Endogenous Antinociceptive Ligands 435 nociception acting on GlyRs (Pellicer et al., 2007). Both IT and intraperitoneal (IP) administration of taurine relieves nociceptive stimulation effects, and pain stimulus releases taurine in the spinal cord (Hornfeldt et al., 1992;Ishikawa et al., 2000; Legendre, 2001; Serrano et al., 1998; Skilling et al., 1990; Smullin et al., 1990). 2.3.6 Kynurenic Acid (KYNA) Degradation of the essential amino acid tryptophan along the kynurenine pathway yields several neuroactive intermediates, including kynurenic acid (4-oxo-1H- quinoline-2-carboxylic acid) (Moroni et al., 1988; Schwarcz and Pellicciari, 2002; Vecsei and Beal, 1991). This is found both centrally and peripherally in low con- centrations (10–150 nM) and is synthesized in the CNS, predominantly by glial cells (Moroni et al., 1988; Pawlak et al., 2000; Schwarcz and Pellicciari, 2002; Turski and Schwarcz, 1988; Urbanska et al., 2000). KYNA at high, nonphysiolog- ical concentrations is a broad-spectrum antagonist of ionotropic excitatory amino acid receptors, acting at the Gly (half-maximal inhibitory concentration: IC 50 ∼ 20 μM) and the NMDA recognition sites (IC 50 ∼ 200 μM) of the NMDA recep- tor complex (Carpenedo et al., 2001; Ganong et al., 1983; Stone, 1993). In higher concentrations (0.1–1 mM), it also antagonizes the AMPA and kainate receptors, and KYNA is a potent noncompetitive antagonist of α7 nAchRs (IC 50 ∼7 μM) too (Hilmas et al., 2001; Stone, 2000). Thus, direct support for its physiological role in glutamatergic and cholinergic neurotransmission has been reported (Carpenedo et al., 2001; Nemeth et al., 2005; Schwarcz and Pellicciari, 2002). A recent study has shown that GPR35, a previously orphan GPCR, functions as a receptor for kynurenic acid (Wang et al., 2006a). KYNA elicits calcium mobi- lization and IP3 production in a GPR35-dependent manner, and it also induces the internalization of this receptor. GPR35 is predominantly detected in immune cells and the gastrointestinal tract, but it has also been found in the DRG on small- to medium-diameter neurons (Ohshiro et al., 2008). The results suggest that GPR35 may modulate nociception and a continued study of this receptor will pro- vide additional insight into the role of KYNA in pain perception, inasmuch as no in vivo data are available regarding the role of GPR35 in the effects of KYNA. Systemic administration of KYNA produced antinociception in acute heat pain tests and attenuated the development of tolerance (Heinricher and McGaraughty, 1998; Marek et al., 1991). Intracisternally administered KYNA effectively inhibited the capsaicin-induced pain behavior (Hajos and Engberg, 1990). Intra-RVM infusion of KYNA inhibited the opioid-induced antinociception, although the baseline pain threshold was unaffected (Heinricher and McGaraughty, 1998; Heinricher et al., 1999). IT administration of KYNA produces antinociception in different models (Raigorodsky and Urca, 1990; Yaksh, 1989; Yamamoto and Yaksh, 1992; Zhang et al., 2003b), and enhanced the effects of EM-1 and AGM (Horvath et al., 2007; Horvath and Kekesi, 2006; Kekesi et al., 2002). Its peripheral administration also 436 G. Horvath produced antinociception in a joint inflammatory model with low potency (Mecs et al., 2009). 3 Purines 3.1 Adenosine The endogenous purine mediator adenosine, originating from adenosine 5- triphosphate (ATP), is a widely distributed neuromodulator with complex effects (Sawynok, 1998; Sawynok and Liu, 2003). Four adenosine receptors have been identified and are termed A1, A2A, A2B, and A3 (Fredholm et al., 2001). They are all GPCRs and couple to classical second messenger pathways; A1 and A3 recep- tor activation decreases the level of cAMP, A2 increases it, whereas A2B receptor stimulates PLC (Sawynok, 1998; Sawynok and Liu, 2003). It has a complex influence on nociceptive pathways because its effects depend on the receptor subtype activated (Sawynok, 1998; Sawynok and Liu, 2003). The acti- vation of the A1 and A3 receptors produces analgesia by both peripheral and central mechanisms, and a variety of molecules is being developed to provide analgesia through this nonopioid mechanism (Poon and Sawynok, 1998; Sawynok, 1998; Sawynok and Liu, 2003). It is very important that adenosine has a significant role in opioid-induced antinociception (Sawynok, 1998). Only a f ew data support its antinociceptive potency at the supraspinal level. Thus caffeine (adenosine antago- nist) decreased the analgesic effects of ICV-administered opioids, and MOR and DOR (but not κ-opioid receptor [KOR]) agonists (Pham et al., 2003). Furthermore, both A1 and A2A agonists produced antinociception in acute pain models (Pham et al., 2003; Regaya et al., 2004). Several reports suggest the antinociceptive effect of synthetic adenosine deriva- tives or adenosine kinase inhibitors in different pain tests at spinal level (Poon and Sawynok, 1998; Sawynok, 1998), however, only few laboratories (those of Sollevi and Eisenach) have investigated adenosine in this regard (Belfrage et al., 1999; Chiari and Eisenach, 1999; Eisenach et al., 2002; Lavand’homme and Eisenach, 1999; Rane et al., 2000; Von Heijne et al., 1999). Most studies have observed effec- tive antinociception in neuropathic pain states, but slight or no effects on normal or inflammatory pain sensitivity have been found (Kekesi et al., 2004; Lavand’homme and Eisenach, 1999; Rane et al., 2000; Sawynok and Liu, 2003). This ineffectivity might be due to uptake and metabolic degradation of adenosine (Sawynok and Liu, 2003). It seems that antinociceptive effects of adenosine are particularly related to the activation of A1 receptors in the spinal cord where inhibition of intrinsic and pri- mary sensory neurons may contribute to this actions (Sawynok and Liu, 2003; Schulte et al., 2003; Sollevi et al., 1995). Moreover, adenosine agonists produce analgesia largely by interacting with the descending inhibitory noradrenergic s ys- tem, and the effect of adenosine is blocked by α 2 -adrenergic antagonists (Gomes et al., 1999; Sweeney et al., 1987). A component of the antinociceptive action Endogenous Antinociceptive Ligands 437 of morphine is also due to the local release of adenosine within the spinal cord (Sawynok et al., 1989; Sweeney et al., 1987). Adenosine acting at its A2A receptor is thought to be pronociceptive, and this effect has been proposed to result from the increase in cAMP levels and activation of NMDA receptors (Hussey et al., 2007; Khasar et al., 1995). As regards the effects of adenosine at peripheral level, the A1 receptor is the predominant receptor subtype mediating antinociception at peripheral level too, whereas the A2A, A2B, and A3 receptors mediate nociception peripher- ally (Sawynok and Liu, 2003). Thus it has been shown that the activation of A1 receptors produced a significant antihyperalgesic effect in the inflammatory pain model (Vuckovic et al., 2006). In contrast, A2A receptor knock-out mice have a higher nociceptive threshold and this has been suggested to be attributable to the lack of peripheral adenosine A2A receptors (Ledent et al., 1997). 3.2 Nucleotides Nucleotides are molecules which comprise the structural units of RNA and DNA. Additionally, nucleotides play central roles in the metabolism. These nucleotides might be adenine-(adenosine di- and triphosphate: ADP and ATP), guanine- (guanosine di- and triphosphate: GDP, GTP), or pyrimidine-nucleotides (uridine di- and triphosphate: UDP, UTP). It has been established that these molecules medi- ate diverse biological effects via P2 purinoceptors (P2Xn: ionotropic and P2Yn: metabotropic receptors) in both the PNS and CNS (Tsuda et al., 2005; Wirkner et al., 2007). The P2X receptors expressed in the brain are primarily distributed throughout the rat hindbrain, including the RVM and LC (Kanjhan et al., 1999; Wirkner et al., 2007). ATP acting on P2X receptors at the supraspinal level pro- duces mechanical and thermal antinociception in rats through the activation of P2X 3 receptors (Fukui et al., 2004; Fukui et al., 2006; Wirkner et al., 2007). It is conceiv- able that the ascending noradrenergic neurons arising from the LC are i nvolved in the supraspinal antinociception by a P2X receptor agonist (Fukui et al., 2004). At the spinal level the expression of the P2X 3 receptors appears selective for a sub- population of small diameter DRG neurons, which are probably associated with nociception (Inoue et al., 2007; Wirkner et al., 2007). As regards the role of P2Y receptors, the activation of some of these receptors by UTP (acting on P2Y 2,4,6,14 ) in the brain had no effect on the mechanical nociceptive threshold (Fukui et al., 2001). Only a few data suggest that the expression of these receptors increased in the microglia at the spinal level during neuropathy, and blocking or lack of this receptor produced antinociception (Gerevich and Illes, 2004; Inoue et al., 2007; Tozaki-Saitoh et al., 2008). Thus, the activation of these receptors contributes to an acute nociceptive behavior, hyperalgesia, and allodynia. In contrast, some data suggest that P2Y receptor agonists can inhibit cytokine release from activated spinal cord microglia (Gerevich and Illes, 2004). This process could interrupt chronic pain development and continuation. Thus, UTP and UDP were shown to be analgesic in the neuropathic pain model (Okada et al., 2002). IT pyrimidine nucleotides elevated the nociceptive threshold in the paw pressure and 438 G. Horvath TF tests, whereas adenine nucleotides (activate: P2Y 1,2,11,12,13 ) lowered it and pro- duced allodynia (Gerevich and Illes, 2004). The ADP analogue ADP-β-S (acting on P2Y 1,12,13 ) has also been found to cause analgesia in TF test. As regards the action mechanism, it seems that both pyrimidine nucleotides and ADP-β-S produce antinociception by the activation of P2Y receptor causing the inhibition of the volt- age gated N-type Ca 2+ channels, and decrease the transmitter release in the spinal cord (Gerevich et al., 2004). However, the specific antagonist of P2Y 12 receptor decreased neuropathic pain, suggesting that this type of purinergic receptors may be critical in the pathogenesis of neuropathic pain (Tozaki-Saitoh et al., 2008). These ligands, mainly the ATP play a facilitatory role in pain t ransmission at peripheral level (Wirkner et al., 2007). Extracellular guanine-based purines (GBP), namely the nucleotide guanosine monophosphate (GMP) and the nucleoside guanosine also exert biological effects. Such actions are unrelated to direct G-protein modulation, but GBPs induce mod- ulation of the glutamatergic system by the inhibition of the binding of glutamate and analogues, and prevent cell responses to excitatory amino acids (Burgos et al., 1998; Morciano et al., 2004). Both neurons and astrocytes release guanosine under basal and toxic conditions (Ciccarelli et al., 2001). ICV administration of guano- sine or its prosubstance, GTP, produced dose-dependent antinociceptive effects in several different acute nociceptive tests (HP, TF, intraplantar: IPL capsaicin or glu- tamate) (Schmidt et al., 2008). The action mechanism of this effect is unknown, but it was not inhibited by the adenosine- or opioid receptor antagonists, and guanosine did not increase the adenosine level in the brain. (Traversa et al., 2003). However, guanosine significantly stimulates glutamate uptake, thereby preventing glutamate toxicity, therefore, it is tempting to suppose that the in vivo antinociceptive effect of guanosine can result from its effect on glutamate removal from the synaptic cleft, leading to less activation of glutamate receptors (Frizzo et al., 2003). 4 Other Nonpeptide Molecules 4.1 Ouabain Endogenous cardiac glycoside inhibitors of Na + /K + -ATPase with structures simi- lar to that of plant-derived ouabain (1β,3β,5β,11α,14,19-hexahydroxycard-20(22)- enolide 3-(6-deoxy-α-L-mannopyranoside)) have been isolated from several tissues, including the adrenal cortex and the brain (Van Huysse and Leenen, 1998). Ouabain, through the inhibition of Na + ,K + -ATPase, may produce several effects including modulation of neural activity and neurotransmitter release. These effects might be related to the pain mechanism; but only few studies investigated its role in this context. ICV-injected ouabain in relatively high doses (μg) exerts an antinocicep- tive effect and potentiates the analgesic activity of morphine (Calcutt et al., 1971). However, lower doses (ng) of ouabain were able to antagonize the antinociception induced by morphine (Masocha et al., 2003). IT administration of ouabain in high doses also produced analgesia and enhanced the potency of morphine (Zeng et al., Endogenous Antinociceptive Ligands 439 1999). Another study found that low doses of ouabain did not modify the acute heat pain latency, but higher doses caused excitation and motor impairment, suggesting that ouabain does not produce a pronounced effect on the pain threshold (Horvath et al., 2003). It is supposed that the controversial results might be due to the fact that ouabain produces its effect on all of the cells, inasmuch as all have Na + ,K + -ATPase, and its net effects on transmitter releases might be dose-dependent. 5Peptides A growing number of peptides have been identified in the CNS and the periphery that relate to pain modulation (Palkovits, 1984). They can originate from neurons, endocrine cells, immunocytes, fat and muscle cells, and so on. There is no clear classification for peptides, and there is some overlap between the different groups. The first group of them contains the hormones, however, now it is well established that most of the hormones can originate from neurons, and they can modify the neuronal functions as well. The second group comprises neuropeptides consisting of short chains of amino acids, with some functioning as neurotransmitters others as hormones; they are often localized in axon terminals at synapses and are classified as putative neurotransmitters. They include endorphins, ENKs, and others. Cytokines (the third group) are a special category of s ignaling molecules that, like hormones and neurotransmitters, are used extensively in cellular communication. Anatomical and structural distinctions between cytokines and classic hormones are fading as we learn more about each. Classic protein hormones circulate in nanomolar (10 -9 ) concentrations that usually vary by less than one order of magnitude. In contrast, some cytokines (such as IL-6) circulate in picomolar (10 -12 ) concentrations that can increase up to 1000-fold during a trauma or infection. 5.1 Peptide Hormones 5.1.1 Oxytocin (OT) Oxytocin (nonapeptide) is mainly synthesized together with arginin-vasopressin (AVP) in magnocellular neurons of the paraventricular (PVN) and supraoptic (SO) nuclei of the hypothalamus, and acts as a neurohormone during parturition and the milk ejection reflex (Gimpl and Fahrenholz, 2001). However, it can subserve a neurotransmitter/neuromodulator function as well. OT exerts its actions via the OT receptor, a GPCR receptor (G q/11 class, stimulates PLC activity), which is local- ized in many parts of the CNS and PNS (Tribollet et al., 1992). Descending OTergic pathways extend from the hypothalamus to the thalamus and brainstem (Sawchenko and Swanson, 1982), thus OTergic terminals and high-affinity binding sites for OT are present in regions involved in pain perception (Tribollet et al., 1992). High con- centrations of OT are found in raphe nuclei, and OT modulates serotonin turnover in the brain (Kovacs, 1986). It has been suggested that a loop exists between the LC and the hypothalamus, a pathway that may be involved in the regulation of the 440 G. Horvath release of OT (and AVP) following painful stimulation (Swanson and McKellar, 1979). At least 25% of OTergic neurons in the hypothalamus project to the SDH, and these projection sites match well OT binding sites in the superficial layers of the SDH and in the autonomic regions (Rousselot et al., 1990). Although some conflicting results have been reported in the literature, analgesic effects of OT have been proved in most studies after systemic and/or central admin- istration. The ICV injection of OT produced a significant antinociceptive effect in different acute pain models (Arletti et al., 1993; Gao and Yu, 2004; Zubrzycka and Janecka, 2008). An OT-sensitive antinociception can be induced by massage-like stimulation, swim stress, and electrical stimulation the PVN in both naive and neu- ropathic rats, suggesting the involvement of an endogenous OT receptor-dependent analgesic system. The OT-induced antinociception might be mediated by MOR and KOR activation, which suggests the release of endogenous opioids after OT recep- tor activation (Zubrzycka et al., 2005). The stimulation of PVN causes OT release in the spinal cord and can influence spinal nociceptive processing (Condes-Lara et al., 2005; Yang, 1994). This OT-specific stimulation of neurons allows the recruitment of GABA-ergic interneurons in lamina II, which produces a generalized elevation of local inhibition. 5.1.2 Vasopressin (Arginine Vasopressin: AVP) AVP (nonapeptide), the other posterior pituitary hormone, is mainly synthesized in the PVN of hypothalamus. Similarly to OT, descending AVPergic pathways extend from the hypothalamus to the thalamus, medulla oblongata, and the sub- stantia gelatinosa of the SDH (Sawchenko and Swanson, 1982). Especially PAG contains many AVP-containing fibers (Pittman et al., 1981). Three s ubtypes of AVP receptors (GPCRs), V1, V2, and V3, have been identified, mediating vasoconstric- tion, water reabsorption, and central nervous system effects, respectively (Holmes et al., 2003). Functionally, V1R activates G-proteins of the G q/11 , whereas V2R stimulates the Gs proteins. A variety of signaling pathways is associated with V1R including the activation of calcium influx, PLA2, PLC, and PLD; in contrast, V2R activates cAMP. More than one G-protein appears to participate in signal trans- duction pathways linked to V3Rs, depending on the level of receptor expression and the concentration of AVP. Many experiments discovered that AVP is related to pain modulation, and pain stimuli elevate AVP concentration in different brain areas (NRM, caudate nucleus, and PVN), furthermore microinjection of AVP into these centers raised pain thresholds (Yang et al., 2007a; Zubrzycka and Janecka, 2007). Stimulation of PVN caused antinociception, which was antagonized by anti- AVP (Yang et al., 2007a). Central injection of AVP (ICV, intra-PAG) increased pain threshold and the level of endogenous opioids, thus its effects were reversed by naloxone suggesting that the release of endogenous opioids plays a significant role in its antinociceptive effect (Yang et al., 2007a, b; Zubrzycka et al., 2005; Zubrzycka and Janecka, 2008). The analgesic effect of AVP in PAG can be reversed by a V2 antagonist and V2 antagonist also reduced the basal pain threshold, suggesting an inhibitory tone of these fibres (Yang et al., 2006). It seems that the antinociceptive Endogenous Antinociceptive Ligands 441 effect of AVP is limited to the brain nuclei, not to the spinal cord and peripheral organs, because IT or intravenous (IV) injection of AVP or anti-AVP serum did not change the pain threshold (Yang et al., 2007a). 5.1.3 Calcitonin (CT)/Parathyroid Hormone Fragments (PTH)/Tuberoinfundibular Peptide of 39 Residues (TIP39) Parathormone (PTH) (84 amino acids) acts in concert with calcitonin (CT; 32-amino acids) to maintain the serum calcium level acting on GPCR receptors. TIP39 was purified based on parathyroid hormone-2 receptor (PTHR2) activation, and it is an endogenous ligand for the PTHR2, a GPCR receptor increasing cAMP level (Usdin et al., 2003). Both ligands and their binding sites can be found in regions that are involved in processing pain-related information (Harvey and Hayer, 1993; Olgiati et al., 1983; Usdin et al., 2003). The ability of CT to produce antinociception has led to the suggestion that CT may serve as a neuromodulator in the CNS. Systemic injection of CT produced antinociception in inflammatory pain models, and this was inhibited by ICV but not by IT application of serotonin-antagonists, and neither IT nor ICV administration of α-adrenoceptor antagonist and systemic injection of opioid antagonist influenced the antinociceptive effects (Yamazaki et al., 1999). ICV administration also was effective in acute heat- and visceral pain models (Welch et al., 1986). One study has investigated the antinociceptive interaction of fragments of the PTH with CT after their ICV administration (Welch and Dewey, 1990). It has been suggested that CT and some PTH fragments interact in the modu- lation of nonopiate antinociception, possibly via actions on the calcium level in the brain. CT is not an effective antinociceptive ligand at the spinal level (Wiesenfeld- Hallin and Persson, 1984). Only few data support the role of TIP39 in the pain processes. ICV administration of TIP39 partially reversed tactile withdrawal hyper- sensitivity following carrageenan administration, but did not change the HP latency or the formalin-induced behavioral responses (LaBuda and Usdin, 2004). TIP39 also decreased the aversiveness of paw stimulation, suggesting that it may modulate an effective component of nociception within the brain, and its IT administration produced pronociceptive action (Dobolyi et al., 2002; LaBuda and Usdin, 2004). 5.1.4 Insulin Insulin (51 amino acids) is produced in the islets of Langerhans in the pancreas. Insulin regulates not only the blood sugar level, but also various CNS functions acting on its tyrosine-kinase receptor (TrK). A few early data suggest that insulin can induce antinociception, and diabetic rats are less sensitive to the antinociceptive effect of morphine (Bodnar et al., 1979; Simon and Dewey, 1981). Both systemic and ICV insulin decreased the formalin-evoked behavior, and the insulin-induced antinociception was independent of hypoglycemic effects, but it could be due to the activation of endogenous dopaminergic, serotoninergic, and opioidergic systems (Anuradha et al., 2004; Takeshita and Yamaguchi, 1997). Insulin inhibits neuronal firing in the hippocampus and hypothalamus, and insulin-induced antinociception 442 G. Horvath could possibly involve all these centers, whereas the IT administration of insulin did not influence the pain threshold (Bitar et al., 1996). 5.1.5 Renin-Angiotensin System (RAS) Angiotensin II (Ang II; octapeptide) is a key regulator of the cardiovascular system, and it is the main effector of RAS. The final steps of its biosynthesis involve con- secutive proteolytic cleavages of its inactive precursors, angiotensinogen (AO) and angiotensin I (Ang I), by renin and angiotensin converting enzyme (ACE), respec- tively (Suarez et al., 2002). Ang II-containing neurons and fibers were identified in the brain, especially in the hypothalamic regions, the nucleus of the solitary tract, and in the PAG (Sofroniew et al., 1981). Two receptor subtypes Ang II type 1 (AT1) and Ang II type 2 (AT2) (GPCRs: increase PLC and/or cAMP formation) have been identified, and the main biological functions exerted by Ang II are medi- ated by the AT1 receptor subtype. Ang III (metabolite of Ang II) binds primarily to AT2 subtypes, and the biological function of AT2 receptors is still controversial (McKinley et al., 2003; Suarez et al., 2002). The localization of Ang II and its recep- tors in the PAG reinforces the suggestion that endogenous Ang II participates in an AT-receptor-mediated modulation of nociception (Pelegrini-da-Silva et al., 2005). AT2 receptor-deficient mice have increased sensitivity to pain and decreased the levels of brain β-endorphin (Sakagawa et al., 2000). Antinociception following ICV administration of Ang II and Ang III has been demonstrated in several rodent pain models (Pelegrini-da-Silva et al., 2005; Raghavendra et al., 1999). Administration of Ang III into the rat nucleus reticularis gigantocellularis also evokes antinociception in TF test (Yien et al., 1993). Microinjection of angiotensinogen, Ang I, Ang II, or Ang III into the PAG produces a dose-dependent antinociceptive effect in the rat TF-test, which was inhibited by Ang- and opioid receptor antagonists (Pelegrini- da-Silva et al., 2005; Prado et al., 2003; Adams et al., 1986). Additional evidence of the involvement of RAS peptides in nociception includes the reduction of the SIA by Ang receptor antagonists (Haulica et al., 1986; Raghavendra et al., 1999). Nerve terminals containing Ang II-like immunoreactivity (LI) have been identified in the primary sensory neurons and in the SDH as well (Buck et al., 1982), and IT administration of Ang II increased the TF latency (Thurston et al., 1992), whereas others did not find this effect (Cridland and Henry, 1988a). 5.1.6 Melanocortin System (MC) Amongst the wide range of modulators, the melanocortin system represents a rela- tively new, intriguing, potential target for pain control (Bertorelli et al., 2005). MCs are a family of endogenous peptides generated by enzymatic cleavage of a com- mon precursor molecule, proopiomelanocortin (POMC). Main members of the MC family are α-, β-, γ-melanocyte-stimulating hormones (MSH: containing 16, 22, 12 amino acids, respectively) and the adrenocorticotropic hormone (ACTH; 39 amino acids). MCs exert their actions through activation of at least five subtypes of recep- tors (MC1–MC5), which are GPCRs, and each of them is positively coupled to AC. A further peculiarity of the system is that, in addition to endogenous agonists, there Endogenous Antinociceptive Ligands 443 are also endogenous antagonists such as agouti protein, which binds preferentially to MC1 receptors and is expressed mainly in the skin, and the agouti-related protein (AgRP), which is an inverse agonist of both MC3 and MC4 receptors and is mainly expressed in the brain (Dinulescu and Cone, 2000; Nijenhuis et al., 2001). Thus, both endogenous agonists and antagonists can be detected in the CNS including the spinal cord (Bertorelli et al., 2005), and the MC system could be under the control not only of an excitatory but also an inhibitory system. MCs and their receptors are mainly present in the periphery where they can be found primarily on melanoma cells and melanocytes (Wikberg and Mutulis, 2008). In the CNS, the MC1 receptor is present on neurons in the PAG of the midbrain, where it is thought to have a role in pain control (Mogil et al., 2003; Palkovits et al., 1987). Furthermore, mainly MC3 and MC4 receptors are found in the spinal cord. A possible link between MCs and nociception was first postulated by pioneering studies in late the 1970s, early 1980s, showing that ICV administration of α-MSH and ACTH causes hyperalgesia, and reverses the analgesic effects of morphine and β-endorphin (Sandman and Kastin, 1981; Williams et al., 1986). Thus, these ligands produce mainly hyperalgesia, whereas the antagonists produce antinoci- ception in inflammatory and nerve injury models (Bellasio et al., 2003; Bertorelli et al., 2005; Mogil et al., 2003; Sandman and Kastin, 1981; Starowicz et al., 2002; Vrinten et al., 2000; Vrinten et al., 2001). However, MSH has an anti-inflammatory potency as well, and it antagonizes the interleukin-1β-induced hyperalgesia in the PVN (Ceriani et al., 1994; Macaluso et al., 1994). Furthermore, MC1 mediates KOR-mediated analgesia in female mice (Mogil et al., 2003). Inasmuch as agouti protein and AgRP are endogenous antagonists of MC receptors, the inhibition of these receptors by these endogenous ligands can produce effective antinociception, as was shown after their IT administration (Bellasio et al., 2003; Bertorelli et al., 2005). 5.1.7 Corticotropin-Releasing Factor (CRF) and Related Peptides CRF (41 amino acids) is best known as the major physiological regulator of pitu- itary ACTH secretion. CRF not only mediates stress responses but also acts as a neuromodulator of synaptic transmission outside of the hypothalamic–pituitary– adrenocortical axis (Ji and Neugebauer, 2008). In addition to CRF, the CRF family encompasses three novel CRF-related mammalian ligands, urocortin 1 (Ucn1), Ucn2, and Ucn3 (Martinez et al., 2004; Perrin and Vale, 1999). Ucn1 contains 40 amino acids, whereas Ucn2 and Ucn3 are composed of 38 amino acids. Each type of Ucns is found in different brain areas including the PAG, but they can be identi- fied peripherally as well (Martinez et al., 2004). These ligands mediate their actions through interaction with two distinct receptor subtypes, CRF1 and CRF2, and both receptors can couple to similar s ignal transduction pathways (AC and PKA). CRF has preferential affinity for CRF1, Ucn1 binds with equal high affinity to both CRF receptors, and Ucn2 and Ucn3 exhibit high selectivity towards CRF2 receptors (Dautzenberg and Hauger, 2002; De Souza et al., 1985; Korosi et al., 2007). The presence of Ucn1 and Ucn3 immunoreactive nerve terminals in association with 444 G. Horvath CRF2 receptors in the PAG and spinal cord suggests a modulatory influence of these receptors in pain (Korosi et al., 2007). Accumulating evidence suggests that peripheral and central CRF are important pain modulators, but the literature on pain-related CRF functions in the CNS is very controversial (Million et al., 2006). CRF, injected ICV, mimicked the effects of stress-induced visceral hyperalgesia, and CRF1/CRF2 antagonists blocked this effect (Gue et al., 1997). However, other data suggest that both CRF1 and CRF2 are involved in SIA (Ji and Neugebauer, 2008; Korosi et al., 2007). A recent study has shown antinociceptive effects of CRF (ICV) in acute and inflammatory pain mod- els, but CRF has also increased pain-related vocalizations and the number of Fos immunopositive spinal neurons (Vit et al., 2006). It seems that the amygdala might be an important site in this respect, because CRF2 receptors could activate inhibitory circuits in the amygdala, whereas CRF1 receptors regulate excitatory processes (Ji and Neugebauer, 2008; Neugebauer et al., 2004). The opposing effects of CRF on nociceptive processing may be mediated through different receptors. Low concen- trations of CRF facilitate nociception through CRF1, whereas higher concentrations have inhibitory effects through CRF2 receptors, and this would be consistent with the higher affinity of CRF for CRF1 than CRF2 receptors (Dautzenberg and Hauger, 2002; Ji and Neugebauer, 2008). CRF, Ucn1, and CRF receptors occur in the spinal cord as well, mainly in laminae VII and X, and occasionally in lamina IX, whereas the receptors have not been identified in the superficial laminae of the dorsal horn (Korosi et al., 2007). It has been suggested that these receptors are also involved in stress adaptation processes, such as modulation of SIA and the mediation of visceral nociceptive information at spinal level (Korosi et al., 2007; Million et al., 2006; Robbins and Ness, 2008). Thus, IT-administered CRF produces antinociception, which is reversible by the CRF2 receptor antagonist (Nijsen et al., 2005). CRF and CRF2 receptor expressions were detected in the periphery as well, and the periph- eral CRF2 may also be involved in visceral sensitivity (Ayesta and Nikolarakis, 1989; Kiang and Wei, 1985; Million et al., 2006; Schafer et al., 1996). Intra-arterial injection of Uc2 reduced visceral hyperalgesia in vitro, and this effect was inhibited by a selective CRF2 antagonist, suggesting that the peripheral activation of CRF2 receptors has a significant role in visceral antinociception (Million et al., 2006). 5.1.8 Thyrotropin-Releasing Hormone (TRH) TRH (tripeptide: Glu-His-Pro-NH 2 ), discovered originally as a hypothalamic hor- mone, is widely distributed in the CNS, and it also coexists with substance P (SP) and 5-HT in the neurons of the medulla oblongata which projects into the spinal cord (Johansson et al., 1981). TRH exerts a variety of CNS effects, through stim- ulation of TRHR1 and TRHR2 receptors belonging to GPCR. TRHR1 receptors are expressed in the pituitary gland to release thyrod stimulating hormone, whereas TRHR2 has been shown in the brainstem nuclei, which are involved in descend- ing pain modulation (Cao et al., 1998). Both systemic and supraspinal activations of TRH receptors produce antinociceptive effects in acute heat, mechanical, and visceral pain tests, whereas IT administration has been ineffective (Tanabe et al., 2007; Webster et al., 1983). Thus, injection of TRH into the lateral ventricle, PAG, . 1997). Insulin inhibits neuronal firing in the hippocampus and hypothalamus, and insulin-induced antinociception 442 G. Horvath could possibly involve all these centers, whereas the IT administration. PTHR2, a GPCR receptor increasing cAMP level (Usdin et al., 2003). Both ligands and their binding sites can be found in regions that are involved in processing pain-related information (Harvey and. the capsaicin-induced pain behavior (Hajos and Engberg, 1990). Intra-RVM infusion of KYNA inhibited the opioid-induced antinociception, although the baseline pain threshold was unaffected (Heinricher