Chronic pain affects millions of individuals around the world and is often accompanied by dramatic changes in quality of life. The needs of patients with chronic pain are largely unmet due to the poor efficacy and side effects of current treatments, and therefore, the development of new treatments is necessary. This study indicated that the hypothalamic neuropeptide MCH mediates the development of pain behaviors in pMCH^−/− mice. Although the i.n. administration of MCH inhibited nociceptive behaviors associated with CFA-induced inflammatory pain, these MCH-induced effects were blocked by a MCH1R antagonist. The effects of MCH were also evident in the PSNL-induced neuropathic pain model. Furthermore, the present study demonstrated the possible roles of the endogenous opioid and endocannabinoid analgesic systems in the antinociceptive effects of MCH. I.n. treatment with MCH increased the number of CB1Rs as well as the activation of this system, and these changes were negatively correlated with the pain behaviors in the mPFC, vlPAG, and dlPAG. To our knowledge, this is the first report suggesting an analgesic effect of MCH in pain models.

MCH is synthesized exclusively in the lateral hypothalamus and zona incerta but MCH neurons project widely throughout the CNS, which suggests that MCH may function as a neurotransmitter and/or neuromodulator to regulate many physiological functions, including sleep, feeding, and reward^12,29–31. While the role of MCH in pain was not revealed until recently, microdialysis studies showed that MCH concentrations in the amygdala are relatively lower in epileptic patients upon feeling pain and that these levels are restored by morphine treatment³². The present study aimed to determine the involvement of MCH in pain modulation by comparing nociceptive behaviors between pMCH^−/− and pMCH^+/+ mice. The results indicated that the baseline pain sensitivity was greater in pMCH^−/− than pMCH^+/+ mice. The time to reach the maximum level of mechanical allodynia was significantly shorter in pMCH^−/− mice than pMCH^+/+ controls in both the CFA-induced inflammatory and PSNL-induced neuropathic pain models. In the later stages of pain (after Days 4 and 7 of pain induction in each model), the levels of pain remained higher in pMCH^−/− mice than in pMCH^+/+ mice, but these differences were not significant. It is possible that the pain response could not increase further after reaching a peak level. Taken together, the present results indicate that MCH plays important roles in the modulation of pain. However, pMCH encodes MCH as well as two neuropeptides, neuropeptide EI (NEI) and neuropeptide GE (NGE)³³, and may cause an alternative splice variant, MCH-gene-overprinted-polypeptide (MGOP), and an encode antisense-RNA-overlapping-MCH (AROM)^34,35. Therefore, MCH as well as other proteins encoded by pMCH, including NEI, NGE, MGOP, and AROM, may contribute to the phenotype of pMCH^−/− mice.

Next, to investigate the role of MCH in the modulation of pain, we tested whether the i.n. administration of MCH could modulate pain behaviors via neurological activity. As expected, administration of MCH (10–25 μg/30 μl, i.n.) exerted dose-dependent antinociceptive effects against thermal noxious stimuli in the hot-plate test.

Additionally, both a single treatment (10 μg/30 μl, i.n.) and the repeated administration (10 μg/30 μl, once per day for 5 days, i.n.) of MCH inhibited pain behaviors after CFA-induced inflammatory pain, suggesting a role of MCH as a potential analgesic agent. In this model, the analgesic effects of i.n. MCH lasted for at least 60 min after its administration, and the effects were equivalent to those of diclofenac (10 mg/kg, i.p.), an NSAID. The present study also compared the effects of MCH (10 μg/30 μl, i.n.) with or without the presence of a MCHR1 antagonist, TC-MCH 7c (10 mg/kg, i.p.), and found that the anti-allodynic effects of MCH totally disappeared following pretreatment with the MCHR1 antagonist. These results indicate that the effects of MCH were dependent on ligand–receptor interactions rather than due to unspecific effects.

Because the pharmacological treatment of neuropathic pain remains challenging^4,36, the present study also examined the effects of MCH in the PSNL-induced neuropathic pain model. The repeated administration of MCH (10 μg/30 μl, once per day for 7 days, i.n.) produced a significant decrease in the mechanical allodynia induced by PSNL surgery and the degree of the anti-allodynic effect gradually increased over time.

Additionally, the present study aimed to characterize the endogenous analgesic mechanisms involved in MCH inhibition of CFA-induced inflammatory and PSNL-induced neuropathic pain using several receptor antagonists. The opioid receptor antagonist naltrexone blocked the anti-allodynic and analgesic effects of MCH in both pain models. Opioid receptors, the mu, kappa, and delta receptors, are widely distributed in the central and peripheral nervous systems. Opioids inhibit pain transmission mainly by interacting with these G protein-coupled receptors at presynaptic and postsynaptic sites in the dorsal horn of the spinal cord and by increasing descending inhibition from the brain and brainstem to the spinal cord³⁷. The endogenous opioid system also inhibits nociceptive transmission in the ascendant pathway, activates the descending inhibitory pathway, and modifies the limbic system to inhibit the emotional perception of pain^38–40. Some interactions between endogenous opioid and MCH systems have been proposed, including prevention of MCH-induced feeding by blockade of the kappa-, mu-, and delta-opioid receptors^41,42, and antagonism of MCH-stimulated hedonic responses by kappa-, mu-, and delta-opioid receptor antagonists⁴³. Further studies are needed to elucidate the roles of the endogenous opioid system in the modulation of pain by MCH.

Next, we demonstrated that the endocannabinoid system is also related to the analgesic effects of MCH, as the effects of i.n. MCH administration were markedly inhibited by a CB1R antagonist in both the inflammatory and neuropathic pain models. Several studies have suggested that there are interactions between the cannabinoid and MCH systems. For example, introduction of the endocannabinoid receptor ligand, 2-arachidonoyglycerol, into the rat lateral hypothalamus increases REM sleep and c-Fos expression in MCH neurons⁴⁴ and combined treatment with CB1R and MCH1R antagonists synergistically induces anti-obesity effects in diet-induced obese mice⁴⁵. The endocannabinoid system is an important regulator of the endogenous pain control system, which, along with the opioid system, plays an important role in the development and resolution of pain as well as in the affective components of pain^46,47. The cannabinoid system also inhibits synaptic transmission and controls synaptic plasticity in pain pathways via activation of the G protein-coupled CB1 and CB2 cannabinoid receptors^28,48. CB1Rs are widely distributed throughout the brain^48–50, whereas CB2 receptors are most highly expressed in peripheral cells⁴⁸. CBR activity inhibits ascending nociceptive transmission, activates the inhibitory descending pathway, and modifies the emotional aspects of pain^50,51. These effects of the endocannabinoid system on the control of pain occur at the peripheral, spinal, and supraspinal levels⁵².

As current knowledge regarding the importance of the cannabinoid system in pain modulation has progressed, the mPFC has received an increasing amount of attention⁵³. It is now known that the descending inhibitory pain system^36,54 as well as higher centers of the brain, including the frontal lobe, anterior cingulate cortex, insula, and amygdala, are involved in the descending modulation of pain⁵⁴. The mPFC is responsible for a variety of executive functions and is closely connected to limbic areas (e.g., amygdala) that play key roles in emotion^24,55,56. Recent evidence has shown that pain-related neuroplastic changes in the mPFC–limbic network contribute to emotional responses and enhance pain sensitivity⁵⁷ and that there are functional and structural abnormalities in the mPFC regions of patients with pain⁵⁸. Additionally, rodent models of arthritis⁵⁹ and neuropathic pain⁶⁰ have observed reduced activity in output neurons in the mPFC. CB1Rs in the mPFC are thought to play an important role in the abnormal synaptic inhibition observed in pain models⁶¹; therefore, the restoration of mPFC outputs using CB1R agonists would be expected to alleviate pain^61,62.

Based on the abovementioned findings, the present authors postulated that CB1Rs in the mPFC would be associated with the effects of MCH. MCH is effectively delivered and distributed to regions around the mPFC following the i.n. application of FITC-labeled MCH^7,18, and it is known that MCH1Rs are expressed in this region as well¹⁵. In the present study, MCH treatment significantly restored reductions in the number of CB1Rs in the mPFC in the CFA group. Interestingly, the changes in CB1Rs in this region were negatively correlated with pain behaviors, which implies that the increased expression of CB1Rs may reflect decreases in pain. Furthermore, the MCH-induced increases in CB1Rs were blocked by a MCH1R antagonist, which indicates that MCH1Rs mediate the effects of MCH.

The present study also examined the activation of CB1Rs by assessing the number of cells that co-expressed c-Fos/CB1R. Although a decrease in activated CB1Rs was not evident, MCH treatment increased the activation of CB1Rs compared to the Control and CFA groups, and these changes were correlated with reductions in pain behaviors. It has been previously reported that microinjections of MCH into the dorsal raphe nucleus and basolateral amygdala induce depression-like behaviors^10,63. However, contrary to previous findings, analyses of the TST and FST revealed that treatment with an anti-hyperalgesic dose of MCH for either 7 or 14 days decreased depression-like behaviors in the CFA and PSNL pain models. Thus, it is possible that changes in CB1Rs in the mPFC play important roles in the modulation of emotional deficits as well as in pain sensitivity.

Neuroimaging studies have shown that pain in humans is associated with activity in the mPFC and PAG⁶⁴. These brain areas exhibit strong connectivity and form a network that allows for the top-down modulation of nociception⁶⁵. Additionally, robust projections between the mPFC and PAG have been identified in rodents⁶⁶, and the PAG is known to play an essential role in descending modulation through projections onto the rostral ventromedial medulla, which can exert either facilitating or inhibitory effects on the spinal cord^67–69. Although MCH1Rs are not expressed in the mouse PAG¹⁵, it is possible that molecular changes in the mPFC (i.e., increased expression levels and the activation of CB1Rs) affect PAG activity and result in the modulation of pain transmission. In the present study, CB1R levels in the mPFC were highly correlated with those in the vlPAG and dlPAG. It is well known that CB1Rs activate the descending inhibitory pathway via the inhibition of GABA release in the PAG^48,70. The immunofluorescence analyses performed in the present study demonstrated that the expression levels of CB1Rs within the vlPAG and dlPAG were lowered in the inflammatory pain model, which is consistent with previous findings^71–73. In contrast, the decreases in the expressions of CB1Rs were attenuated following i.n. MCH treatment but were inhibited by pretreatment with a MCH1R antagonist. Furthermore, MCH treatment enhanced the activation of CB1Rs in these areas. Consequently, it can be postulated that the i.n. administration of MCH modulated higher centers of the brain, including the mPFC, which subsequently induced changes in the PAG and resulted in the activation of the descending inhibitory pathway. However, further studies are required to gain a better understanding of the contribution of the cannabinoid system to the analgesic effects of MCH.

In the present study, MCH was delivered via i.n. administration, which allowed the peptides to rapidly enter the brain and directly bypass the blood–brain barrier^74–76. The intranasally administered FITC-labeled MCH (10 μg/30 μl) was distributed to various brain regions, including the PFC, nucleus accumbens, and hippocampus, within 30 min of administration (data not shown). This finding is consistent with those of previous studies showing that peptides delivered via i.n. administration are detected at maximal levels after 30 min and can be detected after 6 h^76,77. Although changes in CB1R levels were observed in the mPFC, vlPAG, and dlPAG following the administration of i.n. MCH, other brain regions (including the spinal cord, striatum, nucleus accumbens, and amygdala) may also contribute to the blockade of allodynic and hyperalgesic abnormalities associated with inflammatory and neuropathic pain.

Intranasally administered MCH can spread to various parts of the brain as well as to the blood stream. In addition to its localization in the CNS, MCH1Rs are also found in peripheral tissues, such as the thyroid, kidney, adipose tissue, lung, testes, and tongue^78,79, where they perform various physiological functions. Although the peripheral actions of MCH are not yet well understood relative to its functions in the CNS, lower levels of intestinal inflammation⁸⁰ and intestinal tumorigenesis⁸¹ and an increased mortality against intestinal pathogens⁸² are evident in MCH KO mice, which suggests that the peripheral functions of MCH should be considered⁸².

The present study also investigated whether the administration of MCH (i.n.) for 5 consecutive days would induce intestinal inflammation by measuring macroscopic damage in terms of hyperemia, changes in the thickness of the colon wall, and the extent of ulceration⁸⁰. There were no abnormal signs of inflammation in the intestines of animals in the CFA-induced pain model (Supplementary Fig. 2). The present study also examined the effects of repeated MCH treatments on body weight and food intake in healthy mice for 3 weeks because MCH was found to increase body weight and food intake in previous studies^29,45. However, there were no significant increases in body weight or food intake in MCH-treated mice compared with controls (data not shown). These findings suggest that i.n. MCH treatment could be used as an analgesic agent without side effects, such as weight gain. In any event, further long-term observation will be necessary to monitor the occurrence of possible adverse events.

In conclusion, the present data demonstrated potential analgesic effects of MCH via i.n. administration on both inflammatory and neuropathic pain, thus suggesting new potential therapeutic applications of this peptide in addition to its previously described effects.