AMPK activation by peri-sciatic nerve administration of ozone attenuates CCI-induced neuropathic pain in rats

Abstract

Neuropathic pain is a debilitating clinical condition with few efficacious treatments, warranting development of novel therapeutics. Ozone is widely used as an alternative therapy for many different pain conditions, with exact mechanisms still elusive. In this study, we found that a single peri-sciatic nerve injection of ozone decreased mechanical allodynia and thermal hyperalgesia, and normalized the phosphorylation of protein kinase C γ, N-methyl-D-aspartate receptor, and extracellular signal-regulated kinase in a chronic constriction injury (CCI) model in rat sciatic nerve. Meanwhile, ozone significantly suppressed CCI-induced activation of spinal microglia. More importantly, the anti-nociceptive effect of ozone depended on the activation of 5′-adenosine monophosphate (AMP)-activated protein kinase (AMPK), which was proved by the fact that the phosphorylated AMPK level increased during the ozone therapy and AMPK antagonist abolished the effect of ozone in vivo and in vitro. In addition, direct injection of AMPK agonist could replicate the anti-nociceptive effect of ozone in CCI rats. In conclusion, our observations indicate that peri-sciatic nerve injection of ozone activates AMPK to attenuate CCI-induced neuropathic pain.

AMPK, ozone therapy, neuropathic pain

Introduction

Neuropathic pain is a devastating chronic pain caused by a primary lesion or dysfunction in both peripheral and central nervous system (Baron, 2006). The symptoms of neuropathic pain are characterized by hyperalgesia, allodynia, and spontaneous pain. Despite multiple approaches were tested, the treatment of neuropathic pain continues a challenge to physicians (Price et al., 2007; Huang et al., 2008; Jimenez-Diaz et al., 2008; Geranton et al., 2009; Price and Geranton, 2009; Melemedjian et al., 2010). In general, neuropathic pain is a debilitating clinical condition with few efficacious treatments, warranting development of novel therapeutic interventions including alternative medicine treatment (Baron et al., 2010; Gaskell et al., 2014).

Ozone is an unstable and irritant gas with a pungent odour, primarily known for its ecological properties and industrial application. Ozone therapy has been developed and well known as an alternative and effective approach in certain clinical circumstances (Wells et al., 1991). Recently, ozone therapy has gained more prominence and credibility in the treatment of chronic pain, particularly in Europe, Russia, and China (Bassi et al., 1982; Agrillo et al., 2012; Niesters and Dahan, 2012; Zhang et al., 2012; Bertolotti et al., 2013; Clavo et al., 2013; Neimark et al., 2014; Al Habashneh et al., 2015). For example, ozone therapy suppressed low back pain in lumbar and cervical disk herniation in patients without any significant complications when topically applied (Magalhaes et al., 2012). Despite the widespread use of ozone to treat a variety of clinical conditions, questions persist concerning its potential toxicity as an oxidant agent vs. its reported therapeutic efficacy. Therefore, continuous efforts are needed to elucidate the mechanisms underlying its biological action and to avoid potential adverse effects.

Ozone therapy has also been proposed as an immunomodulator and activator of cellular metabolism that shows long-term anti-inflammatory effects (Holz et al., 2005). It also increases the endogenous antioxidant system activities in endotoxic and septic shock models (Zamora et al., 2005). Moreover, a single subcutaneous injection of ozone attenuated mechanical allodynia induced by surgical procedure and decreased overexpression of pro-inflammatory caspases in the orbito-frontal cortex of neuropathic mice (Fuccio et al., 2009). However, these changes in the brain cannot completely explain the analgesia effect of ozone when it is conventionally applied in close proximity to the locus of pain. In the present study, we aim to systemically examine the inhibitory effects of ozone on the pain hypersensitive states in rats received chronic constriction injury (CCI) and further explore the underlying mechanisms.

Results

Local injection of ozone attenuates CCI-induced pain-related behaviours in rats

In our study, rats received CCI exhibited thermal hyperalgesia and mechanical allodynia. These pain-related behaviours were greatly suppressed by a single peri-sciatic nerve administration of ozone (3.75, 7.5, or 15 μg of ozone per 0.5 ml oxygen for each single injection) at postoperative day 14 in a dose-dependent manner, with the analgesia effect of ozone lasting for >24 h (Figure 1A and B). In addition, repetitive daily administration of ozone (15 μg/0.5 ml/day for 7 days) from postoperative day 14 produced a sustain inhibition in mechanical allodynia and thermal hyperalgesia (Figure 1C and D). There were no significant differences in pain-related behaviours among sham and oxygen- or ozone-treated naïve rats (Figure 1F), either between air- and oxygen-treated CCI rats (Figure 1E). Furthermore, none of these treatments significantly changed the thermal or mechanical threshold of the feet contralateral to those subjected to CCI.

Figure 1

Intramuscular injection of ozone attenuates CCI-induced mechanical allodynia and thermal hyperalgesia in rats. (A) Effects of a single injection of ozone (3.75, 7.5, or 15 μg/0.5 ml) on mechanical allodynia in CCI rats. (B) Effects of a single injection of ozone (15 μg/0.5 ml) on thermal hyperalgesia induced by CCI in rats. (C) Effects of multiple injections of ozone (15 μg/0.5 ml/day for 7 days) on mechanical allodynia in CCI rats. (D) Effects of multiple injections of ozone (15 μg/0.5 ml/day for 7 days) on thermal hyperalgesia in CCI rats. (E) Effects of a single injection of air (0.5 ml) and oxygen (0.5 ml) on mechanical allodynia in CCI rats. (F) Effects of a single injection of oxygen (0.5 ml) and ozone (15 μg/0.5 ml) on sham rats. The diagram shows the paw withdrawal latencies in response to mechanical stimulation as assessed with Von Frey fibres in A, C, E, F or thermal stimulation in B, D (n = 6 rats/group). Each arrow indicates an injection. Two-way ANOVA revealed a significant difference at *P < 0.05 and **P < 0.01 vs. CCI group.

Ozone suppresses CCI-induced upregulation of phosphorylated NR1, NR2B, PKCγ, and ERK in the spinal cord

The upregulated phosphorylation of N-methyl-D-aspartate receptor (NMDAR) and the aberrant activation of extracellular signal-regulated kinase (ERK) and protein kinase C γ (PKCγ) in the spinal cord have a well-established role of central sensitization in neuropathic pain (Weyerbacher et al., 2010). Here, as shown in Figure 2, CCI caused significant increases in the levels of phosphorylated NR1, NR2B, PKCγ and ERK, which was in the same pattern as the occurrence of neuropathic pain-related behaviours in rats. Treatment with ozone (3.75, 7.5, and 15 μg/0.5 ml) by a single peri-sciatic intramuscular injection at postoperative day 14 markedly and dose-dependently inhibited the phosphorylation of these proteins (Figure 2A and B).

Figure 2

Ozone suppresses CCI-induced phosphorylation of NR1, NR2B, PKC, and ERK in the spinal cord. Ozone (3.75, 7.5, or 15 μg/0.5 ml) was administrated at postoperative day 14. Spinal cord tissues were collected at 2 h after ozone treatment for western blotting. Representative western blots and quantification for pNR1, pNR2B bands (A) and pPKC, pERK bands (B) (n = 4 each group) are shown. One-way ANOVA revealed a significant difference at **P < 0.01 vs. control and ^##P < 0.01 vs. CCI group.

Ozone suppresses CCI-induced activation of spinal microglia

Both western blotting and immunofluorescence assays were performed to determine whether ozone suppressed the activation of spinal microglia, which was reported to mediate CCI-induced hyperalgesia and allodynia. Compared to the naive control group, CCI rats showed a significant increase in spinal IBA-1 level, indicative of microglia activation in the spinal cord (Figure 3A and B). The co-instantaneous administration of ozone reversed these pathological changes in CCI rats, but did not affect the spinal IBA-1 level in naïve rats (P > 0.05), suggesting that ozone suppresses CCI-induced spinal microglia activation (Figure 3).

Figure 3

Ozone suppresses CCI-induced activation of spinal microglia. Ozone (15 μg/0.5 ml) was administrated at postoperative day 14. Spinal cord tissues were collected at 2 h after ozone treatment. (A) Representative western blots and quantification for IBA-1 bands (n = 4 each group) are shown. (B) Immunofluorescence of IBA-1 in the spinal cord. Left, representative confocal images; right, quantification for IBA-1 fluorescence. Magnification, 100× and 200×. One-way ANOVA revealed a significant difference at ^##P < 0.01 vs. control and **P < 0.01 vs. CCI group.

AMPK activation is required for ozone to attenuate CCI-induced neuropathic pain in rats

It has been reported that 5′-adenosine monophosphate (AMP)-activated protein kinase (AMPK) could be directly activated by H₂O₂ through reversible oxidation of cysteine residues (Zmijewski et al., 2010). Ozone has a strong oxidizing capacity as H₂O₂, suggesting a possible effect of ozone on directly activating AMPK. In addition, AMPK was recently recognized as a potential target for the prevention and intervention for injury-induced pain (Price and Dussor, 2013). Therefore, we hypothesized that ozone might suppress neuropathic hyperalgesia and allodynia by activating AMPK. Here we showed that AMPK phosphorylation in the sciatic nerve was significantly increased after ozone administration in a dose-dependent manner (Figure 4A), which was inhibited by the AMPK inhibitor Compound C (100 μg, peri-sciatic intramuscular injection) to the basal level (Figure 4B). These results indicated that ozone could directly activate AMPK.

Figure 4

AMPK activation is required for ozone to attenuate CCI-induced neuropathic pain in rats. (A) Local administration of ozone (3.75, 7.5, or 15 μg/0.5 ml) activated AMPK in the sciatic nerve in a dose-dependent manner. (B) Local administration of ozone (15 μg/0.5 ml) activated AMPK in the sciatic nerve, which could be blocked by AMPK inhibitor Compound C. Control rats were administrated with 0.5 ml of air. Representative western blots and quantification for AMPK and pAMPK bands (n = 4 each group) are shown. One-way ANOVA revealed a significant difference at **P < 0.01 vs. control and ^##P < 0.01 vs. ozone-treated group. (C) AMPK inhibitor Compound C (30 μg/300 μl or 100 μg/300 μl) reversed the effects of ozone (15 μg/0.5 ml). Two-way ANOVA revealed a significant difference at ^#P < 0.05, ^##P < 0.01 vs. ozone-treated group and *P < 0.05, **P < 0.01 vs. CCI group. (D) Effects of AMPK activators AICAR (0.25 mg/kg) and A769662 (0.25 mg/kg) on CCI-induced mechanical allodynia in rats. Two-way ANOVA revealed a significant difference at *P < 0.05, **P < 0.01 vs. CCI group. (E) Effects of AMPK inhibitor Compound C (100 μg/300 μl) on CCI-induced mechanical allodynia in rats. (F) Effects of AMPK siRNA on CCI-induced mechanical allodynia and ozone treatment in rats. Two-way ANOVA revealed a significant difference at ^#P < 0.05, ^##P < 0.01 vs. negative siRNA-treated group and *P < 0.05, **P < 0.01 vs. Day 14 after CCI. All drugs were administrated intramuscularly to peri-sciatic nerve. The diagram shows the paw withdrawal latencies in response to mechanical stimulation as assessed with Von Frey fibres (n = 6 rats/group). An arrow indicates an injection.

We then questioned whether AMPK is required for ozone-mediated inhibition of CCI-induced neuropathic pain and thus examined whether the AMPK inhibitor Compound C could reverse the effects of ozone. As expected, Compound C (30 and 100 μg, peri-sciatic intramuscular injection) significantly reversed ozone-induced alleviation of mechanical allodynia on Day 14 after CCI (Figure 4C). There were no significant differences in pain-related behaviours between CCI rats with or without Compound C alone treatment (Figure 4E). Knockdown of AMPK by peri-sciatic nerve administration of siRNAs against AMPKα also abolished ozone-mediated suppression of mechanical allodynia in CCI rats (Figure 4F), confirming that AMPK activation is required for ozone to attenuate CCI-induced neuropathic pain. Moreover, peri-sciatic intramuscular injection of the AMPK activator AICAR (0.25 mg/kg) or A769662 (0.25 mg/kg) at postoperative day 14 mimicked the effects of ozone on ameliorating neuropathic pain in CCI rats (Figure 4D).

AMPK activators suppress CCI-induced upregulation of protein phosphorylation and microglia activation in the spinal cord

Next, we questioned whether AMPK activators could resemble ozone to inhibit the activation of NMDAR, PKCγ, ERK, and IBA-1 induced by CCI in rats. Indeed, peri-sciatic intramuscular injection of the AMPK activator AICAR (0.25 mg/kg) or A769662 (0.25 mg/kg) markedly inhibited the phosphorylation of NR1, NR2B, PKCγ, and ERK (Figure 5A and B) and the activation of IBA-1 in rat spinal cord on Day 14 after CCI (Figure 5C), in accordance with their ameliorative effects on neuropathic pain in CCI rats.

Figure 5

AMPK activators resemble the effects of ozone to suppress CCI-induced phosphorylation of NR1, NR2B (A), PKC, ERK (B), and expression of IBA-1 (C) in the spinal cord. AICAR or A769662 (0.25 mg/kg) was administrated at postoperative day 14. Spinal cord tissues were collected at 2 h after AMPK activator treatment. Representative western blots and quantification for pNR1, pNR2B, pPKC, pERK, and IBA-1 bands (n = 4 each group) are shown. One-way ANOVA revealed a significant difference at **P < 0.01 vs. control and ^##P < 0.01 vs. CCI group.

Ozone suppresses CCI-promoted ERK phosphorylation in the sciatic nerve

Peripheral sensitization contributes to the pain hypersensitivity found at the site of tissue damage and inflammation. Accumulating evidence shows that ERK-mediated modulation of Nav1.7 may play a key role in neuropathic pain in the sciatic nerve (Tillu et al., 2012; Yan et al., 2012). Therefore, we further examined whether ozone could regulate the ERK phosphorylation in injured sciatic nerve.

Previous studies suggest that peripheral nerve injury induced reorganization of translation machinery in the peripheral nervous system of rats and mice along with enhanced ERK activity, which could be inhibited by AMPK activators, such as metformin and A769662 (Melemedjian et al., 2011). Our data showed that the level of phosphorylated ERK in the sciatic nerve was significantly increased after CCI, which was notably suppressed by ozone (15 μg/0.5 ml) (Figure 6A). Moreover, the AMPK inhibitor Compound C completely reversed the inhibitory effects of ozone on CCI-induced increase in ERK phosphorylation in the sciatic nerve (Figure 6A). In contrast, AMPK activators AICAR and A769662 significantly suppressed CCI-induced upregulation of phosphorylated ERK in the sciatic nerve (Figure 6B), mimicking the effects of ozone.

Figure 6

Ozone suppresses CCI-induced phosphorylation of ERK in the sciatic nerve. Ozone (15 μg/0.5 ml), AICAR (0.25 mg/kg), or A769662 (0.25 mg/kg) was administrated at postoperative day 14. Sciatic nerve tissues were collected at 2 h after the treatment with ozone, AICAR, or A769662. (A) Effects of ozone (15 μg/0.5 ml) on CCI-induced phosphorylation of ERK in rat sciatic nerve. (B) Effects of AMPK activators AICAR (0.25 mg/kg) and A769662 (0.25 mg/kg) on CCI-induced phosphorylation of ERK in rat sciatic nerve. Representative western blots and quantification for pERK bands (n = 4 each group) are shown. One-way ANOVA revealed a significant difference at **P < 0.01 vs. control, ^#P < 0.05 and ^##P < 0.01 vs. CCI group, and ^&&P < 0.01 vs. ozone-treated group.

Ozone activates AMPK and suppresses lipopolysaccharide-induced NF-κB activation in macrophages

In the rat sciatic nerve CCI model, macrophages produce pro-inflammatory cytokines that contribute to neuropathic pain. Since p65 NF-κB translocation from the cytosol to the nucleus was shown to promote the expression of pro-inflammatory cytokines in macrophages, Raw264.7 cells were stimulated with lipopolysaccharide (LPS) to evaluate the effects of ozone on NF-κB activation in macrophages.

We showed that ozone increased AMPK phosphorylation in Raw264.7 cells, which could be blocked by Compound C (Figure 7A). LPS promoted p65 NF-κB translocation from the cytosol to the nucleus, which was significantly reduced by a post-treatment (6 h) with ozone (Figure 7B), in accordance with ozone-induced upregulation of AMPK phosphorylation.

Figure 7

Ozone activates AMPK and suppresses LPS-induced NF-κB activation in macrophages. (A) After a pre-treatment of LPS (1 μg/ml, 12 h), RAW264.7 cells were treated with different doses of ozone for 2 h. Ozone increased AMPK phosphorylation in Raw264.7 cells, which could be blocked by Compound C. Representative western blots and quantification for AMPK and pAMPK bands (n = 4 each group) are shown. (B) After a pre-treatment of LPS (1 μg/ml, 12 h), RAW264.7 cells were treated with ozone (30 μg/ml) for 6 h. Ozone inhibited the LPS-promoted NF-κB translocation from the cytosol to the nucleus, as demonstrated by the immunofluorescence of p65 in RAW264.7 cells. Left, representative confocal images; right, quantification for cells with p65 translocation. Magnification, 200x. One-way ANOVA revealed a significant difference at *P < 0.05 and **P < 0.01 vs. control and ^#P < 0.05 and ^##P < 0.01 vs. ozone- or LPS-treated group.

Discussion

Our study demonstrated that peri-sciatic intramuscular injection of ozone dramatically suppressed mechanical allodynia and thermal hyperalgesia induced by CCI in the sciatic nerve. Ozone therapy significantly normalized the NR1, NR2B, ERK, and PKC phosphorylation levels in the spinal cord evoked by CCI. Furthermore, ozone profoundly upregulated AMPK phosphorylation levels in the sciatic nerve, and the analgesic effects of ozone were completely prevented by AMPK inhibitor. In addition, ozone significantly suppressed CCI-induced activation of spinal microglia. Take together, our results illustrate, for the first time, that AMPK activation plays an essential role in analgesic effect of ozone therapy and further suggest AMPK activator as an alternative analgesic option for ozone therapy.

Ozone, which has been widely used in Europe, China, and Africa against chronic pain in clinic, is considered to have anti-oxidation, anti-inflammatory, and analgesic effects (Bonforte et al., 2013; Ucar et al., 2013; Kazancioglu et al., 2014; Menendez et al., 2014). In 1998, Muto treated patients with ozone injection in intervertebral disc and vertebral side to cure prolapsed of lumbar intervertebral disc, and 78% patients were recovered (Muto and Avella, 1998). Accumulating evidences also demonstrate that ozone is effective to treat prolapse of lumbar intervertebral disc (Gazzeri et al., 2007; Paoloni et al., 2009; Gautam et al., 2011; Zhang et al., 2012). It has been reported that, from 1966 to September 2011, ozone therapy appeared to yield positive results and low morbidity rates when applied percutaneously for the treatment of chronic low back pain in the multicenter trials of thousands of cases (Magalhaes et al., 2012).

Ozone therapy has been used and heavily studied for more than a century. A number of randomized clinical trials have been carried out or are being considered to assess the effect of ozone (D'Erme et al., 1998; Bonetti et al., 2005; Zambello et al., 2006; Gallucci et al., 2007; Paoloni et al., 2009; Elvis and Ekta, 2011). Bonetti et al. (2005) reported that in the randomized series of 306 patients, 57.5% of 80 patients in the disc disease group treated with steroid deemed the clinical outcome to be excellent, as did 62.8% of 70 patients in the group with no disc disease after steroid infiltration. In contrast, 74.4% of 86 patients with disc disease reported complete remission of pain after ozone therapy, as did 75.0% of 70 patients with no disc disease. In another randomized study, Zambello et al. (2006) randomized 351 patients with low back pain for treatment with either ozone or steroid (epidural) and planned a crossover during the follow-up to the other group in case of failure to respond to treatment after 4 weeks of therapy. The long-term outcome remained excellent or good in 47.3% of 171 patients treated by epidural steroid injections and in 77.1% of 180 patients treated with O₂–O₃. Recently, Paoloni et al. (2009) conducted a multicenter, randomized, double-blinded, ‘simulated therapy’-controlled clinical trial. A greater percentage of patients became pain-free (61% vs. 33%, P < 0.01) in the ozone group. Although the analgesic effect of ozone is well documented, little is understood about its biological basis.

Our study gave the direct evidence that ozone could suppress CCI-induced mechanical allodynia and thermal hyperalgesia in rats. A single injection of ozone has shown a significant therapeutic effect (Figure 1A and B). Moreover, repetitive daily administration of ozone was more effective in the aspect of the analgesic effect and no analgesic tolerance (Figure 1C and D). Consistent with our results, it was reported in 2009 that a subcutaneous injection of ozone suppressed mechanical allodynia in neuropathic mice (Fuccio et al., 2009). Since air or oxygen alone did not show any analgesic effect, it could be inferred that it was ozone, but not oxygen, that attenuated the mechanical allodynia and thermal hyperalgesia induced by CCI. It also ruled out the possibility that the analgesic effect of ozone was due to acupuncture.

It has been widely accepted that upregulated phosphorylation of NMDAR, activation of spinal microglia, and aberrant activation of ERK and PKCγ play essential roles in central sensitization in neuropathic pain (Basbaum et al., 2009). Our results showed that ozone provides an inhibition in NMDAR (NR1 and NR2B) phosphorylation, IBA-1 upregulation, and aberrant activation of PKCγ and ERK in spinal cord of CCI rats.

However, the peripheral efficacy of ozone in injured sciatic nerve might be more important for its analgesic effect, since it was injected intramuscularly. We further identified that AMPK activation was the central mechanism for ozone to produce antinociception in pain hypersensitivity states. AMPK is a heterotrimeric Ser/Thr protein kinase activated by alterations in cellular AMP: adenosine triphosphate (ATP) ratio and servers as an energy sensor that regulates energy homoeostasis and metabolic stress. Once activated, AMPK inhibits ATP-consuming anabolic processes such as protein translation (Carling et al., 2012; O'Neill and Hardie, 2013). The evidences for the role of AMPK in pain modulation came from a variety of studies in animals (Melemedjian et al., 2011; Roy Chowdhury et al., 2012; Russe et al., 2013). Here we showed that peri-sciatic intramuscular administration of ozone induced AMPK phosphorylation in the sciatic nerve (Figure 4A). In addition, the analgesic effect of ozone was inhibited by the AMPK inhibitor Compound C in a dose-dependent manner (Figure 4C). Local injection of the AMPK agonist AICAR or A769662 could mimic the analgesic effect of ozone (Figure 4D). Altogether, these results strongly suggest that AMPK activation plays pivotal roles in the analgesia effect of ozone. Since a number of studies have shown that AMPK activation has other positive effects, such as antineoplasticity and regulation of glucose and lipid metabolism (O'Neill and Hardie, 2013; Vincent et al., 2015), our findings indicate a new direction for applying ozone therapy in treating more diseases.

Several lines of studies show that pharmacological AMPK activation negatively regulates aberrant translation control after nerve injury, resolves neuropathic allodynia, and decreases sensory neuron excitability through suppressing mTOR and MAPK, especially ERK signalling (Price and Dussor, 2013). This is important because the ERK pathway is linked to neuronal excitability and is known to play an important role in the pain neuronal axis as an inducer of cell plasticity (Melemedjian et al., 2011; Price and Dussor, 2013). ERK modulation of several channel types has been well documented, including Kv4.2 (Adams et al., 2000; Schrader et al., 2006), Cav2.2 channels (Martin et al., 2006), and Nav1.7, a channel thought to contribute to amplification of generator potentials to reach the threshold for action potential firing (Stamboulian et al., 2010; Yan et al., 2012). ERK phosphorylation of Nav1.7 mediates an increase in action potential firing and a decrease in latency to first action potential in sensory neurons stimulated with depolarizing ramp current injections (Stamboulian et al., 2010). Human clinical findings and pharmacological inhibition of Nav1.7 indicate a compelling rationale for targeting Nav1.7 in human pain disorders. It has been recently shown that AMPK activators decrease both ERK phosphorylation and the excitability of sensory neurons in response to ramp currents, an effect consistent with decreased phosphorylation of the voltage-gated sodium channel Nav1.7 (Melemedjian et al., 2011). Hence, we further demonstrated that ozone strongly suppressed aberrant phosphorylation of ERK evoked by CCI, providing further evidence that AMPK has been implicated in the analgesic effect of ozone.

Our study offers the first evidence that ozone profoundly upregulates AMPK phosphorylation levels (Figure 4A and B). Consistent with our findings, Hulo's work also suggests that ozone can activate AMPK in lung epithelial cells. However, the mechanism underlying AMPK activation by ozone is still unknown (Hulo et al., 2011). One decade ago, Choi et al. (2001) proposed that AMPK activation coupled with increased levels of H₂O₂. In agreement, a recent study has demonstrated that AMPK could be directly activated by H₂O₂ through reversible oxidation of cytokine residues (Zmijewski et al., 2010). Ozone has a strong oxidizing activity as H₂O₂, suggesting its possible mechanism involving AMPK. In addition, AMPK has been recently reported to sense and respond to pro-oxidant conditions that are induced by reactive oxygen species (ROS) (Cardaci et al., 2012). ROS can act as positive regulators of AMPK activity by means of inducing oxidation reactions at the level of mitochondrial complexes, which finally results in a decrease of the ratio between AMP and ATP. As known to all, ozone has a strong oxidizing capacity, and injection of ozone might increase the production of ROS. Therefore, AMPK activation by ozone may be elicited in a ROS-dependent manner. The detailed molecular mechanism of AMPK regulation by ozone needs to be further studied in future.

Macrophages produce pro-inflammatory cytokines in rat sciatic nerve CCI model of neuropathic pain. Phosphorylation of MAPK, such as ERK and P38, and activation of NF-κB were shown to promote the expression of pro-inflammatory cytokines in macrophages. Consistent with the results in vivo, our in vitro results demonstrated that ozone increased AMPK phosphorylation in Raw264.7 cells, which was blocked by Compound C (Figure 7A). AMPK activator has shown anti-inflammatory effects on LPS-induced inflammation in several cell lines (Yi et al., 2011; Chen et al., 2014; Han et al., 2014), such as BV2 and RAW264.7. Our results showed that LPS could promote p65 NF-κB translocation from the cytosol to the nucleus, and ozone significantly reduced these effects (Figure 7B).

In summary, our data demonstrate that ozone attenuates neuropathic pain via AMPK activation. We clarify the underlying mechanism of the analgesic effect of ozone and provide more powerful evidence supporting the use of ozone. Our results also support AMPK activation as a potential strategy for the treatment of neuropathic pain.

Materials and methods

Ethics statement

All procedures were strictly carried out in accordance with the Regulations of the Ethics Committee of the International Association for the Study of Pain and the Guide for the Care and Use of Laboratory Animals of the Ministry of Science and Technology of China (2006 Edition). All animal experiments were approved by Nanjing Medical University Animal Care and Use Committee and were designed to minimize suffering and the number of animals used.

Animals

Sprague-Dawley rats (weight 180−220 g) were provided by the Experimental Animal Center at Nanjing Medical University, Nanjing, China. Animals were housed five to six per cage under pathogen-free conditions with soft bedding under controlled temperature (22°C ± 2°C) and photoperiods (12-h/12-h light/dark cycle). They were allowed to acclimate to these conditions for at least 2 days before inclusion in experiments. For each group of experiments, the animals were matched by age and body weight. All surgeries were performed under anaesthesia with pentobarbital (50 mg/kg, intraperitoneally, Sigma).

Drugs and reagents

Ozone/oxygen mixture was generated by Medozon compact (Herrmann Apparatebau). Ozone obtained from medicinal grade oxygen was used immediately. AICAR and Compound C were purchased from Sigma. A769662 and antibodies for p65/RelA and IBA-1 were purchased from Abcam. Antibody for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was from Sigma. Antibodies for phosphorylated NMDAR NR1 subunit (Ser896), phosphorylated NMDAR NR2B subunit (Tyr1472), phosphorylated PKCγ (Thr514), phosphorylated ERK (Thr202/Tyr204), phosphorylated p38 (Tyr182), and phosphorylated AMPK (Thr172) were from Cell Signaling Technology. Secondary antibodies were from Chemicon. All other chemicals were purchased from Sigma Chemical Co.

Neuropathic pain model

Peripheral nerve injury was produced using the CCI model of the sciatic nerve. In brief, the left common sciatic nerve of each rat was exposed at the mid-thigh level. Approximately 7 mm of the nerve proximal to the trifurcation of sciatic nerve was separated from adhering tissue, and four ligatures (4-0 chronic gut) were tied loosely around it with ~1 mm between ligatures. Animals in the sham group received identical surgery but without nerve injury.

Assessment of peripheral nerve injury-related pain behaviours

Mechanical allodynia was determined by measuring incidence of foot withdrawal in response to mechanical indentation of the plantar surface of each hindpaw with a sharp, cylindrical probe with a uniform tip diameter of ~0.2 mm provided by an Electro Von Frey (ALMEMO 2390-5, Anesthesiometer IITC, Inc.), using a protocol similar to the one previously described (Liu et al., 2013). The probe was applied to 10 designated loci distributed over the plantar surface of the foot. The minimal force (in grams) that induced paw withdrawal was read off of the display. The threshold of mechanical withdrawal for each rat was calculated by averaging the 10 readings, and the force was converted into millinewtons. For the results expressing mechanical allodynia, the values are mean values of ipsilateral feet. The rats were tested on each of three successive days prior to surgery. To examine the immediate effect of drugs and vehicles on the persistence of tumour pain, tests were conducted daily on Days 14−18 postoperative (2 h after drugs were given) and at 0.5, 2, 4, 8, 24 h after a single drug injection on Day 14 postoperative.

Thermal hyperalgesia was assessed by measuring foot withdrawal latency to heat stimulation using a protocol as previously described (Wang et al., 2005; Song et al., 2003, 2006). An analgesia metre (model 37370; Ugo Basile Biological Instruments) was used to provide a heat source. In brief, each rat was placed in a box containing a smooth, temperature-controlled glass floor. The heat source was focused on a portion of the hindpaw, which was flush against the glass, and a radiant thermal stimulus was delivered to that site. The stimulus shut off when the hindpaw moved (or after 25.1 sec to prevent tissue damage). The intensity of the heat stimulus was maintained constant throughout all experiments. The elicited paw movement occurred at latency between 9 and 14 sec in control animals. Thermal stimuli were delivered three times to each hindpaw at 5- to 6-min intervals.

Western blot

To identify the phosphorylated levels of proteins, whole protein samples were analysed. In brief, samples (sciatic nerve or spinal cord segments at L1−L6) were collected and washed with ice-cold phosphate buffered saline (PBS) before being lysed in radio immunoprecipitation assay lysis buffer. Then, whole sample lysates were separated by SDS-PAGE and electrophoretically transferred onto PVDF membranes (Millipore Corp.). The membranes were blocked with 5% bovine serum albumin (BSA) for 1 h at room temperature, probed with primary antibodies overnight at 4°C and then incubated with horseradish peroxidase-coupled secondary antibodies (Chemicon). The primary antibodies used include pAMPK (Thr172), 1:1000; pNR1 (Ser896), 1:1000; pNR2B (Tyr1472), 1:1000; pERK1/2 (Thr202/Tyr204), 1:1000; p-p38 (Tyr182), 1:1000; GAPDH, 1:5000; pPKC (Thr514), 1:1000; IBA-1, 1:200. The filters were then developed by enhanced chemiluminescence reagents (PerkinElmer). Data were analysed with the Molecular Imager (Gel DocTM XR, 170-8170) and the associated software Quantity One-4.6.5 (Bio-Rad Laboratories).

Immunofluorescence

Mice under deep anaesthesia were transcardially perfused with PBS followed by 4% paraformaldehyde. L4 and/or L5 lumbar segment was dissected out and post-fixed in the same fixative. The embedded blocks were sectioned for 30-μm thickness. Sections from each group (5 rats in each group) were incubated with IBA-1 antibody (1:200). Then, the free-floating sections were washed with PBS and incubated with the secondary antibody for 2 h. After washing three times with PBS, the samples were studied under a confocal microscope (Olympus FV1000 Confocal System) for the immunofluorescence staining. Images were randomly coded and transferred to a computer for further analysis.

siRNA mediated knockdown of AMPKα

PRKAA1 and PRKAA2 siRNAs (GenePharma) were used for knockdown of AMPKα1 and AMPKα2, respectively. On Day 14 after CCI surgery, 10 μg siRNA was dissolved in 30 μl Lipofectamine 2000 (Invitrogen) according to manufacturer's instructions and was administered by a single peri-sciatic nerve intramuscular. The combination of both siRNAs proved most effective in reducing AMPKα1/α2 expression.

The siRNA sequences were as follows: Prkaa1: GGGAACACGAGUGGUUUAATT, UUAAACCACUCGUGUUCCCTT; Prkaa2: GCAGUGGCUUAUCAUCUUATT, UAAGAUGAUAAGCCACUGCTT; negative control: UUCUCCGAACGUGUCACGUTT, ACGUGACACGUUCGGAGAATT.

Cell preparation and stimulation

RAW264.7 cells were maintained in humidified 5% CO₂ at 37°C in Dulbecco's modified Eagle's Medium supplemented with 10% (v/v) fetal bovine serum. To induce inflammasome activation, 1 × 10⁵ cells were plated in 6-well plate overnight. The medium was changed to serum-free medium in the following morning, and the cells were treated with LPS (1 μg/ml) with or without ozone for 12 h. The air-only treatment was used as the control. Cell extracts and precipitated supernatants were analysed by immunoblotting.

NF-κB activation assay

RAW264.7 cells were plated on coverslips in cell culture dishes and treated with LPS (1 μg/ml) for 12 h with or without ozone. Then, the cells were fixed with ice-cold methanol and were permeabilized with 0.25% Triton X-100/PBS (PBST). After blocking with 1% BSA in PBST for 1 h, the coverslips with RAW264.7 cells were incubated for 2 h at room temperature with the p65/RelA antibody diluted in 1% BSA (1:50). Then, the coverslips were exposed to the FITC-conjugated anti-rabbit IgG (1:100) at room temperature for 1 h and rinsed three times with PBS. Finally, the coverslips were stained with 1 μg/ml 4,6-diamidino-2-phenylindole (DAPI), a fluorescent DNA dye to mark nucleus, for 1 min. Confocal microscopy analysis was carried out using Olympus FV1000 Confocal System.

Statistical analyses

SPSS Rel 15 (SPSS Inc.) was used to conduct all the statistical analyses in this study. The alteration of protein expression and the differences in animal behavioural responses over time were tested with one-way and two-way analysis of variance (ANOVA), respectively, followed by Bonferroni post hoc tests. Results are expressed as mean ± SEM of three independent experiments in triplicates. Results described as significant are based on a criterion of P < 0.05.

Acknowledgements

We thank Dr Xue-Feng Wu (Nanjing University) for her comments on the manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (81471142, 81200860, 81300899) and the Open Project Program of the State Key Laboratory of Natural Medicines, China Pharmaceutical University (SKLNMKF201306).

Conflict of interest: none declared.

Author notes

†

These authors contributed equally to this work.