Introduction

There is a type of pain that does not go away even though the tissue has already healed. One type of this pain is called neuropathic pain that typically develops when peripheral nerves are damaged such as through surgery, bone compression in cancer, diabetes, or infection. Neuropathic pain is a major factor causing impaired quality of life in millions of people worldwide and is frequently resistant to all known analgesic drugs. Over the last decade, accumulating evidence concerning how peripheral nerve injury creates neuropathic pain has suggested that nerve injury produces molecular and cellular alterations that result in multiple forms of neuronal plasticity and anatomical reorganization in the dorsal horn of the spinal cord. These alterations have been proposed to be crucial in the pathogenesis of neuropathic pain [1, 2]. While the dominant theme in research on neuropathic pain has been to understand the roles of neurons in the peripheral nervous system and the dorsal horn, there is a rapidly growing body of evidence indicating that spinal glial cells play a critical role in the pathogenesis of neuropathic pain.

Recently, growing evidence has indicated that neuron-glia interaction is a key idea to understand functions of the central nervous system (CNS). Especially glia play important roles in pathophysiological situations of the CNS including psychiatric disorders, physical trauma, and infections [3]. Glia consist of three members: astrocytes, oligodendrocytes, and microglia. Accumulating findings also indicate that nucleotides play an important role in neuron-glia communication through P2 purinoceptors, even though ATP is recognized primarily to be a source of free energy and nucleotides are key molecules in cells. Microglia, which are thought to be residential macrophages in the CNS, express P2 purinoceptors, mainly P2X4 and P2X7 as well as P2Y2, P2Y6, and P2Y12 [4]. Damaged neurons release or leak nucleotides including ATP and UTP to stimulate microglia [5, 6]. It is clear that these nucleotides trigger the release of various neurotoxic and neuroprotective cytokines and growth factors via different purine receptors [4, 7] or induce P2Y12-dependent chemotaxis in cultured cells [8, 9]. It was shown in an animal model of neuropathic pain that microglial P2X4 and P2X7 receptors are crucial in pain signaling after peripheral nerve lesion [10–12]. Other purine receptors are upregulated in microglia in response to neuronal injury, as was recently demonstrated for P2Y6 [6]. The P2Y6 receptor triggers microglial phagocytosis [6]. In this paper, we review the modification of pain sensation through microglial P2X4 and P2X7, and also point out the possibility of P2Y6 and P2Y12 involvement in pain signaling.

P2X4 in neuropathic pain

We found that the marked tactile allodynia that develops following nerve injury was reversed by acutely administering TNP-ATP intrathecally but was unaffected by administering PPADS [10]. From the pharmacological profiles of TNP-ATP (blocking P2X4 at high concentration) and PPADS (not blocking P2X4), it was suggested that tactile allodynia depends upon P2X4 in the spinal cord. The expression of P2X4 protein, normally low in the naïve spinal cord, progressively increased in the days following nerve injury with a time course parallel to that of the development of tactile allodynia. Double immunolabeling analysis demonstrated that not neurons or astrocytes but activated microglia in the dorsal horn were intensely positive for P2X4 protein [10]. Moreover, intrathecally administered antisense oligodeoxynucleotide for P2X4R reduced the expression of P2X4 protein in spinal microglia and prevented the development of the nerve injury-induced tactile allodynia. In naïve rats, intrathecal administration of cultured microglia that were preincubated with ATP to activate P2X4 produced tactile allodynia over the 3–5 h after the administration [10].

Since it was already reported that the nerve injury-induced tactile allodynia depends on a depolarizing shift in the Eanion of spinal lamina I (LI) neurons in the dorsal spinal cord, resulting in converting the GABAA receptor- and glycine receptor-mediated inhibition to excitation [13], it was considered that microglia may affect Eanion in LI neurons. To investigate this possibility, microglia were administered to the lumbar spinal level of naïve rats by an intrathecal catheter as described [11]. Administering microglia stimulated with ATP caused a progressive tactile allodynia over the 5 h after injection. Eanion in LI neurons from rats administered ATP-stimulated microglia was shifted to −61.6 mV from −68.3 mV of normal rats. In addition, we found that GABA response switched from hyperpolarizing in control rats to depolarizing in microglia-treated rats. Activated microglia secrete various biologically active molecules, one of which, BDNF, was implicated in the hypersensitivity of dorsal horn neurons that follows sensitization and inflammation [14–16] and in anion gradient shifts in the hippocampus [17]. Indeed, intrathecal administration of recombinant BDNF produced tactile allodynia comparable to that produced by ATP-stimulated microglia [11]. Eanion of LI neurons in slices treated with BDNF (>90 min, in vitro) was significantly less negative than that of LI neurons from control slices. The rise in [Ca2+]i was prevented by the GABAA receptor blocker bicuculline, confirming that the effect was mediated by GABAA receptors. Thus, acute administration of BDNF in slices caused a depolarizing shift in Eanion and caused GABA to produce net excitation [11]. Moreover, a function-blocking antibody against the TrkB receptor (anti-TrkB) and a BDNF-sequestering fusion protein (TrkB-Fc) acutely inhibited the allodynia and the shift of Eanion of LI neurons [11]. The administration of ATP-stimulated microglia with either anti-TrkB or TrkB-Fc did not develop tactile allodynia. After pretreatment of microglia with double-stranded short interfering RNA directed against BDNF (BDNF siRNA), the ATP-stimulated microglia injected intrathecally into normal rats did not cause the allodynia [11]. Anti-TrkB and BDNF siRNA prevented the shift in Eanion induced by ATP-stimulated microglia. ATP stimulation caused release of BDNF from microglia in culture. This effect of ATP was blocked by treating the cultures with the P2X receptor blocker TNP-ATP. In addition, pretreatment of the microglia with BDNF siRNA prevented release of BDNF by ATP stimulation. By bath application of TNP-ATP to spinal slices taken from allodynic rats 2 weeks after nerve injury, Eanion of LI neurons was returned to normal value [11]. These findings indicate that P2X4-dependent release of BDNF from microglia is necessary to sustain both the tactile allodynia and the depolarizing shift in Eanion in LI neurons that result from nerve injury (Fig. 1).

Fig. 1
figure 1

Schematic illustration of potential mechanisms by which P2X/Y receptors in activated microglia modulate neuropathic pain signaling in the dorsal horn. Activated microglia in the spinal cord after nerve injury express ionotropic ATP receptors [e.g., P2X4 receptor (P2X4R) or P2X7R]. P2X4R or P2X7R activation leads to the release of bioactive diffusible factors such as BDNF and other proinflammatory factors (cytokines and chemokines). BDNF causes a collapse of transmembrane anion gradient in dorsal horn lamina I neurons presumably through the downregulation of KCC2, which in turn renders GABA and glycine effects depolarizing, rather than hyperpolarizing, in these neurons. Microglial factors may also interact with excitatory synapses of neighboring dorsal horn neurons and enhance the excitability in dorsal horn neurons. The net hyperexcitability in the dorsal horn pain network by these factors from activated microglia may be responsible for neuropathic pain. Microglia also express G protein-coupled ATP receptors [e.g., P2Y6 receptor (P2Y6R) and P2Y12R]. Activating P2Y12R and P2Y6R leads to chemotaxis and phagocytosis, respectively, but their functional relevance to neuropathic pain remained to be determined

Full size image

P2X7 in neuropathic pain

P2X7 purinoceptors are a ligand-gated nonselective cationic channel and are expressed predominantly on immune cells [18]. Stimulation of the P2X7 receptors on microglia is associated with release of cytokines including tumor necrosis factor-α (TNF-α) [5, 19], interleukin-6 (IL-6) [20], and interleukin-1β (IL-1β) [21–23]. Several cytokines such as IL-1β, IL-6, and TNF-α in the dorsal horn are increased after nerve lesion [24–26] and have been implicated in contributing to neuropathic pain [24–27]. Recent evidence indicates the relationship between TNF-α and neuropathic pain [28–32], and TNF-α released after injury is proposed as an initiator of abnormal pain sensation. TNF-α is upregulated after nerve injury in both dorsal root ganglion (DRG) and spinal cord [33]. The inhibition of TNF-α reduces the hyperalgesia in neuropathic pain models [34]. Recent evidence indicates the relationship between inflammatory cytokines including IL-1β and neuropathic pain [25, 26, 28]. The expression of IL-1β is upregulated in the spinal cord of several rat neuropathy models [24–26]. These findings further support a role for central IL-1β in the development and maintenance of neuropathic pain through induction of a proinflammatory cytokine cascade (Fig. 1).

Recently, it was reported that in mice lacking P2X7 inflammatory and neuropathic hypersensitivity is completely absent to both mechanical and thermal stimuli, whilst normal nociceptive processing is preserved [12]. Contribution of P2X7 receptor to neuropathic pain is also demonstrated by using the recently developed selective antagonist for P2X7 receptor A-740003 [35]. The knockout animals were unimpaired in their ability to produce mRNA for pro-IL-1β, and cytometric analysis of paw and systemic cytokines from knockout and wild-type animals following adjuvant insult suggested a selective effect of the gene deletion on release of IL-1β and IL-10, with systemic reductions in adjuvant-induced increases in IL-6 and MCP-1. In addition, P2X7 receptors were upregulated in human dorsal root ganglia and injured nerves obtained from chronic neuropathic pain patients [12]. It was hypothesized that the P2X7 receptor plays a common upstream transductional role in the development of pain of neuropathic and inflammatory origin via regulation of mature IL-1β production.

In addition, a recent study has shown that activation of P2X7 receptors expressed on satellite glia that enwrap each DRG neuron leads to a release of TNF-α, which in turn increases the excitability of DRG neurons [36]. P2X7 receptors are upregulated in surrounding satellite glial cells in the DRG in humans [12]. Implication of TNF-α has been reported in neuropathic pain [28–32]. Therefore, the contribution of P2X7 receptor to neuropathic pain might be related not only to the activation of microglial P2X7 receptors, but also to P2X7 receptors on other cell types such as satellite glia.

P2Y6 and P2Y12 in neuropathic pain?

When neurons are injured or dead, microglia are activated, resulting in their interaction with immune cells, active migration to the site of injury, release of proinflammatory substances, and the phagocytosis of damaged cells or debris. For such activation of microglial motilities, extracellular nucleotides have a central role. Extracellular ATP functions as a chemoattractant. Microglial chemotaxis by ATP via P2Y12 receptors was originally found by Honda et al. [8] and has recently been confirmed in vivo in P2Y12 receptor knockout animals [37]. Neuronal injury results in the release or leakage of ATP that appears to be a “find-me” signal from damaged neurons to microglia to cause chemotaxis. In addition to microglial migration by ATP, another nucleotide, UDP, an endogenous agonist of the P2Y6 receptor, greatly activates the motility of microglia and orders microglia to eat damaged neurons. UDP does not cause chemotaxis, but instead causes phagocytosis by microglia [6]. Phagocytosis, a specialized form of endocytosis, is the uptake by the cell of relatively large particles (>1.0 μm) into vacuoles and has a central role in tissue remodeling, inflammation, and the defense against infectious agents [38]. Phagocytosis is initiated by the activation of various cell surface phagocytosis receptors, including Fc receptors, complement receptors, integrins, endotoxin receptors (CD18, CD14), mannose receptors, and scavenger receptors [39], which are activated by corresponding extracellular ligands. In the CNS, a full innate immune system, i.e., Fc receptors, complement system, scavenger receptors, and Toll-like receptors etc., has been described, and microglia reveal related roles as dedicated phagocytes. Since recognition is the first and the most important step for phagocytosis, extensive studies on phagocytosis receptors have been reported. It is well-known that dying cells express so-called eat-me signals such as phosphatidylserine on their surface membrane [39], by which microglia recognize the apoptotic cells to catch and remove them [39]. We first found that exogenously applied UDP caused microglial phagocytosis in a concentration-dependent manner, which was P2Y6 receptor-dependent. We found that neuronal injury caused by kainic acid (KA) upregulated P2Y6 receptors in microglia, the KA-evoked neuronal injury resulted in an increase in extracellular UTP, which was immediately metabolized into UDP in vivo and in vitro. Moreover, UDP leaked from injured neurons caused P2Y6 receptor-dependent phagocytosis in vivo and in vitro. These results suggest that UDP could be a molecule that signals the crisis of damaged neurons to microglia, triggering phagocytosis. It should be noted that nucleotides could be both “find-me” and “eat-me” signals. Cells release ATP, and we also found that KA caused an increase in extracellular UTP/UDP. Therefore microglia might be attracted by ATP/ADP [8, 40, 41] and subsequently recognize UDP, leading to the removal of the dying cells and their debris.

Are P2Y6 and P2Y12 receptors involved in neuropathic pain? For this question, there is a report indicating that the UDP-sensitive P2Y6 receptor produces inhibitory effects on spinal pain transmission in a neuropathic pain model [42]. In the neuropathic pain model, in which the sciatic nerves of rats were partially ligated, UDP (30 and 100 nmol/rat) produced significant antiallodynic effects. UDP (100 nmol/rat) caused no motor deficit in the inclined plane test [42]. The mechanism of this UDP-evoked inhibitory effect on neuropathic pain is unknown. There is no evidence to show the site (neuron or glia) of action of UDP. We are now investigating the involvement of P2Y12 in pain signaling and have obtained evidence suggesting the role of P2Y12 in the neuropathic pain state.

Conclusion

Neuropathic pain is a major factor causing impaired quality of life in millions of people worldwide. We have to try to reveal the mechanism of this pain in order to develop effective drugs against the pain. Recently, there has been a rapidly growing body of evidence indicating that spinal glia, especially microglia, play a critical role in the pathogenesis of neuropathic pain and that nucleotides play an important role in neuron-glia communication through P2 purinoceptors. We described the modification of neuropathic pain signaling through microglial P2X4 and P2X7, with the possibility of P2Y6 and P2Y12 involvement (Fig. 1). It was shown in an animal model of neuropathic pain that microglial P2X4 and P2X7 receptors are crucial in pain signaling after peripheral nerve lesion. Microglial P2Y6 and P2Y12 play very interesting roles for phagocytosis and chemotaxis, respectively. Besides, there is a possibility that these receptors are involved in pain sensation. Additional experiments are needed to clarify the possibility.