw withdrawal thresholds <8 g) and without any motor impairments were selected for further study. The rats were randomly assigned to each group. Compared with the SNL group, sham rats were operated on without L5 and L6 nerve ligation.
Before mechanical threshold testing, rats were well handled by investigators for at least 4 days, 41 and were placed in a plastic box and acclimated individually for 30 min. A 2290 CE electrical von Frey hair (IITC Life Science), ranging from 0.1 to 90 g, was applied to measure the plantar surface of the rats' hind paws until they suddenly withdrew, while the rats stood on the grid. The lowest force was recorded as the threshold. The procedure was repeated 3 times with a 1-min interval. For thermal latency testing, rats were acclimated the atmosphere and transferred to an elevated plastic box on glass. Thermal threshold was determined using a 390G plantar test analgesic meter (IITC Life Science Inc.), the typical withdrawal behaviors were record as thermal threshold, and three withdrawal latency measurements were taken for each rat with a 5-min interval.
A 30-sec cutoff was confined to prevent tissue damage, and the data were averaged for each test. All experiments were performed by investigators who were blinded to the treatments.

| Intrathecal catheterization and injection in rats
A catheter with 0.28 nm inner diameter and 0.61 nm outer diameter (PE-10, Clay Adams) was placed in the rat lumbar spine under inhaled isoflurane anesthesia as previously described. 42 The rat lum-

| Immunofluorescence staining
The rats were anesthetized by intraperitoneal injection of pentobarbital sodium (50 mg/kg) and perfused with 50 ml ice-cold 0.9% NaCl solution followed by 50 ml 4% paraformaldehyde (w/v) in viewer was also used to generate merged-images in which colocalized areas appeared as yellow.

| Spinal slice preparation
Spinal slices obtained from the rats 14 days after surgery were used for electrophysiological assessments. All the rats were tested by mechanical threshold testing and signed with numbers before re- for 30 minutes at 32℃ and cooling to room temperature for one hour, and then transferred to the recording chamber. Data acquisition was conducted by using an Axonpatch 200 B amplifier (Axon Instruments), and data were filtered at 2 kHz and digitized at 5 kHz using pClamp10 software. 43

| Data evaluation and statistics
GraphPad Prism 7.0 (GraphPad Software) was used for data analysis.
The data were analyzed using two independent sample Student's t test, one-way or repeated-measures two-way ANOVA, followed by Sidak's post-tests for multiple comparisons. The results are represented as mean ± standard error of the mean (SEM). Before statistical analysis, the data were tested for Gaussian distribution assessed using Shapiro-Wilk normality test after transform to logarithms. Furthermore, the data failed to show Gaussian distribution has been assessed by nonparametric tests. p values <0.05 were considered statistically significant. All the data were processed by CorelDraw 2019.

| Phenotypic characterization of neuropathic pain
Peripheral nerve injury-induced neuropathic pain in animals is characterized by alterations in synaptic transmission followed with abnormalities in pain sensation or response to stimulus. 4,[45][46][47] This study firstly assessed the time courses of mechanical allodynia and thermal hyperalgesia in L5/L6 spinal nerve ligation induced neuropathic rats. Compared to sham rats, neuropathic rats exhibited time-dependent thermal hyperalgesia and mechanical allodynia to innocuous mechanical and radiant stimuli in the ipsilateral hind paws, with peak effects 9 days after surgery that were maintained for 14 days ( Figure 1A Figure 1D). Patch-clamp recordings showed that both frequency and amplitude of mEPSCs were significantly increased compared with sham group (Figure 1E,F, Frequency, t 13 = 3.860, p = 0.0020; Amplitude, t 13 = 8.124, p < 0.0001, two-tailed unpaired t test), which was in agreement with the previous finding of excessive glutamatergic synaptic transmission in pain hypersensitivity states. 49,50 Disinhibition is also known to contribute to pain hypersensitive states. 49,51 TTX, CNQX, and D-AP5 were used to block glutamate transmission and sodium channels; patch-clamp recordings  To examine the effects of IL-10 during pain states on inhibitory neurotransmission, whole-cell recording of mIPSCs was applied ( Figure 2F). However, bath application of IL-10 (100 ng/ml) did not alter the frequency or amplitude of mIPSCs in either the SNL rats or sham rats ( Figure 2G

| Reduced excitatory synaptic transmission mediated exenatide-induced pain antihypersensitivity.
The activation of GLP-1 receptors has been reported to inhibit thermal hyperalgesia and mechanical allodynia in neuropathic rats through spinal expression of IL-10 and subsequent β-endorphin. 23,29,31 To

| Exogenous β-endorphin ameliorated pain hypersensitivity and spinal synaptic excitatory transmission.
The MOR is a well-known target for pain relief. 55

| DISCUSS ION
Our study confirmed that SNL induced profound and long-lasting Disinhibition of inhibitory transmission closely related from descending inhibitory system were also arisen and also implicated the trans-synaptic effects induced by enhanced excitatory transmission. In behavioral tests, no matter the drugs applied, mechanical thresholds of neuropathic rats could not totally reverse, it might be inferred that those impairments induced neuronal organization and synaptic arrangements might be implicated in the behavioral observations.
In addition, our previous study manifested that dose-dependent injection of IL-10 and exenatide could reduce pain states in rats. It might be inferred that agonists, we applied in our study, have been extensively studied by using dosage-dependent application in behavioral and primary culture tests. 30,35 Furthermore, the antagonists were also critically identified by the minimum, and sufficient dosages applied in general. 32,33 Thus, the results are sufficient to support our results. Some studies have suggested gender difference in neuropathy of microglia, where nerve damage-induced neuroinflammatory cytokines expression and pain like behavers were showed to be dominant in male animals. In contrast, other findings have revealed that no sex dimorphism is found in microglial mediation of anti-hypersensitivity effects on neuropathic pain conditions such as gabapentin, IL-10, electroacupuncture, α7nAChR agonists, and/ or genetic deletion of the microglia selective molecules involving CX3CR1, P2Y12, and TMEN16F. 23,34,61,[69][70][71][72] It is clear that the strengthened neuronal connections between primary afferents and dorsal horn neurons have led to a state of hyperalgesia through sensory circuits in the spinal dorsal horn, accompanied by an imbalance between glutamatergic and GABAergic/glycinergic transmissions. 51 However, activation of microglia and interactions between microglia and neurons might shape the function and synaptic circuits in neuropathic pain. At the early stage of neuropathic pain, microglia-derived inflammatory cytokines, including IL-6, IL-1β, and TNFα, participate in synaptic plasticity by modulating the functions of synaptic receptors, channels, and enzymes such as voltage-gated Ca 2+ channels, the Src family of kinases, NMDA receptors, and metabotropic glutamate receptors, which eventually potentiated neuronal transmission of pain messages to higher neurons. 36,[73][74][75][76][77][78] Meanwhile, the CXCL12/CXCR4 signaling pathway also contributed to neuropathic pain through regulation of central sensitization. 79 Silencing of microglia-specific Ca 2+ -activated K + auxiliary β3 subunit significantly increases neurotransmission and attenuates antinociceptive tolerance. 80 Additionally, the regulation of microglia-specifically expressed P2Y12 reduces inflammation, pain sensation, and related glutamatergic transmission in spinal dorsal horn. 81 In contrast, our data provided evidence that activation of microglial IL-10 and GLP-1 receptors inhibited the frequencies and amplitudes of F I G U R E 6 Schematic diagram showing the role of microglial expression of β-endorphin in IL-10-and specific GLP-1 receptor agonist exenatide-induced inhibition of spinal excitatory synaptic transmission and pain hypersensitivity in neuropathic pain. Following activation of GLP-1 receptors, IL-10 is released and then activates IL-10 receptors via a microglial autocrine mechanism. Afterward, the β-endorphin is released to microglial neuronal synapses and activates neuronal presynaptic and postsynaptic μ-opioid receptors (MORs) to inhibit the enhanced glutamatergic transmission, leading to pain anti-hypersensitivity mEPSCs via the expression of β-endorphin, which was specifically and completely abolished by the microglial metabolic inhibitor minocycline, although minocycline did not affect β-endorphininhibited neurotransmission. In addition, aconitum-derived bulleyaconitine A was recently reported to stimulate microglia to release dynorphin A, which specifically activates presynaptic κ-opioid receptors in afferent neurons of substantia gelatinosa (SG) and inhibits spinal synaptic plasticity. 82 These results could provide evidence for interactions between microglia and neurons through endogenous peptides, which might primarily inhibit spinal synaptic plasticity and pain transmission and transduction.

| CON CLUS IONS
Our results have illustrated that treatment with IL-10 and exenatide produces pain anti-hypersensitivity in neuropathic rats and inhibits pre-and postsynaptic glutamatergic transmission in the spinal dorsal horn, which was blocked by minocycline, β-endorphin antiserum, and CTAP. Figure 6 schematically presents the proposed role of microglial expression of β-endorphin through autocrine IL-10and exenatide-induced inhibition of spinal synaptic plasticity and pain anti-hypersensitivity in neuropathic pain. Our results provide evidence that activation of microglial IL-10/β-endorphin signaling contributes to pain management and amelioration of maladaptive circuits in neuropathic pain.

ACK N OWLED G EM ENT
This work was financially, in part, supported by Yangzi River Pharmaceuticals Group.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no competing financial interests in this work.

E TH I C S A PPROVA L S TATE M E NT
All experiments were performed in accordance with the Animal Care and Welfare Committee of Shanghai Jiao Tong University and followed the animal care guidelines of the National Institutes of Health.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data, models, or figures generated or used during the study are available from the corresponding authors.