摘要
神經胜肽S (neuropeptide S)是由20個胜肽所組成，因為其N端殘基在不同物種中都是絲胺酸 (serine)，故命名為神經胜肽S。神經胜肽S主要是由腦幹製造，並投射至含有神經胜肽S受體 (neuropeptide S receptor; NPSR)之腦區以產生多種生理作用，包括焦慮、覺醒、活動力、食物攝取、成癮以及止痛等。目前神經胜肽S已經被研究出在脊髓給予時可以發揮止痛效果以及與壓力所致行為改變有關，主要作用於神經激肽S受體上，然而，它的實際作用機轉仍不清楚，但食慾素 (orexin) 系統被認為有參與其中的機制。在先前的研究中我們發現食慾素可藉由活化在中腦環導水管灰質腹外側區 (ventrolateral periaqueductal gray; vlPAG)中物質P (substance P)神經元上的食慾素1受體 (orexin 1 receptor; OX1R)來促進物質P分泌；物質P活化同個核區內的麩胺酸神經元 (glutamatergic neuron)，並釋放出大量麩胺酸 (glutamate) 作用在突觸周圍的第5型代謝性麩胺酸受體 (type 5 metabolic glutamate receptor; mGluR5) 產生內生性大麻酯 (endocannabinoid; eCB)，並藉由作用於γ-氨基丁酸 (γ-aminobutyric acid; GABA) 神經終端突觸上的大麻酯1受體 (cannabinoid 1 receptor; CB1R) 而抑制GABA釋放，出現逆行性去抑制 (disinhibition) 現象。 因此，我們在這篇論文中使用藥理學的方法驗證神經胜肽S活化在下視丘側區 (lateral hypothalamus; LH) 的食慾素神經元並釋放食慾素A至中腦環導水管灰質腹外側區，並活化此區內含有神經激肽的神經 (neuroknin-containing neuron) 以釋放物質P，並經由麩胺酸-mGluR5-內生性大麻酯-CB1R的訊息路徑來產生止痛現象，而且此止痛機轉是藉由神經胜肽S所誘導。 首先，我們發現在腦室內注射神經胜肽S可以顯著增加縮腳潛伏期 (withdrawal latency) 的時間且不會影響老鼠的活動力，而此止痛效果可以被[tBu-D-Gly5]NPS所阻斷。不過在vlPAG內注射神經胜肽S並沒有產生顯著的止痛效果，代表神經胜肽S的止痛作用並不適藉由vlPAG來調控。第二，腦室注射神經胜肽S的止痛作用能被在vlPAG內注射SB-334867、L-703,606、MPEP和AM251所阻斷，指出OX1R、NK1R、mGluR5和CB1R皆參與在神經胜肽S的止痛作用中。第三，我們過去已經建立一個對小鼠使用束縛壓力 (restraint stress) 所造成之緊張所致止痛的模式，即將小鼠束縛在50 ml的離心管後其縮腳潛伏期會顯著增長，而在實驗中這樣的止痛模式可以藉由先在腦室投與[tBu-D-Gly5]NPS被抑制掉，而其使用的劑量並不會影響老鼠的活動力。第四，經過束縛壓力處理過後的小鼠相較於控制組其LH均質液內神經胜肽S蛋白質量顯著增加。 本篇論文的實驗結果和先前的電生理和藥理實驗結果可以推測出神經胜肽S所產生的止痛是活化LH內食慾素神經並釋放食慾素至vlPAG，而食慾素活化物質P神經並促進物質P分泌且活化麩胺酸神經釋放麩胺酸，進而活化mGluR5，產生內生性大麻酯逆行性去抑制。既然緊張所致止痛可以被NPSR、NK1R、mGluR5和OX1R拮抗劑所抑制，由此證明緊張所致止痛是在束縛壓力時釋放神經胜肽S產生止痛效果並促進食慾素釋放至vlPAG內，再經由釋放物質P透過mGluR5-內生性大麻酯機制造成去抑制現象。
並列摘要
Neuropeptide S (NPS), an icosapeptide identified in 2004, was named due to its conserved N-terminal residue, serine, in all species. NPS-containing neurons are mainly distributed in the brainstem and projects to many brain regions. Thus, NPS plays a role in several physiological functions like anxiety, arousal, locomotion, food intake, addiction and antinociception via its specific Gq/Gs-protein-coupled- receptor, neuropeptide S receptor (NPSR). NPS was antinociceptive when given at the supraspinal level while its antinociceptive action mechanism(s) remain unclear. Previously, we have found that orexin can produce antinociceptive effect through the activation of orexin 1 receptors (OX1Rs) on substance P neuron in the ventrolateral periaqueductal gray (vlPAG) and release substance P. Substance P then, via neurokinin 1 receptors (NK1Rs), activates glutamatergic neurons in the vlPAG to release massive glutamate that activates perisynaptic metabotropic glutamate receptor 5 (mGluR5), yielding endocannabinoids that engage on the cannabinoid 1 receptor (CB1R) of presynaptic GABAergic terminals to inhibit GABA release, producing retrograde disinhibition in the vlPAG. NPS has been found to activate orexin-expressing neurons in multiple hypothalamic regions, suggesting this OX1R-initiated signaling cascade mediated by substance P (NK1R)-glutamate (mGluR5)-endocannabinoid (CB1R) in the vlPAG contributes to NPS-induced antinociceptive effect. We, therefore, validated a hypothesis in this study that NPS activates orexin neurons in the LH and release orexin A to the vlPAG, and then activates neurokinin-containing neurons in the vlPAG to release substance P and induce antinociception through the glutamate-mGluR5-endocannabinoid-CB1R signaling. We also hypothesized this mechanism may contribute to stress-induced analgesia (SIA) since it has been reported that NPS plays a role in stress-related behaviors, and orexin, substance P and mGluR5 are involved in SIA. The SIA model was induced by giving mice a 30-min restraint stress and the antinociceptive response was accessed by the withdrawal latency of mice in the hot-plate test. Drugs were administered by either intracerebralventricular (i.c.v.) or intra-vlPAG (i.pag.) microinjection. Besides, the level of NPS in the homogenate of the LH in restrained mice was compared with un-restrained mice by western blot analysis. First, we found that i.c.v., but not i.pag. microinjection of NPS significantly increased the withdrawal latency at doses that did not affect the spontaneous locomotor activity, and this effect was reversed by i.c.v. [tBu-D-Gly5]NPS, an NPSR antagonist. Second, this analgesic effect was blocked by, i.pag. microinjection of an antagonist of OX1Rs (SB-334867), NK1Rs (L-703,606), mGluR5s (MPEP) and CB1Rs, suggesting the NPSR in the brain and the OX1R, NK1R, mGluR5 and CB1R in the PAG are involved in the analgesic effect of NPS. Third, we have established a SIA model induced by acute restraint stress in mice, i.e. the withdrawal latency of the mouse restrained in a 50-ml centrifuge tube was significantly longer than in the unrestrained group. This SIA was significantly prevented by i.c.v. pretreatment with [tBu-D-Gly5]NPS at the dose that did not affect the locomotor activity. Forth, the NPS protein level in the LH homogenate of restrained mice was significantly higher than in the unrestrained control group. The latter two findings suggest an acute restraint stress activates NPS neurons, releasing NPS in the lateral hypothalamus to induced analgesia. The results in this study and our previous electrophysiological and pharmacological results suggest that NPS produces supraspinal analgesia through activating orexin neurons in the LH and release orexin to the vlPAG. Orexin activates substance P neurons and release substance P which acts on glutamatergic neurons in the vlPAG to release glutamate that activates mGluR5, resulting in endocannabinoid retrograde disinhibition in the vlPAG. Besides, since SIA can be blocked by NPSR, NK1R, mGluR5 and OX1R antagonists, it is suggested that SIA is mediated by NPS that is released during restraint stress to induce analgesia via releasing orexin that produces disinhibition in the vlPAG via the orexin-substance P-mGluR5-endocannabinoid mechanism.