AICAR activates AMPK to regulate STAT3 nuclear translocation and phosphorylation and iNOS expression in inammatory pain

Background: AMP-activated protein kinase (AMPK) activators can improve inammatory pain and neuropathic pain. Inammation translocate signal transducers and activators of transcription 3 (STAT3) to the nuclei of activated macrophages, and STAT3 phosphorylation promotes the expression of inducible nitric oxide synthetase (iNOS). In this study, we determined whether AMPK activation alleviate inammatory pain via STAT3 nuclear translocation and phosphorylation. Methods: Immunoblotting was used to measure the expression of p-AMPK, and iNOS. Immunoblotting and immunouorescence were used to detect the nuclear translocation of p-STAT3(Ser727) and STAT3 in macrophages of local inammatory tissues. Flow cytometry was used to measure reactive oxygen species (ROS) accumulation and mitochondrial damage. Results: AMPK activation with AICAR signicantly alleviated pain hypersensitivity and inhibited the expression of iNOS in complete Freund's adjust (CFA)-induced inamed skin tissues. CFA caused nuclear translocation of STAT3 and p-STAT3(Ser727) in macrophages of inamed skin tissues. AICAR inhibited nuclear translocation of STAT3 and p-STAT3(Ser727) and promoted the phosphorylation of STAT3(Ser727) in the cytoplasm of macrophages. AICAR also inhibited the expression of iNOS and nuclear translocation of STAT3 and p-STAT3(Ser727), and promoted the phosphorylation of STAT3(Ser727) in NR8383 macrophages treated with CFA. AMPK activation also inhibited the ROS generation and the mitochondrial damage of NR8383 macrophages caused by CFA. In addition, transfection of STAT3 S727D decreased ROS and alleviated mitochondrial damage. Conclusions: Activation of AMPK attenuates inammatory pain and suppresses STAT3 nuclear translocation and phosphorylation of STAT3(Ser727) in macrophages, resulting in reduced iNOS. and phosphorylation. Our provides new information about the signaling mechanisms involved in the analgesic effect of AMPK activation on inammatory

translocation and phosphorylation of STAT3(Ser727) in macrophages, resulting in reduced iNOS. Activation of AMPK also promotes phosphorylation of STAT3(Ser727) in the cytoplasm of macrophages to alleviate ROS accumulation and mitochondrial damage associated with in ammation.

Background
In ammatory pain is a common symptom associated with many clinical conditions [1]. Tissues damage leads to the release of in ammatory mediators from damaged cells [2]. A variety of in ammatory mediators in the extracellular environment causes the sensitization of nociceptive neurons [3]. In formalininduced in ammatory pain model, inhibition of interleukin-1 β (IL-1β), tumor necrosis factor-α (TNF-α ) and inducible nitric oxide synthetase (iNOS) and other in ammatory mediators alleviates in ammatory pain [4]. In addition, activated macrophages release nitric oxide and reactive oxygen species (ROS), which play an important role in in ammation and pain [5,6].
AMP-activated protein kinase (AMPK) is a kinase that regulates energy homeostasis. Phosphorylation of Thr172 site is a marker of AMPK activation [7]. In chronic pain induced by spared nerve injury, activation of AMPK reduces the excitability of dorsal root ganglion neurons to relieve pain [8]. In a mouse model of acute incision-induced pain, local administration of resveratrol cream or systemic administration of metformin activates AMPK to alleviate pain [9]. In lipoteichoic acid (LTA)-induced in ammation, the AMPK activator 5-amino-1 -β-d-ribofuranosyl-imidazole-4-carboxamide (AICAR) inhibits LTA-induced neutrophil in ux and cytokines levels in the bronchoalveolar space to alleviate in ammation [10].
iNOS is an important in ammatory factor produced by macrophages [11]. In lipolysaccharide (LPS)induced macrophages, nuclear translocation of STAT3 promotes the expression of iNOS [12]. Phosphorylated STAT3 (p-STAT3, Ser727) in the nuclei increases the expression of iNOS [12,13]. The stimulation of pro-in ammatory cytokines, such as TNF-α and IL-1 β, leads to the production of ROS in the mitochondria [14,15]. Excessive ROS activation of MPTP(mitochondrial permeability transition pore) results in mitochondrial dysfunction [16]. In complete Freund's adjust (CFA)-induced in ammatory pain, cytokines released from activated macrophages play an important role in mediating the peripheral in ammatory response and sensitization of peripheral sensory nerves [17,18]. In the model of cerebral ischemia injury, electroacupuncture treatment promotes the phosphorylation of STAT3 (Ser727) signi cantly, and participates in the protective effect of electroacupuncture on cerebral ischemia injury [19]. In the model of myocardial ischemia-reperfusion injury of rats, ZnCl 2 promotes the phosphorylation of STAT3(Ser727) in cytoplasm, and attenuates mitochondrial damage and accumulation of mitochondrial ROS (reactive oxygen species) [20].
It is unclear whether STAT3 phosphorylation in activated macrophages play a role in the analgesic effect of AMPK activation on in ammatory pain. In this study, we determined whether AMPK activation reduces in ammatory pain by regulating STAT3 nuclear translocation and phosphorylation. Our study provides new information about the signaling mechanisms involved in the analgesic effect of AMPK activation on in ammatory pain.

Mouse models
The experimental procedures were approved by the ethics committee of Tongji Medical College of Huazhong University of Science and Technology and carried out in strict accordance with the ethical guidelines of the International Association for the Study of Pain. Male C57BL/6 mice (8-9-week old) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. Four to six mice were housed in each cage (12 h-light and 12 h-dark, ambient temperature of22-24℃) and had free access to drinking and food. Before experiments, all mice were adapted to the environment. As reported previously [3], 25 µl of CFA (Sigma, F5881-10 ml) was subcutaneously injected into the plantar side of the left hindpaw to induce in ammatory pain. The control group received an injection of 25 µl normal saline (N.S.).

Nociceptive Behavioral Tests
The withdrawal threshold of pain was measured three days before injection of CFA or vehicle. After 30 minutes of acclimation, we applied von Frey laments (Stoelting, wood Dale, Italy) to the plantar surface of the left hindpaw vertically. Each lament was bend for 5 s with enough force, and the rapid withdrawal of the hindpaw was considered a positive reaction. The "up-down" method was used to determine the mechanical pain threshold of mice [21]. Each test was repeated twice, and the average value was used.
Hot-plate was used to measure the thermal pain threshold, and the surface temperature was maintained at 53 °C. When the mice were placed on the hotplate, the latency of rapid paw withdrawal was recorded. The mice were removed immediately after the withdrawal response was observed. We set 20 seconds as the cut-off to prevent tissue damage [22]. The hot-plate test was repeated every 5 minutes for three times, and the averaged value was calculated.
Secondary antibodies were anti-rabbit HRP and Anti-mouse HRP (1:20000). Quantitative Real-time PCR Total RNA was isolated from local in ammatory skin tissues and NR8383 cells using Trizol Reagent (Invitrogen, TRIzol® Reagent,#15596-018). Spectrophotometer (Thermo Scienti c, USA) was used to quantify the concentration of the total RNA. We used Hifair® 1st Strand cDNA Synthesis SuperMix for qPCR(gDNA digester plus)to reverse total RNA into cDNA. We used Cham QTM Universal SYBR® qPCR Master Mix (Nanjing, China) on the Applied biosystems QuantStudio 7 Flex Thermo Fisher) for qPCR. Expression values of the iNOS mRNA were normalized to the corresponding expression of β-actin mRNA.
We used 2 − ΔΔCt method to calculate relative expression levels of iNOS mRNA. The sequence-speci c primers used are listed below. (Jackson Immuno Research, USA). Sections were incubated with DAPI for the nucleus staining for 5 min and then washed 3 times in 0.01 M PBS for 5 min. Sections were cover-slipped with anti-quenching mounting agent. Images were acquired using uorescence microscope (BX51, Olympus, Japan) and were analyzed using NIH Image J software (Bethesda, MD, USA).
To determine whether AICAR induces activation of AMPK to decrease iNOS expression, NR8383 cells were treated with AICAR (0.5 mM) and Compound C (20 µM) before treatment with N.S. or CFA (100 µg/ml).
NR8383 cells were exposed to CFA for 24 hours. We then collected cells for western blotting and ow cytometry.

Lentivirus Transfection
The control vector (WT) and vectors encoding the STAT3 Ser727 mutant were transfected into NR8383 cells according to the manufacturer's instruction. Brie y, cultured NR8383 cells were transfected for 48 h with the expression vectors for STAT3 S727D (Lentivirus vector encoding an m-Cherry-STAT3 fusion protein that carries a serine-to-aspartate substitution at codon 727, m-Cherry-STAT3 S727D) and WT STAT3 (Lentivirus vector encoding an m-Cherry-STAT3 fusion protein). The expression of m-Cherry-STAT3 and m-Cherry-STAT3 S727D was validated via a uorescence microscope.

Statistical analysis
All data were expressed by mean ± SEM. The two-way analysis of variance (ANOVA) followed by Bonferroni's post hoc test were used to determine statistical difference in the withdrawal thresholds in different groups (SPSS, version 11.0). One-way ANOVA and Newman Keuls post hoc test were used to analyze the expression level of protein and uorescence intensity. The comparison between the two groups was made using non-paired Student's t test. p < 0.05 was considered statistically signi cant.

AICAR reduces CFA-induced pain hypersensitivity and increases AMPK phosphorylation
Mice received CFA injection exhibited mechanical allodynia and thermal hyperalgesia (Fig. 1). A single subcutaneous administration of AICAR (20 µg) at day 7 after CFA injection suppressed mechanical allodynia and thermal hyperalgesia signi cantly (Fig. 1a, b). AICAR also increased the phosphorylation of AMPK (Fig. 1c-e). p-AMPK was present in CD68-positive macrophages of in ammatory skin tissues (Fig. 1f). The results suggest that activation of AMPK with AICAR in in ammatory skin tissues reduces in ammatory pain.

Activation of AMPK with AICAR inhibits the expression of iNOS
Immunoblotting and qPCR were used to test the effect of AMPK activation on iNOS expression in in amed skin tissues. CFA treatment signi cantly increased in the protein and mRNA levels of iNOS (Fig. 1g, h and i). Treatment with AICAR via a single subcutaneous injection at day 7 after CFA injection signi cantly inhibited the expression levels of iNOS (Fig. 1g, h and i).

Activation of AMPK inhibits STAT3 nuclear translocation and promotes cytoplasmic STAT3 (Ser727) phosphorylation in macrophages
We next determined whether activation of AMPK affects STAT3 nuclear translocation and phosphorylation in macrophages. CFA treatment signi cantly increased the protein level of p-STAT3 (Ser727), and AICAR further increased p-STAT3 (ser727) signi cantly in in ammatory skin tissues (Fig. 2a, b). There was no signi cant difference in the total protein level of STAT3 between the control, CFA, and CFA plus AICAR groups (Fig. 2a, c).
In addition, we tested the protein levels of STAT3 and p-STAT3 (Ser727) in the cytoplasmic and nuclear fractions of in amed skin tissues. CFA signi cantly increased the protein level of p-STAT3 (Ser727) and STAT3 in nuclear fraction of in ammatory skin tissues. AICAR reversed this effect of CFA (Fig. 2d-f).
Compared with the CFA model group, AICAR signi cantly increased the protein level of p-STAT3(Ser727) in the cytoplasmic fraction of in ammatory skin tissues (Fig. 2g-i). Immuno uorescence labeling showed the expression of STAT3 and p-STAT3 (Ser727) in macrophages of in ammatory skin tissues (Fig. 2j, k). These results suggest that CFA induces the nuclear translocation of STAT3 and increases STAT3(Ser727) phosphorylation in macrophages of local in ammatory skin tissues. Also, AMPK activation inhibits nuclear translocation of STAT3 and increases the phosphorylation of STAT3(Ser727) in the cytoplasm of macrophages.
The AMPK activation in macrophages promotes the phosphorylation of STAT3 (Ser727) in cytoplasm and inhibit the nuclear translocation of STAT3 and the expression of iNOS NR8383 cells were used to induce macrophage in ammation with CFA treatment [19]. CFA treatment signi cantly increased iNOS protein levels in NR8383 cells (Fig. 3). Compared with the CFA group, AICAR signi cantly inhibited the expression of iNOS and increased the phosphorylation of STAT3 (Ser727).
AICAR had no signi cant effect on total protein levels of STAT3. Compound C, an AMPK antagonist, reversed the effect of AICAR (Fig. 3a-d). These data suggest that activation of AMPK enhances the phosphorylation of STAT3 (Ser727) and inhibits the expression of iNOS in activated macrophages.
Furthermore, in NR8383 cells, we determined whether AMPK activation inhibits STAT3 nuclear translocation. Compared with the control group, CFA treatment signi cantly increased the levels of p-STAT3 (Ser727) and STAT3 in the nuclei and signi cantly decreased the levels of p-STAT3 (Ser727) and STAT3 in the cytoplasm. Treatment with AICAR reversed the effects of CFA on the levels of p-STAT3 (Ser727) and STAT3 in the nuclei and the levels of p-STAT3 (Ser727) and STAT3 in the cytoplasm (Fig. 4).
In addition, the AMPK antagonist Compound C blocked the effect of AICAR in the above experiments ( Fig. 4a-e). Immuno uorescence labeling con rmed the effects of CFA and AMPK activation on the distribution of p-STAT3 (Ser727) and STAT3 in the cytoplasm and nuclei of macrophages (Fig. 4f).
Activation of AMPK alleviates ROS accumulation and mitochondrial damage induced by CFA ZnCl 2 treatment promotes the phosphorylation of STAT3 (Ser727) in the cytoplasm (mitochondria) and attenuates the mitochondrial damage and ROS accumulation to alleviate the in ammatory damage in a rat model of myocardial ischemia-reperfusion injury [20]. We next used NR8383 cells to determine whether AMPK activation reduces ROS accumulation and mitochondrial damage induced by CFA. Compared with the control group, CFA signi cantly increased the uorescence intensity of DCF and signi cantly decreased the uorescence intensity of Mito tracker red CMXRos. Treatment with AICAR reduced CFAinduced changes in DCF uorescence intensity and the uorescence intensity of Mito tracker red CMXRos in NR8383 cells (Fig. 5). Co-treatment with Compound C reversed the effects of AICAR (Fig. 5a-d). These results suggest that in ammation induces ROS accumulation and mitochondrial damage in macrophages and that activation of AMPK attenuates these in ammation-induced effects.
Promoting STAT3(Ser727) phosphorylation alleviates CFA-induced ROS accumulation and mitochondrial damage in macrophages The above experiments showed that AMPK activation promotes the phosphorylation of STAT3 (Ser727) in cytoplasm of macrophages. We then determined whether STAT3 (Ser727) phosphorylation plays a role in the inhibitory effect of AMPK activation on CFA-induced ROS accumulation and mitochondrial damage in NR8383 macrophages. The Ser727 site of STAT3 was mutated to aspartate, S727D, to promote phosphorylation of STAT3, as reported previously [15,23]. We used m-Cherry-STAT3 S727D (and m-Cherry STAT3 wild type(WT) as a negative control) lentiviral vectors to transfect NR8383 macrophages. CFA treatment signi cantly increased the ROS level in NR8383 cells transfected with STAT3 WT vectors. In NR8383 cells transfected with STAT3 S727D vectors, the CFA-induced ROS level was signi cantly decreased ( Fig. 6a and b).
In addition, we found that the uorescence intensity of Mito tracker Red CMXRos was signi cantly more in NR8383 cells transfected with STAT3 S727D vectors than in NR8383 cells transfected with STAT3 WT vectors (Fig. 6c-e). These results suggest that AMPK activation alleviates ROS accumulation and mitochondrial damage in in ammation-induced macrophages through promoting phosphorylation of STAT3(Ser727).

Discussion
In this study, we used an in ammatory pain model to determine the signaling mechanism involved in the analgesic effect produced by local AMPK activation. AMPK is a member of the family of metabolically sensitive protein kinases, which contains α-catalytic subunits and β-and γ-regulatory subunits [24,25]. AMPK activation can inhibit a variety of pro-in ammatory signal cascades, including c-Jun N-terminal kinase (JNK), nuclear factor kappa B (NF kappa B) and JAK-STAT (Janus kinase signal transducer and activator of transcription) [26][27][28][29]. When AMPKα is knocked out in mice, the nociceptive response is enhanced in in ammatory pain [30,31]. In the rat model of osteoarthritis, activation of AMPK/mTOR (mammary target of rapamycin) reduces the in ammatory response of chondrocytes induced by IL-1β and promotes the autophagy to reduce in ammatory injury and osteoarthritic pain [32].
In our study, we showed that activation of AMPK in macrophages inhibited the expression of iNOS and reduced in ammatory pain. We also showed that AMPK activation regulates STAT3 to inhibit iNOS expression in in ammatory pain. As an important neurotransmitter, nitric oxide mediates the pain in peripheral injury model [33,34]. The iNOS is increased in in ammatory and neuropathic pain, which promotes the synthesis of nitric oxide to induce the sensitization of peripheral or central sensory nerves and pain [35,36]. Inhibition of iNOS expression in macrophages is involved in alleviating in ammatory pain [37]. In carrageenan induced hindpaw in ammatory pain mice model, angelica dahurica can alleviate in ammatory pain by inhibiting the production of iNOS and nitric oxide in macrophages [38]. In the rat model of temporomandibular joint osteoarthritis, AMPK activation inhibits IL-1 β and nitric oxide, and alleviates nociceptive pain [39].
STAT3 is a member of the signal transducer and activator of transcription (STAT) family, which includes seven members, STAT 1, STAT 2, STAT 3, STAT 4, STAT 5A, STAT 5B and STAT 6. STAT3 usually transmits signals from activated receptors or intracellular kinases to the nucleus, thus activating and regulating gene transcription [40]. STAT3 activation is induced by a variety of cytokines and growth factors, including IL-6 signal transduction receptor chain gp130 (such as IL-6, tumor suppressor M, interleukin-11) or homodimeric cytokine receptor (such as granulocyte colony stimulating factor G-CSF), as well as growth factors (such as epidermal growth factor) acting through protein tyrosine kinase receptor [41]. Under the stimulation of cytokines, STAT3 tyrosine phosphorylation has transcription activity, and nuclear translocation of STAT3 dimer achieves the transcription of target genes [42][43][44]. Phosphorylation of STAT3 (Ser727) enhances transcription activity of STAT3 [41]. In Raw 264.7 macrophage cell line, LPS causes nuclear translocation of STAT3, and p-STAT3 (Ser727) promotes the transcription of iNOS in nuclei [12]. In LPS-induced sepsis rat model, the activation of STAT3 by miR-34a mediates the expression and secretion of iNOS in pulmonary macrophages [45]. In addition, ZnCl 2 promotes the phosphorylation of STAT3 (Ser727) in the cytoplasm, and improves the mitochondrial damage and accumulation of ROS to alleviate the in ammatory injury [15]. In our study, we found that CFA caused nuclear translocation of STAT3 in activated macrophages, and p-STAT3 (Ser727) in the nuclei promoted the expression of iNOS. AICAR activates AMPK to inhibit nuclear translocation of STAT3 in activated macrophage, and reduces p-STAT3(Ser727) in nuclei to downregulate expression of iNOS.
Reactive oxygen species (ROS) are intermediate products of normal oxygen metabolism, which are involved in the regulation of cell proliferation and in ammatory processes [46]. ROS is also involved in the pain and can recruit CX3CR1-positive monocyte macrophages to the site of peripheral nerve injury after cancer chemotherapy, which can release ROS to activate TRPA1 (transient receptor potential ankyrin 1) channels [47]. In this study, we found that AMPK activation promotes the phosphorylation of STAT3 Ser727 in the cytoplasm of macrophages to reduce the ROS accumulation and mitochondrial damage induced by CFA.

Conclusions
Our study shows new information about the signaling mechanism involved in the antinociceptive effect of activating AMPK on in ammatory pain. Activation of AMPK inhibited the nuclear translocation of STAT3 and decreased the phosphorylation of STAT3(Ser727) in macrophages, thus reducing the expression of iNOS. AMPK activation also promoted cytoplasmic STAT3(Ser727) phosphorylation in macrophages to reduce ROS accumulation and mitochondrial damage (Fig. 7). This signaling cascade likely play a role in the analgesic effect of AMPK activators on in ammatory pain.

Consent for publication
Not applicable.