As a complementary and alternative medicine, EA has been accepted worldwidely, mainly for the treatment of acute and chronic pains. It is well know that EA has a good effect of anti-inflammation and analgesia. From this study, we found that the analgesic effect of EA had the property of a rapid onset, long duration and being strengthened along with EA treatment times, which was consistent with our previous report . Interestingly, it was also found that the role of anti-inflammatory and analgesic effect of EA was not synchronistic. After repeated EA administration, quite a strong difference in the PWTs were found on day 3 between CFA+EA group and CFA group, and statistically significant difference was observed from day 3 to day 21. However, the onset of anti-inflammatory effect of EA appeared later than its analgesic effect. It was not until 14 treatments with EA, we firstly observed the difference in the paw edema of rats in CFA+EA group and that of CFA group. It could be predicted that EA got quicker analgesic effect, maybe through mediating some other way, not or not only by intervening on pro-inflammatory mediators in the treatment of inflammatory pain. Although EA analgesia is well documented, its mechanisms have not been thoroughly clarified. The present study demonstrates, for the first time, that EA suppressed the expression of p-p38 MAPK and its downstream p-ATF-2 as well as gene and protein expressions of VR-1 in the spinal dorsal horn, but had no inhibitory effect on gene and protein expressions of COX-2. Therefore, inhibition of the p38 MAPK activation may be the one of central pain sensitization mechanisms of EA analgesia for inflammatory pain.
The MAPK family includes p38 MAPK, extracellular signal regulated kinase (ERK1/2), c-jun N-terminal kinase (JNK), ERK7/8, ERK3/4 and ERK5 . Initiation of the p38 MAPK cascade involves activation through a classic MAPK kinase kinase (MAP3K) – MAP kinase kinase (MKK) pathway. The p38 MAPK pathway is activated by a wide range of environmental or cellular stresses and proinflammatory cytokines and plays important roles in cell death, neurodegeneration, and inflammation . Recently, p38 MAPK activation was shown to contribute to nociceptive responses in the spinal dorsal horn and dosal root ganglion (DRG) following inflammation and/or nerve injury. Following nerve injury, p-p38 MAPK levels sequentially increased in neurons, microglia, and astrocytes of the spinal dorsal horn; In addition, nerve injury-induced p-p38 MAPK occurred early and was long lasting [24–27]. Moreover, intrathecal injection of p38 MAPK inhibitors attenuated nociceptive response in different neuropathic pain animal models [25, 28–31]. Several reports also indicated that a very robust increase of p-p38 MAPK in spinal dorsal horn induced hindpaw inflammation [4, 6, 32–34]. For example, bee-venom induced rapid increase in p-p38 MAPK in the spinal: p-p38 MAPK began to increase at 1 hour, reached a peak on day 3, and then decreased to normal level on day 7 . In particular, different laboratories have shown that p-p38 MAPK was increased in spinal cord by CFA , formalin , capsaicin , or carrageenan . Besides, intrathecal administration of the p38 inhibitor prevented inflammation-induced thermal and mechanical hypersensitivity [4, 32]. In present study, we demontrated that the numbers of p-p38 MAPK-IR cells in spinal dorsal horn increased and reached at peak at day 3 after CFA injection, maintained for 14 days. The level of p-p38 MAPK protein expression in CFA group have returned to the base at day 21 after CFA-injection, however, inflammation and pain were still existed at that moment. The reason for this is unclear and is worthy of further study.
It is indicated that p38 MAPK is involved in the control of VR-1 mediated glioma apoptotic cell death . Another study reports that p38 MAPK activated in the DRG following peripheral inflammation, by increasing VR-1 levels in nociceptive peripheral terminals, contributes to the maintenance of inflammatory pain . P38 MAPK can potentially regulate protein expression in several different ways, activating transcription factors such as eukaryotic translation initiation factor 4E (eIF4E), Ets likely kinase (ELK-1), and cAMP response element binding protein (CREB) to increase transcription [36–38], increasing mRNA stability  or increasing translation. One major target of p38 MAPK is the translation factor ATF-2. ATF-2 (originally called CRE-BP1) is a member of the ATF/CREB family of transcription factors and is characterized by a basic zipper domain that consists of basic amino acids and a leucine zipper region, which acts as a DNA-binding region. The ATF-2 subfamily contains three members: ATF-2, CRE-BPa, and ATF-7 (originally known as ATF-a) . ATF-2 regulates gene expression via the binding of ATF-2, as a homodimer, to the recognition sequences of the cAMP-response element (CRE) or via formation of a heterodimer with activating protein-1 (AP-1) that binds to CRE sequences. Thus, ATF-2 can interact with other members of the ATF family and with members of the AP-1 family . It has been shown that p-p38 MAPK is involved in regulation of ATF-2 , which then binds to DNA and turns on the target gene . In current report, we found that the numbers of p-p38 MAPK-IR cells at the same time coincided with the numbers of p-ATF-2-IR cells in spinal dorsal horn. So we presumed that ATF-2 as the downstream of p38 MAPK may play an important role in inflammatory pain.
It is now realized that a certain members of the transient receptor potential (TRP) receptors are key molecular integrators of the initiation and maintenance of pain. Vanilloid receptor-1 (VR-1), known as the transient receptor potential ion channel TRPV1, is essential for generating and transmitting pain [44, 45]. It has been found to be expressed within those components of the peripheral and central nervous systems involved in pain detection, transmission and regulation . It is all known that VR-1 can be found in a large population of primary sensory neurons, within the spinal cord itself, and within the brain. It has been reported that deletion of VR-1 is important for mediating thermal hyperalgesia and inflammatory swelling after CFA-induced inflammation . This study was supported by other complementary studies which showed that the severity of arthritis in rodent models was reduced when VR-1 was either blocked or deleted . VR-1 also plays an important role in mechanical hyperalgesia of inflammatory pain. When wild-type and VR-1 null mice are injected in one knee joint with CFA, the VR-1 null mice develops mild joint swelling (evidencing edema) and reduced mechanical hypersensitivity as compared with wild-type animals . In present study, periphery inflammation induced a high expression of VR-1 in ipsilateral spinal dorsal horn on 14 day. EA could down-regulate the expressions of both VR-1mRNA and protein in spinal dorsal horn with increased PWTs.
Several factors have verified that p-p38 MAPK could up-regulate many proinflammatory mediators such as COX-2, IL-1β and VR-1 [6–8]. It has been demonstrated that local induction of COX-2 at the site of peripheral and the subsequent release of prostaglandin E2 (PGE2) has major roles in peripheral pain sensitization, which alters the threshold and excitability of the nociceptive peripheral terminal . In addition to this peripheral action, COX-2 and prostaglandin also have a central function. The central PGE2 produced by COX-2 induced an electrophysiological transmitter release, direct depolarization of postsynaptic membranes and inhibiting glycine receptor activation . The upregulation of central COX-2 level, induced by peripheral inflammatory stimuli such as CFA  and carrageenan , was linked to inflammatory allodynia and hyperalgesia. However, COX-2 just plays an important role in the development, but not maintenance of inflammation. Spinal COX-2 mRNA increased to a peak at 4 h after a intraspinal injection with IL-1α  and similarly increased to a peak at 4 h  or 8 h  after mechanical injury to the spinal cord. The expression of COX-2 protein also increased in the spinal cord during hindpaw inflammation at 4–6 h induced by CFA, paralleled the COX-2mRNA, and returned to baseline within 3 d after induction of inflammation . According to another report, the expression of COX-2 mRNA on the ipsilateral side of spinal cord obtained from CFA injection was significantly increased at both 6 hour and 3 day after CFA injection compared with that from the saline-treated mice . In agreement with above studies, we detected that the level of COX-2 mRNA in ipsilateral spinal dorsal horn was increased respectively at 5 hour and 3 day (Additional file 1: Figure S5), and returned to baseline within 14 day after CFA-induction of inflammation. Whereas, EA treatment did not suppress the increase of COX-2 mRNA expression in spinal dorsal horn at 5 hour, 3 day and 14 day.
As we mentioned above, robust p38 MAPK activation was essential for VR-1 and COX-2 expression in SCDH after peripheral inflammation, and it could regulate ATF-2 phosphorylation in SCDH of CFA-induced inflammation. Our previous report has demonstrated that EA analgesia and anti-inflammation were associated with its inhibition of spinal p38 MAPK activation . In agreement with our previous study, pre-eletroacupuncture treatment has prophylactic analgesic effects on rats suffering from visceral pain by suppressing the local and spinal p38 MAPK . The p38 MAPK signal transduction pathway is partly involved in the regulatory mechanism of this analgesic effect. Another study has reported that EA could improve immune suppression involving in the signaling pathway of p38 MAPK . In this study, we first systemically investigated whether EA treatment might regulate the downstream materials of p38 MAPK. We found that EA did suppress the activation of the p38 MAPK and its downstream p-ATF-2 as well as VR-1mRNA and protein, but had no influence on gene and protein expressions of COX-2 at day 14 after CFA injection. Similarly, this study showed significant analgesic effect of EA at day 14 after CFA-injection, as well as a relative weak anti-inflammatory effect. PWTs of CFA+EA group had no difference with that of control group on day 14 in CFA-induced inflammatory pain, but the percent paw swelling of CFA+EA group had remarkable difference with that of control group at the same time. It could be suggested that both COX-2 and VR-1 play essential roles in inflammatory pain, however, EA just regulates the expressions of VR-1 mRNA and protein, and does not have obvious effect on COX-2 mRNA and protein expressions. It is well known that COX-2 mainly regulates the release of inflammatory mediators, leading to the generation of inflammatory pain, and VR-1 enhances the excitability of nerve resulting in pain [59, 60]. Therefore, we speculate that the analgesic effect of EA, different from the mechanism of NASIDs, is not reached by inhibiting the release of inflammatory cytokines but rather by regulating pain sensitization, because we have not observed a suppressing effect of EA on COX-2 mRNA and protein, but an inhibitory effect on p38 MAPK, ATF-2 and VR-1 expressions.