Electroacupuncture Attenuates CFA-induced Inflammatory Pain by suppressing Nav1.8 through S100B, TRPV1, Opioid, and Adenosine Pathways in Mice

Pain is associated with several conditions, such as inflammation, that result from altered peripheral nerve properties. Electroacupuncture (EA) is a common Chinese clinical medical technology used for pain management. Using an inflammatory pain mouse model, we investigated the effects of EA on the regulation of neurons, microglia, and related molecules. Complete Freund’s adjuvant (CFA) injections produced a significant mechanical and thermal hyperalgesia that was reversed by EA or a transient receptor potential V1 (TRPV1) gene deletion. The expression of the astrocytic marker glial fibrillary acidic protein (GFAP), the microglial marker Iba-1, S100B, receptor for advanced glycation end-products (RAGE), TRPV1, and other related molecules was dramatically increased in the dorsal root ganglion (DRG) and spinal cord dorsal horn (SCDH) of CFA-treated mice. This effect was reversed by EA and TRPV1 gene deletion. In addition, endomorphin (EM) and N6-cyclopentyladenosine (CPA) administration reliably reduced mechanical and thermal hyperalgesia, thereby suggesting the involvement of opioid and adenosine receptors. Furthermore, blocking of opioid and adenosine A1 receptors reversed the analgesic effects of EA. Our study illustrates the substantial therapeutic effects of EA against inflammatory pain and provides a novel and detailed mechanism underlying EA-mediated analgesia via neuronal and non-neuronal pathways.

Several channels, receptors, and signaling molecules within neurons and microglia are responsible for pain transmission. Secreted by astrocytes, S100-B is often implicated in the central nervous system (CNS) 4 . S100-B proteins then activate receptors for advanced glycation end-products (RAGE), which results in acute and chronic diseases 5 . RAGE activation initiates downstream inflammatory cellular responses 6 , and increased levels of RAGE have been reported in neurons and glia after brain injury 7 . The Nav sodium channels are involved in inflammation-induced hyperalgesia 8,9 . Sodium channel-induced currents that significantly influence the threshold for action potential firing have been identified in neurons of the CNS 9 and DRG 8 . Ion channel transient receptor potential vanilloid 1 (TRPV1) plays an important role in both nociceptive 10 and neuropathic pain 11 . TRPV1 is expressed in peripheral dorsal root ganglion (DRG), central spinal cord dorsal horn (SCDH), and brain. Centrally expressed TRPV1 is involved in the detection of thermal and mechanical pain 12 . The PI3K/AKT/ mTOR (mTORC1) signaling pathway is involved in cellular immunity 13 . In addition, the activation of TRPV1 increases the expression of PI3K, AKT, CREB, NF-κ B, Nav1.7, and Nav1.8. The increased expression of these molecules was attenuated in TRPV1 −/− mice 12 .
Acupuncture has been used for over 3,000 years in Asia to treat pain, and the analgesic efficacy of acupuncture is recognized worldwide. Over the past thirty years, studies have investigated the relationship between acupuncture and endogenous central opiates 14 . However, relatively recent studies showed that the antinociceptive effect of acupuncture may be related to changes in the expression of various ionotropic receptor channels and voltage-gated channels, including N-methyl-D-aspartate receptors (NMDARs), acid-sensing ion channel 3 (ASIC3), TRPV1, local adenosine, and Nav channels 12,[15][16][17][18] . Our previous studies demonstrated that EA results in antinociceptive effects and reduces mechanical and thermal hyperalgesia in an inflammatory mouse model via inhibition of TRPV1 and its related pathways 12 . However, the complete mechanism behind the effects of EA on neurons and microglia remains unclear. Thus, we assessed the expression of non-neuronal markers, including GFAP, Iba-1, S100B, and RAGE, and neuronal TRPV1-related molecules during inflammatory pain. This study provides new information on the relationships between EA, inflammatory pain, neurons, and microglia.

Material and Methods
Experimental Animals. All animals were treated in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals, and the study protocol was approved by the ethics committee of the China Medical University, Taichung, Taiwan (permit No. 2016-061). C57/B6 mice weighing approximately g and aged 8-12 weeks were purchased from the BioLASCO Animal Center, Taipei, Taiwan. Animals were housed in Plexiglas cages in a temperature-controlled room (25 ± 2 °C) with a relative humidity of 60 ± 5%, and were fed a diet of standard rat chow and water ad libitum. Approximately hours before the experiment, the rats were fasted but had free access to water.
Inflammatory Pain Model. Based on our previous studies 12 , a total of ten mice per group was the minimum number necessary for fully powered experiments. The mice were subdivided randomly into four groups of: (1) Control group: normal saline injection, (2) CFA group: CFA injection to induce inflammatory pain, (3) EA group: CFA injection and EA manipulation, and (4) TRPV1 −/− group: CFA injection to determine the role of TRPV1 in inflammatory pain. All experiments were performed in the laboratory during daylight hours. Two EA sessions were completed at 24 hours and 48 hours after CFA injection between 9:00 and 10:00 am. We used behavioral testing and the Hargreaves test to assess mechanical and thermal hyperalgesia at baseline, the moment after injection of CFA, and 24 and 48 hours after CFA injection. Two days after the CFA injection and EA sessions, we compared the pain reducing effects of EA and TRPV1 gene knockout. We analyzed pain-related molecules in the DRG and SCDH of mice using western blotting and immunohistochemical staining. Mice were anesthetized with 1% isoflurane and injected with 20 μ l of saline (pH 7.4, buffered with 20 mM HEPES) or CFA (complete Freund's adjuvant; 0.5 mg/ml heat-killed M. tuberculosis; Sigma, St. Louis, MO) in the plantar surface of the hind paw to induce intraplantar inflammation.
Electroacupuncture. EA was applied using stainless steel needles (0.5″ inch, 32 G, YU KUANG, Taiwan) that were inserted into the muscle layer to a depth of 2-3 mm at ST36 acupoint. EA was administered 1 day after the CFA injection every day at the same time (10:00-12:00 AM). A Trio-300 (Japan) stimulator delivered electrical square pulses for 15 min with a 100 μ s duration and a 2 Hz frequency. The stimulation amplitude was 1 mA.

Behavior Test (von Frey test and Hargraves' test).
Behavior tests were conducted at 1-2 day after induction of CFA injection. All stimuli were performed at room temperature (approximately 25 °C) and 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, SERIES8, Model 390 G).
Statistical Analysis. All statistic data are presented as the mean ± standard error. A p value < 0.05 was considered to represent statistical significance. The four groups in this study were: (1) Control group, (2) CFA group, (3) EA group, and (4) TRPV1 −/− group (n = 10/group). Statistical significance between groups was tested using the ANOVA test, followed by a post hoc Tukey's test (p < 0.05 was considered statistically significant). All statistical analyses were carried out using the statistical package SPSS for Windows (Version 21.0, SPSS, Chicago, Illinois, USA).
Opioid and adenosine A1 receptors are critical for electroacupuncture-mediated analgesia in a mouse model of inflammatory pain. We next determined how EA relieves inflammatory pain. EA at acupoint ST36 significantly reduced CFA-induced mechanical hyperalgesia (Fig. 8A, 2.9 ± 0.3 g, p < 0.05 compared with CFA inflammation group, n = 8). This effect was not observed in the sham control (Fig. 8A, 1.2 ± 0.3 g, p > 0.05 compared to CFA inflammation group, n = 8). An i.p. injection of the opioid-specific agonist EM partially reduced mechanical hyperalgesia (Fig. 8A, 2.4 ± 0.3 g, p < 0.05 compared with the CFA group, n = 8). An i.m. injection of the adenosine receptor agonist CPA into acupoint ST36 reduced mechanical hyperalgesia (Fig. 8A, 3.9 ± 0.3 g, p < 0.05 compared with the CFA group, n = 8). Similar results were observed for thermal hyperalgesia. EA significantly attenuated thermal hyperalgesia (Fig. 8B, 9.7 ± 0.2 s, p < 0.05 compared with the CFA group, n = 8), and this effect was not observed in the sham group (Fig. 8B, 6.3 ± 0.2 s, p > 0.05 compared with the CFA group, n = 8). The administration of EM (Fig. 8B, 9.1 ± 0.2 s, p < 0.05 compared with the CFA group, n = 8) and CPA (Fig. 8B, 9.6 ± 0.4 s, p < 0.05 compared with the CFA group, n = 8) significantly reduced thermal hyperalgesia in a mouse model of inflammatory pain.

Discussion
In our current study, we confirmed that a CFA injection successfully evokes inflammatory pain in mice. The mechanical and thermal hyperalgesia observed in the CFA-induced inflammatory pain model were attenuated on days 1 and 2 by EA stimulation at bilateral ST36 acupoints and TRPV1 gene deletion. We also showed that   astrocytes, microglia, S100B, and the TRPV1-related signaling pathway were essential for EA analgesia. EA reduced both mechanical and thermal hyperalgesia via the activation of opioid and adenosine A1 receptors and downregulation of Nav1.8 expression.
A recent study showed that transient hyperalgesia can be partially induced via the acid-sensing ion channel 3 pathway even with TRPV1 gene deletion 19 . However, TRPV1 gradually suppresses pain, which suggests that the TRPV1 pathway is essential but is not the only inflammatory pain pathway. S100B in the DRG and SCDH was increased on day 2 after CFA injection. S100B is mainly secreted by astrocytes 20,21 and stimulates RAGE on neurons and microglia, thus activating these cell types. Using the microglia/macrophage-specific calcium-binding protein Iba-1, we determined that the number of microglia in the DRG and SCDH was increased after CFA injection. Previous studies showed that activated microglia secrete IL-1 22 , IL-6 22 , and TNF-α 22,23 . Our data showed that TRPV1 12,24 , PI3K 12,25 , AKT 12,25 , mTOR, CREB 12,26 , NF-κ B, Nav1.7 2,12 , and Nav1.8 2,12 were increased after CFA injection. Furthermore, the number of Nav1.7 and Nav1.8 channels was increased in DRG neurons. Thus, the intraplantar injection of CFA successfully induced peripheral nociceptive neurons to generate trains of action potentials toward the DRG and SCDH. Next, the neurons and microglia cells of the spinal DRG and SCDH are activated, thus producing mechanical and thermal hyperalgesia. The increased expression of the aforementioned molecules within neurons and microglia was attenuated by EA and TRPV1 gene deletion. To the best of our knowledge, very few studies have explored the effects of EA on inflammation-activated microglia. Our data show that EA suppresses the activation of microglia in an inflammatory pain model. Zhang et al. reported that EA at a frequency of 10 Hz significantly reduced CFA-induced hind paw edema. Moreover, EA attenuated the inflammatory response via the hypothalamus-pituitary-adrenal axis (HPA) and nervous system 27 . Recently, EA was shown to suppress the inflammation-induced expression of neurokinin-1 in the SCDH of rats 28 . These phenomena were not observed in sham groups, which suggests acupoint-specific effects 27,28 . Numerous studies have investigated the role of different Nav channels in pain, neuron excitability, and action potential firing 29,30 . Nav1.7 was greatly expressed in C-fiber free nerve endings and played a crucial role in the propagation of nociceptive information 31 . Recent studies utilizing antisense and knockout mice strongly support the use of Navs as analgesic drugs 32,33 . Derivatives from benzazepinone and imidazopyridine were developed to block Nav1.7 channels for pain treatment 34 . Our results clearly indicate that EA reliably attenuated CFA-induced inflammatory pain by ameliorating Nav1.7 overexpression.
The chronic intrathecal administration of Nav1.8 antisense successfully attenuated Nav1.8-induced currents and decreased mechanical allodynia after intraplantar CFA injections 35 . Specific Nav channel blockers are a potentially useful treatment for inflammatory pain. A-803467, a novel and specific blocker of the Nav1.8 channel, ameliorated inflammatory pain in rats 36 . Thus, our investigation into inflammatory pain using ancient acupuncture mechanisms is highly valuable and our findings can be further applied to clinical medicine. EA and the administration of an adenosine A1 receptor agonist at ST36 attenuated pain behaviors, whereas endomorphin 1 administration at ST36 did not (data not shown). This finding suggests that endomorphin 1 may work in the CNS but not at acupoints, which is supported by a previous study 37 . We injected EM intraperitoneally and showed that Nav1.8 overexpression was reversed. A similar result was observed in the CPA (local injection) group. Activation Figure 10. Schematic illustration of neuronal and non-neuronal mechanisms in EA-mediated analgesia of CFA-induced inflammatory pain. Summary diagram of how astrocyte, glial cells, and TRPV1 is crucial for inflammatory pain and related mechanisms. Our results show that EA can reduce S100B release from nonneuronal cell. We also indicated that EA can trigger the release of opioid and adenosine receptors for relieving inflammatory pain through TRPV1 pathway.
Scientific RepoRts | 7:42531 | DOI: 10.1038/srep42531 of opioid receptors modulates peripheral inflammation in local tissues 38 and suppresses Ca 2+ and/or Na + currents in primary afferent neurons 39 . Stein and Kuchler reported that activating opioid receptors further activates the Gi protein to inhibit TRPV1 40 .
PKA 41 and PKCε 42 phosphorylate Nav1.8, increase Nav1.8 currents, lower the threshold voltage for activation, and produce a depolarizing shift. Inflammatory cytokines, such as interleukin-1β (IL-1β ) 43 and IL-6 44 , activate astrocytes and mediate neuroprotection in a PKCε -dependent manner 45,46 . Our immunostaining data showed that Nav1.8 was increased after CFA-induced inflammation. This effect was ameliorated by EA and the activation of opioid and adenosine receptors. Furthermore, opioid and adenosine antagonists significantly blocked the analgesic effects of EA. Kim et al. suggested that PI3K and Akt phosphorylation increases in response to carrageenan-induced inflammatory pain and is abolished by EA. Furthermore, the PI3K inhibitor significantly attenuates carrageenan-induced hyperalgesia 47 . Xu et al. reported that activation of pPI3K-Akt in the rat DRG and SC is crucial for neuropathic pain after spinal nerve ligation. An intrathecal injection of a PI3K or Akt inhibitor reduces neuropathic pain behavior. Immunohistochemistry indicated that an intrathecal injection of the PI3K inhibitor significantly inhibits activation of Akt in L5 of the DRG and SC 48 . Sun et al. showed that Akt, which is distributed mainly in small to medium DRG, increases in DRG 5 min after an intradermal injection of capsaicin (TRPV1 activator). TRPV1 is also double labeled with Akt. Behavioral tests suggested that an intradermal injection of the PI3K inhibitor was significantly inhibited by capsaicin-induced hyperalgesia. A similar result was observed with an Akt inhibitor 49 . Xu et al. reported that pretreatment with a PI3K inhibitor prevents activation of Akt in a dose-dependent manner in mice with plantar incision-induced pain. Inhibiting spinal PI3K further prevents pain behavior and c-Fos expression 50 . Kay et al. suggested that suppressing endogenous PI3K/Akt activity with the potent PI3K inhibitor LY294002 abolishes pCREB in mice with visceral inflammation-induced cystitis 51 . Our results showed that injecting the PI3K inhibitor significantly reduces inflammatory pain and pAkt-pmTOR protein level.

Conclusion
CFA induced mechanical and thermal hyperalgesia in mice. This effect was accompanied by the activation of astrocytes and glial cells, which then released the RAGE ligand S100B. Inflammation also activated TRPV1 and triggered a signaling pathway that activated nociceptive Nav channels. These phenomena were abolished by EA and deletion of the TRPV1 gene. We demonstrated that EA triggers the release of both endogenous opiates and adenosine to relieve inflammatory pain in mice. This study identified a potential TRPV1-mediated signaling mechanism (Fig. 10). Given the specific analgesic effect of EA on inflammatory hyperalgesia, TRPV1 antagonists could further translate into clinical efficacy for treating pain. Our findings offer insight for EA clinical trials in patients with inflammatory pain.