Toll-like receptor 2 plays an essential role in electroacupuncture analgesia in a mouse model of inflammatory pain

Abstract

Background:

Inflammatory pain occurs when local tissue injury activates macrophages and neutrophils, hence increasing pro-inflammatory cytokine and chemokine levels. Toll-like receptor 2 (TLR2) antagonism reportedly suppresses neuropathic and inflammatory pain.

Aims:

In the present study, we investigated the effect of electroacupuncture (EA) on TLR2 and related signalling molecules in a complete Freund’s adjuvant (CFA)-induced mouse model of inflammatory pain to determine whether EA can attenuate inflammatory pain via the TLR2 signalling pathway.

Methods:

EA significantly reduced mechanical and thermal hyperalgesia in the animal model. A similar effect was produced by TLR2 antagonism induced by CU-CPT22 injection.

Results:

TLR2 expression in the dorsal root ganglia, spinal cord and thalamus increased following induction of inflammation. Expression levels of downstream molecules such as pPI3K, pAkt and pmTOR also increased, as did those of MAPK subfamily members such as pERK, pp38 and pJNK. Transcription factors (pCREB and pNFκB) and nociceptive ion channels (Nav1.7 and Nav1.8) were also involved.

Conclusion:

Increased expression of the above molecules was attenuated by both EA and TLR2 antagonism. Our results show that EA attenuates inflammatory pain via TLR2 signalling.

Introduction

Inflammatory pain occurs as a response to trauma, injury, arthritis or tumours, and changes the properties of peripheral nerves through nerve fibre damage. ¹ This leads to an increased spontaneous firing of nerve fibres or alterations in their conduction properties. Pro-inflammatory mediators alter the transduction and conduction properties of neurons,^2,3 which leads to hypersensitisation in the chemical environment surrounding nerve fibres. ¹ Inflammatory mediators located in the extracellular environment activate numerous receptor classes, inducing downstream signalling cascades,^2,3 which in turn promote the translocation of protein kinases to the plasma membrane and phosphorylation of ion channels, thereby modulating the sensitivity of nociceptive neurons. ⁴ Complete Freund’s adjuvant (CFA) is known to trigger cell-mediated inflammatory pain, resulting in significant mechanical and thermal hyperalgesia.^5,6

Toll-like receptors (TLRs) are type I transmembrane signalling proteins that are expressed in the cells of the innate immune system. ⁷ Ten subtypes of TLRs have been identified, which present distinct ligand specificities.^8,9 TLR2 is expressed in neurons, microglia and astrocytes of the central nervous system (CNS),^10–12 and it exerts a crucial role in linking the innate immune system and the CNS, ¹³ and contributes to CNS autoimmune diseases and neuroinflammation. ¹⁴ TLR2 also regulates nerve injury-induced macrophage infiltration into the peripheral dorsal root ganglia (DRG), subsequently influencing pain sensitivity.^11,15 Jurga et al. showed that rat models of neuropathy present increased levels of TLR2 mRNA and/or protein levels, ¹⁶ whereas Domínguez-Punaro et al. reported that the transcriptional activation of the TLR2 gene was detected at 24 hours in the mouse thalamus. ¹⁷

Acupuncture, a therapeutic technique based on the principles of Traditional Chinese Medicine, has been widely used for over 3000 years. Electroacupuncture (EA) stimulates Aδ-fibres and modulates pain sensation by activating C-fibres. ¹⁸ EA can be utilised in the treatment of systemic diseases, ¹⁸ such as stroke-induced dementia, ¹⁹ epilepsy, ²⁰ weight control ²¹ and pain. ²² Several studies have reported that EA increases the release of endogenous opiates, ²³ serotonin ²⁴ and adenosine, ²⁵ which results in pain reduction, and that the anti-nociceptive role of acupuncture is mediated by various proteins, such as NMDARs, ASIC3, TRPV1, TRPV4 and Nav channels.^26–30 Furthermore, Wang et al. reported that EA decreases TLR2/4 after surgical trauma. ³¹ In this study, we aimed to investigate the effect of EA on inflammatory pain, by specifically focusing on TLR2 and related signalling molecules.

Methods

Animals

Serial experiments were conducted on C57/B6 male mice (ages 8 to 12 weeks) purchased from BioLASCO Co, Ltd, Taipei, Taiwan. Mice were randomly subdivided into four groups (n=8 per group): (1) healthy controls receiving saline injection (Normal group), (2) untreated CFA model (CFA group), (3) CFA model receiving 2 Hz EA (CFA+EA group), and (4) CFA model receiving TLR2 antagonist injection (CFA+CU-CPT22 group). In order to detect an effect size of 0.6 in withdrawal threshold at an α of 0.05 and 80% power it was estimated that eight animals per group would be required. After arrival, the mice were housed under a 12/12 hour light/dark cycle with ad libitum water and food. All procedures were approved by the Institute of Animal Care and Use Committee of China Medical University (permit no. 2016–061) and conducted in accordance with the Guide for the Use of Laboratory Animals of the National Research Council and the ethical guidelines of the International Association for the Study of Pain. The number of animals used and their suffering were minimised with 1% isoflurane. The laboratory workers were kept blind to treatment allocation during the experiments and analysis.

Inflammatory pain model and pharmacological injection

The mice were anaesthetised with 1% isoflurane and received a single injection of 20 µl saline (pH 7.4, buffered with 20 mM HEPES) or CFA (Sigma-F5881, 0.5 mg/mL heat-killed Mycobacterium tuberculosis (Sigma, St Louis, MO, USA)) in the plantar surface of the hind-paw to induce intraplantar inflammation. Behaviour tests were conducted 1 day after induction of inflammation. The TLR2 antagonist CU-CPT22 (Sigma-SML0722, 10 µM; Sigma-Aldrich, USA) was injected at ST36 immediately before CFA injection. A recent article reported that pre-incubation for 30 min with 10 µM CU-CPT22 for 24 hours significantly reduced the expression of TLR2 ³²

Electroacupuncture treatment

EA was applied using stainless steel needles (0.5 inch, 32 G, Yu Kuang, Taiwan) inserted into the muscle layer to a depth of 2–3 mm at bilateral ST36. EA was administered immediately after the CFA injection to determine if EA could prevent inflammatory pain. A stimulator (Trio 300, Ito, Japan) delivered 100 μs square pulses of 1 mA for 15 min at 2 Hz.

Measurement of mechanical and thermal hyperalgesia

Mechanical and thermal sensitivities were tested 1 day after intraplantar injection of CFA. All experiments were performed at room temperature (approximately 25°C) and the stimuli were applied only when the animals were calm but not sleeping or grooming. Mechanical sensitivity was measured by testing the force of responses to stimulation with three applications of electronic von Frey filaments (North Coast Medical, Gilroy, CA, USA). Thermal pain was measured with three applications using Hargraves’ test IITC analgesiometer (IITC Life Sciences, Woodland Hills, CA, USA).

Immunoblotting assay

Animals were anaesthetised with 2% isoflurane followed by cervical dislocation. DRG, spinal cord (SC) and thalamus were immediately excised on day 1 after the behavioural test to extract proteins. Total proteins were prepared by homogenising tissue in lysis buffer containing 50 mM Tris-HCl pH 7.4, 250 mM NaCl, 1% NP-40, 5 mM EDTA, 50 mM NaF, 1 mM Na₃VO₄, 0.02% NaN₃ and 1×protease inhibitor cocktail (AMRESCO). The extracted proteins (30 µg per sample assessed by BCA (bicinchoninic acid) protein assay) were subjected to 8% SDS-Tris glycine gel electrophoresis and transferred to a PVDF (polyvinylidene fluoride) membrane. The membrane was blocked with 5% non-fat milk in TBS-T buffer (10 mM Tris pH 7.5, 100 mM NaCl, 0.1% Tween 20), incubated with primary antibody (Alomone Labs Ltd) in TBS-T with 1% bovine serum albumin, and incubated for 1 hour at room temperature. Peroxidase-conjugated anti-rabbit antibody (1:5000) was used as a secondary antibody. The bands were visualised using an enhanced chemiluminescent substrate kit (Pierce) with LAS-3000 Fujifilm (Fuji Photo Film Co Ltd). Where applicable, the image intensities of specific bands were quantified with NIH ImageJ software (Bethesda, MD, USA).

Statistical analysis

All statistical data are presented as mean±SE. Statistical significance between normal, CFA, CFA+EA and CFA+CU-CPT22 groups were tested using analysis of variance (ANOVA), followed by a post-hoc Tukey’s test (P<0.05 was considered statistically significant).

Results

One day after CFA or saline injection, CFA-induced mice presented significant mechanical hyperalgesia (Figure 1A, mechanical force=1.75±0.17 g, n=8) and thermal hyperalgesia (Figure 1B, thermal latency=5.01±0.42 g, n=8), whereas these were not observed in the control mice (Figure 1A, mechanical force=3.40±0.11 g, n=8; Figure 1B, thermal latency=11.65±0.84 g, n=8). An attenuation of CFA-induced mechanical hyperalgesia was observed after 2 Hz EA (Figure 1A, 3.19±0.18 g, n=8). Furthermore, EA significantly reduced CFA-induced thermal hyperalgesia (Figure 1B, 11.22±0.99 g, n=8), and a similar effect was observed in TLR2 antagonist-injected mice (Figure 1B, 10.6±0.76 g, n=8). We further assessed whether thermal hyperalgesia was similarly affected. Injecting normal saline did not induce thermal hyperalgesia (Figure 1B, thermal latency=11.65±0.84 g, n=8), whereas injecting CFA did induce thermal hyperalgesia (Figure 1B, 5.01±0.42 g, n=8). Furthermore, EA significantly reduced CFA-induced thermal hyperalgesia (Figure 1B, 11.22±0.99 g, n=8). The same effect was obtained in the TLR2 antagonist-injected mice (Figure 1B, 10.6±0.76 g, n=8). The above results show that EA or TLR2 blockade attenuated mechanical and thermal hyperalgesia in a mouse inflammatory pain model.

Figure 1.

Mechanical and thermal withdrawal thresholds in the hind paws of mice following normal saline injection (Normal group, n=8), or CFA injections with no treatment (CFA group, n=8), after 1 day EA (CFA+EA group, n=8), or after 1 day TLR2 antagonist injection (CFA+CU-CPT22 group, n=8). *P<0.05 vs Normal group. #P<0.05 vs CFA group. CFA, complete Freund’s adjuvant; EA, electroacupuncture.

Western blot analysis results revealed expression of TLR2 in normal DRG (Figure 2A, 100.1±3.6%, n=6), and demonstrated that its expression level increased following CFA induction (Figure 2A, 143.6±4.8%, n=6). EA successfully attenuated TLR2 overexpression (Figure 2A, 95.9±17.5%, n=6), and a similar attenuation effect was observed after administration of the TLR2 antagonist CU-CPT22 (Figure 2A, 99.1±7.0%, n=6). Western blot analysis results also indicated an increase in pPI3K levels in CFA-injected mice (Figure 2A, 187.7±15.6%, n=6), which was reduced by both EA (Figure 2A, 134.4±8.3%, n=6) and CU-CPT22 administration (Figure 2A, 117.3±14.6%, n=6). A similar effect was observed for pAkt signalling (Figure 2A, CFA: 152.1±1.5%, EA: 110.0±16.2%, Cu-CPT22: 97.6±14.5%, n=6). Expression levels of the downstream molecule pmTOR were increased in CFA-injected mice (Figure 2A, 127.6±4.4%, n=6), and were reduced by both EA (Figure 2A, 96.5±14.1%, n=6) and CU-CPT22 treatment (Figure 2A, 93.4±7.7%, n=6). Results also revealed that the expression levels of pERK, pp38 and pJNK increased in the DRG of CFA-injected mice (Figure 2B, pERK: 156.1±4.9%, pp38: 193.4±36.8%, and pJNK: 131.5±13.3%, n=6), and that both EA and CU-CPT22 attenuated that increase (Figure 2B, 127.7±2.2%, n=6). Expression levels of the transcription factors pCREB and pNKκB also increased in CFA-induced mice (Figure 2C, pCREB: 204.1±12.3% vs pNKκB: 135.4±7.0% n=6), and were attenuated by both EA and CU-CPT22 administration (Figure 2C, n=6). Furthermore, expression levels of the nociceptive channels Nav1.7 and Nav1.8 were augmented in CFA-injected mice (Figure 2C, Nav1.7: 132.3±10.5% and Nav1.8: 130.2±5.0%, n=6), and were reduced by both EA and CU-CPT22 administration (Figure 2C, n=6).

Figure 2.

Expression levels of TLR2-associated signalling pathway proteins in the mouse dorsal root ganglia (DRG). (A) TLR2, pPI3K, pAKT and pmTOR, (B) pERK, pp38 and pJNK, and (C) pCREB, pNFκB, Nav1.7 and Nav1.8 expression levels in tissues from the Normal, CFA, CFA+EA and CFA+CU-CPT22 groups (from left to right). CFA=complete Freund’s adjuvant; EA= electroacupuncture; CU-CPT22= TLR2 antagonist. *P<0.05 vs Normal group. #P<0.05 vs CFA group. The Western blot bands at the top show the target protein. The lower bands are internal controls (β-actin or α-tubulin).

TLR2 expression was observed in the normal SC (Figure 3A, 100.0±1.8%, n=6), and was found to be increased by CFA induction (Figure 3A, 131.0±14.6%, n=6). EA significantly reduced TLR2 overexpression in the SC (Figure 3A, 93.6±12.6%, n=6), and a similar effect was observed in CU-CPT22-treated mice (Figure 3A, 95.5±6.4%, n=6). We next showed that pPI3K expression levels increased in CFA-treated mice (Figure 3A, 133.7±7.1%, n=6), and were reduced by both EA (Figure 3A, 117.3±15.8%, n=6) and CU-CPT22 administration (Figure 3A, 108.6±14.9%, n=6). A similar effect was observed for pAkt (Figure 3A, CFA: 137.5±10.9%; EA: 111.3±9.1%; Cu-CPT22: 95.0±9.3%, n=6) and pmTOR (Figure 3A, CFA: 139.6±7.7%; EA: 103.2±3.1%; Cu-CPT22: 105.0±5.2%, n=6). We further showed that pERK, pp38 and pJNK expression levels increased in the SC of CFA-injected mice (Figure 3B, pERK: 123.2±2.9%; pp38: 129.3±4.7%; pJNK 137.6±5.8%, n=6), and that the increase was attenuated by both EA and CU-CPT22 administration (Figure 3B, n=6). The expression levels of pCREB and pNKκB increased in CFA-injected mice (Figure 3C, pCREB: 131.1±10.2%; pNKκB: 126.4±9.8%, n=6), and were reduced by both EA and CU-CPT22 administration (Figure 3C, n=6). The expression levels of Nav1.7 and Nav1.8 were increased in CFA-induced mice (Figure 3C, Nav1.7: 138.2±4.4%; Nav1.8: 164.5±8.3%, n=6), and this increase was attenuated by both EA and CU-CPT22 administration (Figure 3C, n=6).

Figure 3.

Expression levels of TLR2-associated signalling pathway proteins in the mouse spinal cord (SC). (A) TLR2, pPI3K, pAKT and pmTOR, (B) pERK, pp38 and pJNK, and (C) pCREB, pNFκB, Nav1.7 and Nav1.8 expression levels in tissues from the Normal, CFA, CFA+EA and CFA+CU-CPT22 groups (from left to right). CFA=complete Freund’s adjuvant; EA=electroacupuncture; CU-CPT22=TLR2 antagonist. *P<0.05 vs Normal group. #P<0.05 vs CFA group. The Western blot bands at the top show the target protein. The lower bands are internal controls (β-actin or α-tubulin).

TLR2 expression was observed in the normal thalamus (Figure 4A, 99.9±1.2%, n=6), and its expression level increased following CFA induction (Figure 4A, 138.3±4.3%, n=6). TLR2 overexpression in the thalamus was significantly reduced by EA (Figure 4A, 98.9±3.5%, n=6), with a similar effect occurring in CU-CPT22-treated mice (Figure 4A, 96.7±4.3%, n=6). We further showed that pPI3K expression levels increased in CFA-treated mice (Figure 4A, 136.7±4.4%, n=6), and were reduced by both EA (Figure 4A, 112.3±4.3%, n=6) and CU-CPT22 administration (Figure 4A, 106.7±4.4%, n=6). A similar effect of both treatments was found in pAkt (Figure 3A, CFA: 139.7±4.6%; EA: 106.4±7.3%; Cu-CPT22: 98.7±6.4%, n=6) and pmTOR (Figure 3A, CFA: 169.6±8.3%; EA: 106.3±4.7%; Cu-CPT22: 102.9±4.9%, n=6). Increased expression levels of pERK, pp38 and pJNK were observed in the thalamus of CFA-injected mice (Figure 4B, pERK: 128.7±3.1%; pp38: 138.7±2.1%; pJNK 142.4±2.7%, n=6), and were attenuated by both EA and CU-CPT22 administration (Figure 4B, n=6). Expression levels of pCREB and pNKκB were increased in CFA-induced mice (Figure 4C, pCREB: 139.7±3.7%; pNKκB: 136.7±6.4%, n=6), and reduced by both EA and CU-CPT22 administration (Figure 4C, n=6). Expression levels of Nav1.7 and Nav1.8 were increased in CFA-injected mice (Figure 4C, Nav1.7: 140.4±3.7%; Nav1.8: 162.4±6.7%, n=6), and the increase was similarly attenuated by both EA and CU-CPT22 administration (Figure 4C, n=6).

Figure 4.

Expression levels of TLR2-associated signalling pathway proteins in the mouse thalamus. (A) TLR2, pPI3K, pAKT and pmTOR, (B) pERK, pp38 and pJNK, and (C) pCREB, pNFκB, Nav1.7 and Nav1.8 expression levels in tissues from the Normal, CFA, CFA+EA and CFA+CU-CPT22 groups (from left to right). CFA=complete Freund’s adjuvant; EA=electroacupuncture; CU-CPT22=TLR2 antagonist. *P<0.05 vs Normal group. #P<0.05 vs CFA group. The Western blot bands at the top show the target protein. The lower bands are internal controls (β-actin or α-tubulin).

Discussion

Inflammatory pain frequently follows peripheral nerve damage, and is caused by the infiltration of peripheral macrophages, neutrophils and other immune cells into the damaged site and adjacent DRG. ³² A simultaneous increase in the amount of pro-inflammatory cytokines and chemokines occurs in the peripheral DRG. ³³ TLR2 has previously been shown to play a crucial role in the DRG following peripheral nerve damage, ³⁴ and to be essential for the development of thermal hyperalgesia-related neuropathic pain. ³⁵ Kim et al. found that TLR2 knockout mice exhibited attenuation of the overexpression of nerve injury-induced pro-inflammatory genes, such as tumour necrosis factor α (TNFα), interleukin (IL)-1β and IL-6 in the SC. ³⁶ There is also evidence supporting the crucial role of TLR2 in inflammatory pain, wherein its activation by lipopolysaccharides significantly induces hyperalgesia and its antagonism alleviates hyperalgesia in animal models of neuropathic inflammatory pain. ¹¹

Several studies have suggested that acupuncture increases the release of endogenous opiates, ²³ serotonin ²⁴ and adenosine, ²⁵ hence attenuating nociception. Pre-treatment with EA can reduce postoperative analgesic requirements and adverse effects in patients undergoing lower abdominal surgery. ³⁷ Furthermore, intrathecal injection of μ opioid receptor antiserum or an antagonist in the cerebral ventricle attenuates 2 Hz EA-elicited analgesia. ³⁸

Zhang et al. report that TLR2 plays an essential role in controlling innate and inflammatory responses, and that morphine-induced activation of microglia and pro-inflammatory cytokines is inhibited by TLR2 deficiency. ³⁹ A recent study showed that TLR2 is crucial for the occurrence of neuropathy, through the activation of IBA-1/CD40-positive cells, and that TLR2 blockade produces significant analgesia and enhances the effectiveness of buprenorphine. ¹⁶ Wang et al. suggest that EA inhibits TLR2 and pro-inflammatory cytokines, hence producing an anti-inflammatory effect in a surgical trauma stress model. ⁴⁰ In this study, we investigated the effects of EA on TLR2-mediated inflammatory pain. Our results show that both EA and TLR2 antagonism can reduce inflammatory pain through TLR2 and related signalling pathways. Previous studies have shed light on the mechanisms underlying some of acupuncture’s beneficial effects. For example, Zhang et al. reported that CB1 receptors mediate the anti-inflammatory effects of EA in a rat model of migraine, ³¹ Xu et al. reported that the beneficial effects of moxibustion in rats with collagen-induced arthritis (CIA) are mediated by the periodxiredoxin I (PRDX1) and inositol 1,4,5-triphosphate receptor (IP3R) pathways, ⁴¹ and Zhao et al. suggested that EA increases the polarisation of M2 macrophages, which exerts a positive impact on rat models of spinal cord injury (SCI). ⁴² However, the mechanism underlying EA’s effect on inflammatory pain remains unknown.

The findings of this study suggest that EA can relieve inflammatory pain by downregulating the increased signalling of TLR2 pathway members, from the peripheral DRG to the CNS. Moreover, when applied at ST36, EA significantly reduced CFA-induced mechanical and thermal hyperalgesia. These results show that the TLR2 signalling pathway was significantly up-regulated in the inflamed DRG in the CFA mouse model. Furthermore, a similar result was produced by CU-CPT22 administration, which blocked TLR2. Another important finding of our study was that TLR2 and related molecules are altered in the SC and the thalamus of mice with CFA-induced inflammation, hence supporting their role in central sensitisation. The present study provides evidence that TLR2 and relevant molecules underlie the beneficial effects of EA in a mouse model of inflammatory pain (Figure 5), and hence support the clinical application of EA for the reduction of inflammatory pain. One limitation of the present study is the fact that only the TLR2 to pPI3K signalling pathway was investigated. Future studies are required to expand the present results by focusing on other mechanisms.

Figure 5.

Schematic illustration of possible mechanisms of TLR2 in complete Freund’s adjuvant (CFA)-induced inflammatory pain. We suggest that TLR2 acts as a receptor in inflammatory pain. Activation of TLR2 increases the expression of pPI3K, pAkt and pmTOR. Furthermore, pERK, pp38, pJNK, pNFκB and pCREB are also increased and nociceptive Nav1.7 and Nav1.8 are increased for pain conduction.