Odour-induced analgesia mediated by hypothalamic orexin neurons in mice

Abstract

Various folk remedies employ certain odorous compounds with analgesic effects. In fact, linalool, a monoterpene alcohol found in lavender extracts, has been found to attenuate pain responses via subcutaneous, intraperitoneal, intrathecal, and oral administration. However, the analgesic effects of odorous compounds mediated by olfaction have not been thoroughly examined. We performed behavioural pain tests under odourant vapour exposure in mice. Among six odourant molecules examined, linalool significantly increased the pain threshold and attenuated pain behaviours. Olfactory bulb or epithelium lesion removed these effects, indicating that olfactory sensory input triggered the effects. Furthermore, immunohistochemical analysis revealed that linalool activated hypothalamic orexin neurons, one of the key mediators for pain processing. Formalin tests in orexin neuron-ablated and orexin peptide-deficient mice showed orexinergic transmission was essential for linalool odour-induced analgesia. Together, these findings reveal central analgesic circuits triggered by olfactory input in the mammalian brain and support a potential therapeutic approach for treating pain with linalool odour stimulation.

Figure 1. Linalool vapour exposure induces analgesic effects in the hot plate test.

(A) Increase in pain threshold to a noxious heat stimulus during linalool vapour exposure. Among the odour molecules examined, linalool exposure significantly increased latency to pain responses during the hot plate test; n = 10 for each group. (B) Concentration-dependent increase of the analgesic effects; n = 10 for each concentration. The data set for 100% linalool was the same as the data for linalool group in (A) and shown again for comparison. (C) No significant changes of spontaneous activity under linalool vapour exposure; n = 9 for each group. MO, diluent mineral oil; 1% L, 1% linalool; 10% L, 10% linalool; 100% L, 100% linalool; Air, odourless air-exposed mice; Lin, linalool-exposed mice. Bars indicate the median of each group. *p < 0.05; **p < 0.01; ****p < 0.0001; n.s., not significant.

Figure 2. Attenuation of formalin-evoked pain responses during linalool vapour exposure.

(A) Time course of formalin-evoked pain responses among odourless air-exposed (white squares) and linalool-exposed (black squares) mice. Traces are presented as mean ± SEM. (B) A population analysis of linalool-induced analgesia during the formalin test. Linalool significantly suppressed pain responses during both Phases 1 and 2 during the formalin test. Air, odourless air-exposed mice; Lin, linalool-exposed mice; n = 18 for odourless air-exposed mice and n = 19 for linalool exposed mice. Bars in (B) indicate the median of each group. ***p < 0.005; ****p < 0.0001.

Figure 3. Anosmic model mice do not show linalool-induced analgesia.

(A) Time course of formalin-evoked pain responses among olfactory bulbectomy (OB) mice. Linalool-induced analgesia is not observed among OB mice. Pain responses are attenuated after subcutaneous morphine injection in OB mice during odourless-air exposure. (B) A population analysis of linalool-induced analgesia among OB mice. The time spent on pain behaviour is unchanged in the linalool-exposed group but significantly reduced in the morphine-administered, odourless air-exposed group during both Phases 1 and 2. (C) Time course of formalin-evoked pain responses among olfactory epithelium deprived (3-MI) mice. Attenuation of pain responses induced by linalool-exposure is not observed in 3-MI mice. (D) A population analysis of linalool-induced analgesia in 3-MI mice. Time spent on pain behaviour in linalool-exposed mice does not show significant differences compared with control mice. Air, odourless air-exposed mice; Lin, linalool-exposed mice; Mor, morphine-administered mice. Traces are presented as mean ± SEM. Bars in plots indicate the median of each group; n = 7 for odourless air-exposed OB mice; n = 7 for linalool-exposed OB mice; n = 5 for morphine-administered OB mice; n = 6 for odourless air-exposed 3-MI mice; n = 6 for linalool-exposed 3-MI mice.

Figure 4. Orexinergic transmission is essential for linalool odour-induced analgesia.

(A) Linalool odour exposure activates hypothalamic orexin neurons. Immunohistochemical analyses indicate that a subpopulation of orexin neurons (red) expressed c-Fos (green), even under odourless-air exposure (A1). The distributions of c-Fos positive (black dots) and c-Fos negative (white dots) orexin neurons are shown in (A2). (B) During linalool exposure, c-Fos expressing orexin neurons are widely distributed in the hypothalamus (B1, B2). (C) A population analysis indicates the number of c-Fos expressing orexin neurons significantly increased in the linalool-exposed group (p < 0.05). (D,F) Time course of formalin-evoked pain responses among orexin neuron-ablated (AB) mice (D) and orexin peptide-deficient (KO) mice (F). Attenuation of pain behaviours under linalool odour exposure was not observed in both mutant mouse strains. (E,G) Population analyses reveal linalool-induced analgesia disappearance in the orexin mutant mice. Traces are presented as mean ± SEM. Bars in plots indicate the median of each group. Air, odourless air-exposed mice; Lin, linalool-exposed mice; n = 5 in each group in (A–C); n = 5 for AB mice, n = 6 for KO mice in (D–G); f, fornix; ic, internal capsule. Bars in pictures indicate 200 μm.

Figure 5. Linalool odour exposure does not induce aversive stress responses.

(A) Innate odour preference test. Time spent investigating the odourant-scented filter paper is plotted. Kruskal-Wallis test reveal the significant difference among the three groups (H = 8.447, p < 0.05). Bars indicate the median. n = 14 for DDW group, n = 7 for linalool (Lin) group, n = 8 for TMT group (n = 8). *p < 0.05; n.s., not significant. (B) Innate odour avoidance test. Staying time ratio (see Methods) before and after linalool (B left) or TMT (B right) ventilation is shown. n = 6 (linalool), n = 7 (TMT). *p < 0.05; n.s., not significant (Wilcoxon matched pairs test). (C) Corticosterone assay. Plasma corticosterone concentration after 60 min odour exposure is shown. Data are represented as mean ± SEM. n = 8 for Air group, n = 7 for linalool (Lin) group, n = 8 for TMT group). **p < 0.01; ***p < 0.001; n.s., not significant (ANOVA with post-hoc Tukey’s multiple comparisons test).