Volume 98, Issue 2 p. 372-384
Free Access

The glutamatergic system as a target for neuropathic pain relief

Maria Osikowicz

Maria Osikowicz

Department of Pain Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Krakow, Poland

Search for more papers by this author
Joanna Mika

Joanna Mika

Department of Pain Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Krakow, Poland

Search for more papers by this author
Barbara Przewlocka

Barbara Przewlocka

Department of Pain Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Krakow, Poland

Search for more papers by this author
First published: 04 October 2012
Citations: 87
B. Przewlocka: Department of Pain Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Krakow, Poland. Email: [email protected]

New findings

  • What is the topic of this review?

    This paper presents a review of the literature of glutamate receptors and transporters in neuropathic pain and the role of glia in these effects. Specifically, pharmacological interventions aimed at inhibiting group I mGluRs and/or potentiating group II and III mGluR-mediated signalling is discussed.

  • What advances does this highlight?

    Recent discoveries show that metabotropic glutamate receptors play a role in neuropathic pain, and give substantial evidence of glial participation in these effects.

Glutamate is the major excitatory neurotransmitter in the mammalian CNS. The understanding of glutamatergic transmission in the nervous system has been greatly expanded with the discovery and investigation of the family of ionotropic and metabotropic glutamate receptors (mGluRs). Metabotropic glutamate receptors are localized at nerve terminals, postsynaptic sites and glial cells and thus, they can influence and modulate the action of glutamate at different levels in the synapse. Moreover, there is substantial evidence of glial participation in glutamate nociceptive processes and neuropathic pain. Metabotropic glutamate receptors have been shown to play a role in neuropathic pain, which is one of the most troublesome illnesses because the therapy is still not satisfactory. Recently, the development of selective mGluR ligands has provided important tools for further investigation of the role of mGluRs in the modulation of chronic pain processing. This paper presents a review of the literature of glutamate receptors in neuropathic pain and the role of glia in these effects. Specifically, pharmacological interventions aimed at inhibiting group I mGluRs and/or potentiating group II and III mGluR-mediated signalling is discussed. Moreover, we introduce data about the role of glutamate transporters. They are responsible for the level of glutamate in the synaptic cleft and thus regulate the effects of all three groups of mGluRs and, in consequence, the activity of this system in nociceptive transmission. Additionally, the question of how the modulation of the glutamatergic system influences the effectiveness of analgesic drugs used in neuropathic pain therapy is addressed.

Recently, the role of glutamatergic pathways in nociceptive processes has been associated with different types of pain (Smith et al. 1994, 2002; Baron, 2000; Devor, 2001; Jensen & Baron, 2003). The level of glutamate is elevated in the spinal cord of rats during inflammation (Dmitrieva et al. 2004; Pitcher et al. 2007) and following nerve injury in neuropathic pain (al-Ghoul et al. 1993; Hudson et al. 2002). The role of ionotropic glutamatergic receptors in the mechanisms of neurotransmitter release, the transmission of nociceptive stimuli, morphine analgesia and tolerance has been well demonstrated in many studies. However, severe side-effects diminished interest in this direction of research (Eide et al. 1995; Jevtovic-Todorovic et al. 1998) and led to the study of the therapeutic application of metabotropic glutamate receptor (mGluR) ligands and glutamate transporters (Budai & Larson, 1998; Neugebauer & Carlton, 2002; Chiechio et al. 2004).

Elevated glutamatergic neurotransmission in the CNS is observed during neuropathic pain, which is accompanied by lower effectiveness of opioid antinociceptive drugs and often, a lower mood of patients, sometimes even depression. The modulation of the activity of the glutamatergic system could be beneficial for the potentiation of an analgesic effect of opioid drugs in neuropathic pain and possibly also other drugs used in therapy of neuropathic pain, such antidepressants. Recently, it was shown that opioid-induced hyperalgesia in neuropathic pain could be inhibited by upregulation of spinal glutamate transporter-1 (GLT-1) expression (Chen et al. 2012). Moreover, the effect of other drugs used in neuropathic pain therapy could be potentiated by the use of mGluR ligands and/or inhibitors of excitatory amino acid transporters (EAATs). Thus, substances that lead to a decrease in glutamatergic neurotransmission appear to be an attractive target for the future therapy of neuropathic pain, including other aspects of the illness in addition to pain, such as depression. Some data strongly suggest that negative allosteric modulators [2-Methyl-6-(phenylethynyl)pyridine (MPEP) and 3-((2-Methyl-4-thiazolyl)ethynyl)pyridine (MTEP) and fenobam] of the metabotropic glutamate 5 (mGlu5) receptor display analgesic and antidepressant effects (Osikowicz et al. 2008; Montana et al. 2009). Interestingly, group III mGluR ligands, both agonists and antagonists, also possess both analgesic and antidepressant-like potential, although selective receptor subtype substances that can penetrate the brain are necessary to establish their possible usefulness in the therapy of neuropathic pain. The cellular distribution of mGluRs in the CNS and the importance of glial activity favour the use of combined pharmacology to reduce nociception at different levels of the pain transmission and modulation pathway (Fundytus et al. 1998; Karim et al. 2001). This review of recent knowledge concerning metabotropic glutamate receptors and transporters includes discussion of glial cells involvement in glutamate effects in neuropathic pain, as well as the possible beneficial action of glutamate system on drugs used in the therapy of neuropathic pain.

Glutamate system

Glutamate is a widely distributed excitatory neurotransmitter in the mammalian nervous system. It participates in all functions of the nervous system and affects nervous system development at all stages, from neuron migration, differentiation and death to the formation and elimination of synapses. Previous studies have described the various glutamate transporters and numerous receptors for glutamate (Fig. 1). Glutamate receptors are broadly classified as ionotropic glutamate receptors and metabotropic glutamate receptors.

Details are in the caption following the image

Classification of glutamate transporters and receptors

Glutamate receptors Ionotropic glutamate receptors have been identified in the CNS based on their pharmacology and subsequently through molecular biology (Dingledine et al. 1999; Meldrum, 2000; Tzschentke, 2002). Their names are based upon the pharmacological agonist that binds to the specific receptor subtype and selectively opens the associated ion channel, as follows: the N-methyl-d-aspartate (NMDA) receptor; the kainic acid (KA) receptor; and the α-amino-3-hydroxy-5-methyl-4-isoxazole proprionate receptor (AMPA or non-NMDA receptor; Dingledine et al. 1999; Meldrum, 2000; Tapiero et al. 2002). The ionotropic receptors appear to be tetrameric or pentameric, and the subunits that comprise these receptors are specific for each of the three families (Dingledine et al. 1999; Meldrum, 2000). These receptors play critical roles in the basic transmission of nervous signals; they initiate neuroplastic changes in the CNS and are responsible for many diseases, including chronic pain.

Metabotropic glutamate receptors are a family of seven transmembrane domain G-protein-coupled receptors. Eight mGluR subtypes have been cloned and classified, as follows (see Fig. 2): group I (mGlu1 and mGlu5); group II (mGlu2 and mGlu3); and group III (mGlu4 and mGlu 6–8). The mGluRs are classified based on their sequence and homology, signal transduction mechanisms and pharmacological profile (Hollmann & Heinemann, 1994; Schoepp et al. 1999; Carlton, 2001; Fundytus, 2001; Karim et al. 2001; Neugebauer, 2001; Gasparini et al. 2002).

Details are in the caption following the image

Ligands of metabotropic glutamate receptors (mGluRs)

A main distinction between group I and groups II and III mGluRs is the different synaptic distribution of these receptors, which is also a major factor that determines their function. Group I receptors are primarily localized postsynaptically, apart from the synaptic active zone (Shigemoto et al. 1997), which suggests that the activation of group I mGluRs occurs mostly through high concentrations of glutamate in the synaptic space; however, there are also reports of presynaptic localization of group I mGluRs (Moroni et al. 1998). Group I mGluRs are positively coupled to phospholipase C through G-proteins of the Gq/G11 type. Activation of group I receptors leads to stimulation of phospholipase C, production of inositol trisphosphate, release of Ca2+ from intracellular stores and production of diacylglycerol, which in turn activates protein kinase C (Masu et al. 1991; Abe et al. 1992; Kew & Kemp, 2005).

In contrast, group II and III mGluRs are localized presynaptically and are therefore regarded as autoreceptors (Ohishi et al. 1995; Petralia et al. 1996a). The group III mGluRs are located at the synaptic active zone, while group II mGluRs are located on the outskirts or even outside of synapses on the neurons and glial cells (especially mGlu3 receptors; Petralia et al. 1996b). Group II and III mGluRs are negatively coupled to adenylyl cyclase activity through G-proteins of the Gi/Go type (Pin & Duvoisin, 1995). The activation of presynaptic mGluRs localized on glutamatergic nerve terminals causes a decrease in glutamate release both in vitro (Attwell et al. 1995) and in vivo (Battaglia et al. 1997); therefore, the stimulation of presynaptic group II (mGlu2 and mGlu3) and group III (mGlu4 and mGlu6–8) autoreceptors may produce functional antagonism to the glutamatergic system (Gerber et al. 2000) and may thus attenuate pain.

Glutamate transporters Excitatory amino acid transporters, also known as glutamate transporters, belong to the family of neurotransmitter transporters. The EAATs serve to terminate the excitatory signal by removal (uptake) of glutamate from the neuronal synapse into glia and neurons. Five glutamate transporter proteins have been identified, of which EAAT3 (EAAC1) is found in neurons (Masson et al. 1999; Attwell, 2000; Meldrum, 2000), while EAAT1 (glutamate/aspartate transporter; GLAST) and EAAT2 (GLT-1) are expressed in glial cells (Fig. 3). Both glial and neuronal glutamate transporters play an important role in various physiological functions, such as plasticity, glutamate receptor activation, and the maintenance and duration of extracellular glutamate concentrations (Fig. 4) (Sung et al. 2003; Binns et al. 2005). Transporter-mediated glutamate homeostasis fails dramatically after injury of the CNS; instead of removing extracellular glutamate to protect neurons, transporters release glutamate, triggering neuronal death (Rossi et al. 2000).

Details are in the caption following the image

The uptake of glutamate (Glu) by excitatory amino acid transporter (EAAT) glutamate transporters from the synaptic cleft and its metabolism to glutamine (Gln) in glial cells

Details are in the caption following the image

Synaptic distribution of mGluRs
In general, group I mGluRs (shown in blue) are localized postsynaptically, whereas group II (shown in red) and III receptors (shown in green) are present in presynaptic locations, although exceptions occur. In presynaptic parts, mGluRs 2, 3, 4 and 8 are generally found in extrasynaptic locations, and mGlu7 receptor is highly localized to the active zone. At the postsynaptic terminal, the glutamate-gated ion channels (NMDA) respond to glutamate with increases in intracellular calcium, promoting cell excitability. Postsynaptic mGlu3 receptors are coupled to cAMP inhibition (–). Group II and III mGluRs are found in GABAergic terminals, where they inhibit release of GABA (–). Expression of mGlu3 and mGlu5 recptors on glia is now emerging as another site for mGluR regulation of synaptic activity, though the signalling pathways and consequences of receptor activation on these cells are not currently well understood.

Vesicular glutamate transporter (VGLUT) is responsible for the active transport of l-glutamate into synaptic vesicles; therefore, it is essential for glutamatergic transmission in the peripheral and central nervous system. Vesicular glutamate transporter is also expressed and localized in various secretory vesicles in non-neuronal peripheral organelles and can act as a paracrine and/or autocrine modulator to regulate cellular functions (Moriyama & Omote, 2008). Li et al. (2003) have shown that in laminae I–II of the medullary and spinal dorsal horns, both VGLUT1 and VGLUT2 are expressed in axon terminals of primary afferent fibres, including substance P-containing nociceptive fibres, and that VGLUT2 is primarily located in unmyelinated primary afferent fibres terminating in lamina II (Li et al. 2003).

The involvement of the glutamatergic system in acute and chronic nociceptive processes, especially neuropathic pain

Glutamate receptors The antagonists of ionotropic glutamate receptors, including antagonists of AMPA (2,3-benzodiazepine), kainate (LY382884, LY466195 and NS3763), AMPA/KA (NBQX and CNQX) and NMDA (ketamine, MK-801, dichlorokynurenic acid, L-701,324 and GV196771), decrease nociceptive transmission (Bleakman et al. 2006), but because of their side-effects, the main interest of researchers has been focused on mGluRs. For several years, numerous pharmacological tools have been available, such as orthosteric ligands that can either directly activate (agonists) or block (antagonists) specific mGluR subtypes. In addition, allosteric ligands that positively or negatively modulate mGluR function are also available (Fig. 2). Several behavioural (Fundytus et al. 1998; Zhou et al. 2001; Dolan et al. 2003) and electrophysiological studies (Young et al. 1997) have demonstrated a specific role for group I mGluRs in nociceptive processing in the CNS. The administration of a group I mGluR agonist [(S)-3,5-Dihydroxyphenylglycine (3,5 DHPG)] induces spontaneous nociceptive behaviours (Bhave et al. 2001; Lorrain et al. 2002), whereas group I mGluR antagonists, such as MPEP and MTEP, produce antiallodynic and antihyperalgesic effects in different animal models of pain (Bhave et al. 2001; Walker et al. 2001a,b; Zhu et al. 2004; Osikowicz et al. 2008; Ren et al. 2012). Targeting mGlu5 receptors has been shown to elicit analgesia in different models of persistent and neuropathic pain, as observed with the selective antagonists MPEP (Urban et al. 2003; Osikowicz et al. 2008, 2009), SIB-1757 (Dogrul et al. 2000), full non-competitive antagonist (NAM) and fenobam (Montana et al. 2009). Recently, the new negative allosteric antagonists of the mGlu5 receptor, NAM and ADX10059, have been proposed for the treatment of migraines (Marin & Goadsby, 2010).

In contrast, the role of group II and III mGluRs in nociceptive processing is less well established. In contrast to group I mGluR agonists (3,5-DHPG), the application of the group II (2S,1′S,2′S)-2-(Carboxycyclopropyl)glycine (l-CCG-I) and group III agonists [L-(+)-2-Amino-4-phosphonobutyric acid (l-AP4)] does not produce excitation of dorsal horn neurons (Young et al. 1997) and does not have any behavioural nociceptive effects in naive animals (Fisher & Coderre, 1996a,b). In conditions of tonic pain, however, group II and III mGluR agonists have been shown to inhibit dorsal neuronal responses after injection of carrageenan into the rat hindpaw (Stanfa & Dickenson, 1998) and to produce antinociceptive effects in the rat formalin test (Fisher & Coderre, 1996b), respectively. The activation of group II mGluRs by several mGlu2/3 agonists, such as LY354740, LY379268 and LY389795, has been shown to induce analgesia in inflammatory and neuropathic pain models (Fisher & Coderre, 1996b; Sharpe et al. 2002; Simmons et al. 2002; Zammataro et al. 2011). Group II mGluR agonists [both the mGlu2 and mGlu3 receptor; LY379268, (2R,4R)-4-Aminopyrrolidine-2,4-dica rboxylate (APDC) and N-Acetyl-L-aspartyl-L-glutamic acid (NAAG)] have been shown to decrease the development of inflammatory and neuropathic pain symptoms, such as allodynia and hyperalgesia (Mills et al. 2002; Yamamoto et al. 2004; Chen & Pan, 2005; Osikowicz et al. 2008).

Recent evidence suggests that group III mGluRs, such as mGlu4 and mGlu6–8 receptors, regulate pain transmission and are potential targets for analgesic drugs (Goudet et al. 2008). However, the individual role of mGlu4 and mGlu6–8 receptors in the regulation of the pain threshold is not unequivocal. It has been documented that the group III mGluR non-selective agonist l-AP4 and the selective mGlu7 recptor agonist AMN082 decrease the development of inflammatory and neuropathic pain symptoms, such as allodynia and hyperalgesia (Mills et al. 2002; Yamamoto et al. 2004; Chen & Pan, 2005; Osikowicz et al. 2008). The intrathecal injection of a positive allosteric modulator of mGlu8 receptor N-Phenyl-7-(hydroxyimino)cyclopropa [b]chromen-1a-carboxamide (PHCCC) also causes analgesia in models of inflammatory or neuropathic pain (Goudet et al. 2008), suggesting that mGlu8 receptors in the spinal cord attenuate pain transmission. Interestingly, the spinal administration of the positive allosteric modulator of mGlu4 receptors (VU0155041) attenuated hyperalgesia in a rat model of neuropathic pain, whereas the positive allosteric modulator of mGlu7 receptors (AMN082) did not.

In summary, the recent availability of subtype-selective ligands has shown that group III mGluRs play distinct and sometimes opposite roles in nociception depending on the site of activation and pain modality (Chen & Pan, 2005). It has also been demonstrated that the effects of mGluR ligands in alleviating neuropathic pain symptoms differ depending on the site of administration, e.g. antinociceptive effects were greater after systemic drug delivery (Osikowicz et al. 2008), because a higher number of mGlu5, 2/3 and 7 receptors localized not only on the spinal but also on the supraspinal and peripheral neurons might have been affected (Bond et al. 2000; Tamaru et al. 2001; de Novellis et al. 2003; Mitsukawa et al. 2005).

Glutamate transporters Excitatory amino acid transporters play in important role in the effects of glutamate (Danbolt, 2001; Chen et al. 2004; Gras et al. 2012; López-Redondo et al. 2000). Recent evidence has shown that a deficiency or downregulation of GLT-1 or GLAST in the spinal dorsal horn is associated with the development of neuropathic pain induced by peripheral nerve injury (Sung et al. 2003; Binns et al. 2005). Sung et al. (2003) reported that after injury to the sciatic nerve in rats, GLAST and GLT-1 expression was upregulated during days 1–4 postoperatively but downregulated during days 7–14. A marked reduction in glutamate uptake activity was also observed in the dorsal and ventral horn 4–6 weeks after nerve injury in the spinal nerve ligation model (Sung et al. 2003). The mechanism of spinal glutamate uptake downregulation is still not clear, but it is thought to contribute to the central mechanisms of nociception (Sung et al. 2003; Binns et al. 2005). Recent studies have indicated that the GLAST and GLT-1 transporter inhibitor dl-threo-β-benzyloxyaspartate (TBOA) and the selective GLT-1 inhibitor dihydrokainate (DHK) induce analgesia in neuropathic pain (Weng et al. 2007). The mechanism of spinal glutamate uptake is still not clear, but it is thought to contribute to the central mechanisms of nociception (Sung et al. 2003; Binns et al. 2005).

In pain states, sensory neurons activate dorsal spinal neurons through the release of glutamate, which is stored in synaptic vesicles by VGLUT transporters. The myelinated Aβ fibres terminate in layers III–IV and II of the spinal cord; these fibres are responsible for touch sensation and contain VGLUT1. The myelinated Aδ fibres and non-myelinated C fibres transmit the majority of mechanothermal impulses and end in layers I and II, and they contain VGLUT2. Polymodal C fibres terminate in layer I and on the border between layers II–III and contain VGLUT3 (Drew & MacDermott, 2009; Leo et al. 2009). Studies in knock-out mice confirmed that VGLUT2 and VGLUT3 transporters were important modulators of neuropathic pain. VGLUT1−/− knockout mice survived for only 3 weeks, whereas VGLUT1+/− knockouts developed neuropathic pain in the normal manner. VGLUT2−/− knockout mice died immediately after birth, whereas VGLUT2+/− knockouts did not develop mechanical allodynia or hypersensitivity to low temperatures. VGLUT3−/− knockout mice, like normal mice, developed hypersensitivity to high temperatures but did not develop hypersensitivity to mechanical stimuli (Drew & MacDermott, 2009). Summing up, the involvement of VGLUT2 and VGLUT3 in the nociceptive signalling seems to be very important, especially in neuropathic pain.

Glial participation in glutamatergic effects during neuropathic pain

Glutamate receptors Glial activation is a typical reaction to injury of the peripheral and central nervous system. Many studies have shown an increase in the immunoreactivity of both microglia and astrocytes in the ipsilateral spinal cord segment (Mika et al. 2009; Osikowicz et al. 2009; Wen et al. 2011). Astrocytes activated after nerve injury release glutamate, which enhances neuronal excitability in dorsal horn neurons by activating ionotropic glutamate receptors (Piani et al. 1993) or mGluRs (Mills et al. 2000). Activated microglia express cell surface markers and receptors and release substances such as glutamate, cytokines, nitric oxide and prostaglandins (DeLeo & Yezierski, 2001; Hanisch, 2002). Pathological changes within glia, which occur duirng the development of neuropathic pain, cause dysfunction of the astroglial network, which in turn activates mGluRs, particularly mGlu5 receptors (Ciccarelli et al. 1997; Zonta et al. 2003).

Recently, the pharmacological inhibition of glial activation with minocycline, pentoxifylline or propentofylline was suggested as a novel and effective therapy for controlling neuropathic pain syndromes (Sweitzer et al. 2001; Ledeboer et al. 2005; Mika et al. 2007; Tawfik et al. 2008). The analgesic effects of glial inhibitors in neuropathic pain have been suggested to occur through the attenuation of pro-inflammatory cytokine activity (Hanisch, 2002; González et al. 2007). Interestingly, it has also been shown that inhibition of glial activation by propentofylline reverses the downregulation of glial glutamate transporters caused by L5 nerve transection (Tawfik et al. 2008). Furthermore, a combined pharmacology study has shown that repeated pentoxifylline or minocycline administration enhances the analgesic effectiveness of intraperitoneally and intrathecally injected mGluR ligands, such as MPEP, LY379268 and AMN082. Moreover, the glial inhibitors reduced the chronic constriction injury (CCI)-induced changes in microglial markers and the levels of mGluR proteins (Osikowicz et al. 2009). Based on behavioural pharmacology, Osikowicz et al. (2009) suggested that the majority of mGlu5 receptors that are upregulated after nerve injury are localized on glial cells, because pharmacological inhibition of glia eliminated most of the changes in mGlu5 receptor mRNA and protein levels. These results validate previous findings that mGluRs are present on neurons and glial cells (Petralia et al. 1996a; Biber et al. 1999) and that distinct glial mGluR subtypes localized on glial cells may be involved in the interaction between glia and neurons (Ciccarelli et al. 1997; Agrawal et al. 1998; Taylor et al. 2002).

Many investigators have studied the mechanism of action of glial inhibitors in neuropathic pain. González and colleagues showed that minocycline blocks voltage-dependent Ca2+ channels, thereby decreasing the intracellular Ca2+ concentration and causing a reduction in glutamate release (González et al. 2007). Some data indicate that stimulation of mGlu5 receptors leads to the activation of extracellular signal regulated kinases 1 and 2 (ERK1 and 2) in the spinal cord (Ferraguti et al. 1999). These kinases are activated in both neurons and glial cells as a result of nerve injury (Ji et al. 1999; Zhuang et al. 2005). Activation of the ERK pathway increases the synthesis of multiple inflammatory mediators. Such activation may also regulate the expression of glutamate transporters and the release of glutamate, thereby contributing to the maintenance of neuropathic pain. It has been suggested that following combined treatment with a glial inhibitor (minocycline) and an mGlu5 receptor antagonist (MPEP), activation of the ERK pathway is inhibited, which may cause a reduction in the changes mentioned above and possibly contribute to the analgesic effects observed after co-administration of the two drugs. The results indicate that glial activation can modulate the effectiveness of mGluR ligands and that attenuation of injury-induced glial activation reduces glutamatergic activity, which is an important factor in the pathogenesis and maintenance of neuropathic pain.

Glutamate transporters Glia contribute to synaptic homeostasis by preventing glutamate excitotoxicity by promoting Na+-dependent glutamate uptake (Liberto et al. 2004). It seems that among EAATs, EAAT1/GLAST and EAAT2/GLT-1 play important roles in the effects of glutamate. These transporters are thought to be localized primarily on astrocytes, although recently they have been demonstrated to be expressed on microglia, macrophages (Lopez-Redondo et al. 2000; Danbolt, 2001; Gras et al. 2012) and neurons (Chen et al. 2004). However, not much is known about how the glutamate transporters are regulated in microglial cells. Also, the precise significance of glutamate transporters on astrocytes vs. microglia related to pain still remains poorly understood; therefore, it is not known whether the attenuation of glial activation is sufficient to lower the glutamatergic tone and consequently attenuate neuropathic pain. Tawfik et al. (2008) demonstrated that propentofylline, which exhibits antiallodynic properties in a rodent model of neuropathic pain, inhibits astrocytic activation and modulates spinal astrocytic promoter activation for GLT-1 and GLAST in vivo. Recently, Zhang et al. (2012) have shown that minocycline prevents the downregulation of both GLAST and GLT-1 in the spinal dorsal horn.

Vesicular glutamate transporters are responsible for vesicular glutamate storage and exocytotic glutamate release in neurons and astrocytes. Astrocytes can respond to synaptic activation (Dani et al. 1992) and they can modulate synaptic neurotransmission by releasing glutamate in a Ca2+-dependent manner (Araque et al. 1998). Given that the astrocytic internal Ca2+ levels necessary for glutamate release are within the physiological range (Parpura & Haydon, 2000), this release can be used as a signalling pathway that may influence nociceptive signalling. The role of VGLUT in astrocytes during neuropathic pain remains unclear; however, it is known that VGLUT1, VGLUT2 and VGLUT3 are each expressed in approximately one-third of the astrocytic population (32, 35 and 41%, respectively). It is tempting to suggest that astrocytes could show highly complementary distributions of these three proteins (Herzog et al. 2001; Fremeau et al. 2002). An alternative extreme possibility is that only one-third of astrocytes express all isoforms of VGLUTs. Consequently, it should become possible to manipulate individual VGLUTs selectively in astrocytes to test for their roles in glutamate-mediated astrocyte–neuron signalling in neuropathic pain. Thus, the glutamate effects may depend on glial activation, which occurs when the nervous system is damaged. Glia may be an additional factor to consider when searching for new treatments of neuropathic pain.

The modulatory effect of the glutamatergic system on the effects of analgesic drugs used in neuropathic pain therapy

Glutamate receptors Neuropathic pain has often been classified as opioid resistant, because it can be only partly relieved by high doses of opioids; such a phenomenon is also observed in animal models (Ossipov et al. 1995; Mika et al. 2007). The mechanisms underlying this effect are not fully understood, but appear to be composed of two types of plasticity or counter-adaptation, at the cellular level and through neuronal circuits. The important role in the effects of opioids during neuropathic pain plays so-called anti-opioid glutamate/NMDA receptors. Furthermore, NMDA antagonists (ketamine, MK-801, dextromethorphan, memantine and MRZ 2/579) potentiate the effects of morphine. Methadone, which has an affinity for opioid receptors and has antagonist activity against NMDA receptors, is used in the treatment of neuropathic pain. This characteristics determines the suitability of methadone in the treatment of patients with pain refractory to the action of morphine or intolerance (Christoph et al. 2006; Nichols et al. 1997). There is evidence in the literature indicating that the reduction in the antinociceptive efficacy of morphine in nerve-injured animals in addition to the reported reduction in the number of presynaptic opioid receptors (Porreca et al. 1998) may be due to persistent activation of NMDA receptors, which colocalize with mGluRs on neurons (Benquet et al. 2002). Recent data have shown that mGlu5 receptor antagonists, such as MPEP, attenuate the development of opioid tolerance if co-administered with morphine (Kozela et al. 2003; Narita et al. 2005; Osikowicz et al. 2008). Mayer et al. (1995) suggested similar cellular mechanisms (which may be mediated by protein kinase C) for the development of neuropathic pain and the development of tolerance to the analgesic effects of morphine. The mechanisms through which inhibition of group I mGluRs might influence neuropathic pain and morphine effectiveness are still unclear; however, it has been suggested that group I mGlu, opioid and NMDA receptors might colocalize in the same cells (Fundytus et al. 2001). Furthermore, stimulation of group I mGluRs and, as result of interactions between NMDA receptors, the activation of protein kinase C, leads to an increase in the concentration of intracellular Ca2+. Activated protein kinase C has been shown to phosphorylate opioid receptors and induce desensitization (Narita et al. 1996), reducing the analgesic efficacy of morphine. As a consequence of mGluR group I antagonists injections, a decrease in protein kinase C activation and attenuation of the phosphorylation and desensitization of opioid receptors occurs, thus possibly increasing the efficacy of opioid analgesics. However, intracellular Ca2+ initiates numerous second messenger-mediated intracellular signal transduction cascades.

The effects of group II and III mGluR agonists (LY379268 and AMN082, respectively) on the morphine response in neuropathic pain were demonstrated in a pharmacological study, although until now, the mechanism of such interactions has been unclear. Given that mGlu2, mGlu3, mGluR7 and opioid receptors belong to the G-protein-coupled receptor family, it is possible that both drugs affect the same points in the intracellular messenger pathway (Berenbaum, 1989). Thus, LY379268, AMN082 and morphine may independently alter intracellular second messenger systems coupled with G-protein activation, mediating synergistic interactions. Other hypotheses that may explain the enhanced effect of morphine by mGluR group II and III agonists include presynaptic mGluR agonists, which are thought to inhibit NMDA receptor function (Glaum & Miller, 1994; Pin & Duvoisin, 1995) via decreasing protein kinase C activation and consequently attenuating the phosphorylation and desensitization of opioid receptors, which results in the increased efficacy of opioid analgesics, as described above.

Glutamate transporters Not much is known about the role of EAATs and VGLUTs in the effects of opioids during neuropathic pain; therefore, this requires investigation, because their role in nociceptive transmission and the effects of opioids (Chen et al. 2012) is very crucial. It has been shown that chronic intrathecal morphine administration downregulates GLAST in a dose-dependent manner in the superficial spinal cord dorsal horn of the rat (Mao et al. 2002). Recently, Yang et al. (2011) have shown that ultra-low dose naloxone enhanced the antinociceptive effect of morphine in neuropathic pain, possibly by restoration of GLAST and GLT-1 expression in astrocytes. Also, Shen et al. (2011) demonstrated that the tumour necrosis factor-α antagonist etanercept attenuated the downregulation of membrane glutamate transporters GLT-1 and GLAST and partly restored the antinociceptive effect of morphine.

Conclusion

In order to develop novel therapeutic strategies for neuropathic pain, a better understanding of the molecular and cellular mechanisms underlying the pathogenesis of neuropathic pain is required in order to inhibit its induction and maintenance. Recent results have provided new information on the involvement of mGluRs and also glutamate transporters in the development and maintenance of neuropathic pain. These receptors undergo changes in the lumbar spinal cord after sciatic nerve injury and, interestingly, these changes are reversed after the administration of inhibitors of glial activation, which indicates the important role of glial mGluRs in these effects. Furthermore, some mGluR ligands improve the efficiency of morphine and other drugs that are used in therapy and prevent the development of morphine tolerance in neuropathic pain. Improving the efficiency of morphine and other drugs (such as antidepressants) is extremely important, because the prolonged use of high doses of traditional opioid analgesics in patients with neuropathic pain may cause the development of side-effects. Thus, the search for new treatment strategies is an important problem in both the clinical and the scientific fields.

Call for comments

Readers are invited to give their opinion on this article. To submit a comment, go to: http://ep.physoc.org/letters/submit/expphysiol;98/2/372

Appendix

Acknowledgements

This study was supported by grant of the Ministry of Science and Higher Education NN405375937 and by the grant from Polish Ministry of Science and Higher Education Scientific Network Fund POIG.01.01.02.-12-004/09 Demeter 1.1.2.5.4.