A neural circuit from hypothalamic paraventricular oxytocin neurons to trigeminal nucleus caudalis GABAergic neurons modulates pain sensitization in a mouse model of chronic migraine.

Objectives: To investigate the role of a neural pathway from oxytocin (OXT) neurons in the hypothalamic paraventricular nucleus (PVN) to γ-aminobutyric acid (GABA) neurons in the trigeminal nucleus caudalis (TNC) in regulating pain sensitization in a mouse model of chronic migraine and to explore the underlying mechanisms.

Methods: A chronic migraine model was established by intraperitoneal injection of nitroglycerin (NTG, 10 mg/kg) on days 1, 3, 5, 7, and 9. The study consisted of four parts: PartⅠ: Wild-type C57BL/6J mice were divided into 4 groups (=6 in each), receiving single or repeated injection of NTG or saline, respectively. Immunofluorescence was used to detect c-Fos and OXT expression in the PVN. Part II: OXT-Cre transgenic mice (n=6) were used for anterograde monosynaptic tracing combined with RNAscope and immunofluorescence to identify neural projections from PVN OXT neurons to TNC GABA neurons. Part III: 30 male OXT-Cre transgenic mice were bilaterally injected Cre-dependent chemogenetic activation virus into the PVN. These mice were randomly divided into five groups, with six mice in each group. Mice in CNO group and control group were intraperitoneally injected with 0.1 mg/mL of clozapine -oxide (CNO) solution (1 mg/kg) and the same volume of isotonic normal saline, respectively. 3 hours after the injection, the brain tissues were harvest and c-Fos immunofluorescence staining was performed to verify the efficiency of chemogenetic activation virus. Mice in model control group and CNO activate model group were subjected to chronic migraine modeling, with bilateral TNC injection of isotonic normal saline and CNO, respectively, on day 10. The mice in Negative control group were bilaterally intraTNC injected with isotonic normal saline. After 30 minutes, the von Frey filament and acetone tests were used to assess the mechanical pain threshold and cold pain response time in the periorbital region of mice in these three groups. Part IV: 24 male OXT-Cre transgenic mice were bilaterally injected the Cre-dependent chemogenetic activation virus into the PVN. These mice were randomly divided into the four groups, with six mice in each group. Mice in model control group, CNO activate model group and Atosiban group were subjected to chronic migraine modeling. On day 10, mice in Negative control group and Model control group were intraperitoneally injected with isotonic normal saline, while mice in CNO activate model group and Atosiban group were intraperitoneally injected with CNO. After 15 minutes, mice in the Atosiban group were bilaterally intraTNC injected with atosiban, while mice in other three groups were bilaterally intraTNC injected with isotonic normal saline containing 1% dimethyl sulfoxide. After 15 minutes, the von Frey filament and acetone tests were used to assess the mechanical pain threshold and cold pain response time in the periorbital region of the mice. The GABA content in the bilateral TNC was detected by high-performance liquid chromatography (HPLC).

Results: Compared with the Single injection control and Repeated injection control groups, c-Fos neurons and percentage of c-FosOXT neurons in the PVN of Single injection NTG and Repeated injection model groups were increased (<0.05). Compared with Single injection model group, c-Fos neurons and percentage of c-FosOXT neurons in the PVN of Repeated injection model group was decreased (<0.05). The anterograde tracing virus and RNAscope combined immunofluorescence staining showed that PVN OXT neurons projected to TNC GABA neurons. As the results of immunofluorescence staining, compared with Control group, the percentage of c-FosOXT neurons in the PVN of CNO group was significantly increased (<0.05). In bilateral intraTNC drug administration experiments, compared with the Negative control group, the periorbital mechanical pain threshold decreased, and the periorbital cold pain reaction time increased in Model control and CNO activate model groups (<0.05). Compared with model control group, the periorbital mechanical pain threshold increased, and the periorbital cold pain reaction time decreased in CNO activate model group (<0.05). In intraperitoneal drug administration experiments, compared with the negative control group, the periorbital mechanical pain threshold decreased, and the periorbital cold pain reaction time increased in model control group, CNO activate model group and Atosiban group (<0.05). Compared with model control group, the periorbital mechanical pain threshold increased, and the periorbital cold pain reaction time decreased in CNO activate model group. Compared with CNO activate model group, the periorbital mechanical pain threshold decreased, and the periorbital cold pain reaction time increased in Atosiban group (<0.05). Compared with negative control group, Model control group and Atosiban group, GABA level of TNC in CNO activate model group was increased (<0.05).

Conclusions: PVN OXT neurons exert a descending facilitatory effect on GABAergic neurons in the TNC via OXT release, thereby ameliorating pain sensitization in chronic migraine.