ObjectiveTo establish a mouse model of trigeminal neuropathic pain induced by chronic constriction injury of the infraorbital nerve (CION), screen the optimal frequency of electroacupuncture (EA) for trigeminal neuralgia (TN), so as to explore the peripheral-central interaction mechanism of the “Hegu” (LI 4)-ventral posterior nucleus (VP)-trigeminal ganglion (TG) pathway.MethodsMale C57BL/6J mice were used to establish the TN model via the CION method. EA (2 Hz / 100 Hz / 2 /100 Hz) was applied to bilateral LI 4. Mechanical withdrawal thresholds (MWTs) of the vibrissal pad were measured using von-Frey filaments. Anxiety-like behaviors were assessed by the elevated plus maze (EPM) test, and depression-like behaviors by the tail suspension (TS) test and sucrose preference (SP) test. Neuronal activity in eight brain regions, including the dorsal raphe nucleus (DRN), VP, and spinal trigeminal nucleus caudalis (Sp5C), was evaluated by c-Fos immunofluorescence. Double immunofluorescence labeling was used to detect co-expression of c-Fos with 5-hydroxytryptamine (5-HT), glutamate (Glu), or γ-aminobutyric acid (GABA) in the DRN to identify activated neuron subtypes. Neural connectivity between “Hegu”-dorsal root ganglion (DRG)-spinal dorsal horn (DH)-VP and TG-Sp5C-VP was traced using chemical or viral tracers.ResultsCompared with the normal group, the MTs of mice were significantly decreased in the model group (P<0.0001). The MTs in the 2 Hz, 100 Hz, and 2 Hz/100 Hz EA groups were all increased (P<0.05, P<0.001), and the analgesic effect of the 2 Hz/100 Hz EA group was superior to that of the 2 Hz and 100 Hz EA groups (P<0.05, P<0.01). In the EPM test, the time spent in the open arms was significantly decreased in the model group (P<0.0001), while it was increased in the 100 Hz EA and 2 Hz/100 Hz EA groups (P<0.01, P<0.0001). Moreover, the time spent in the open arms was longer in the 2 Hz/100 Hz EA group than in the 2 Hz EA group (P<0.01). In the TS test, the immobility time was increased in the model group (P<0.0001), whereas EA reversed this change (P<0.0001). The 2 Hz/100 Hz EA group showed a more significant improvement in depression-like behavior than the 2 Hz and 100 Hz EA groups (P<0.05). In the SP test, the sucrose preference was decreased in the model group (P<0.0001), whereas EA reversed this change (P<0.001, P<0.0001). The 2 Hz/100 Hz EA group showed a more significant improvement in depression-like behavior than the 2 Hz and 100 Hz EA groups (P<0.01, P<0.05). Tracing results showed that after injection of a chemical tracer into LI 4, tracer signals were observed in the C6 and C7 DRG. After injection of a retrograde trans-multisynaptic virus into the VP, viral expression was observed in the DH, Sp5C, and TG. Immunofluorescence results showed that neuronal activity in the Sp5C, VP, ventral posteromedial nucleus (VPM), parafascicular nucleus (PaF), and DRN was increased in the model group (P<0.01, P<0.0001, P<0.05), while EA reversed these changes (P<0.0001, P<0.01, P<0.001). Compared with the normal group, the model group showed decreased 5-HT neuronal activity (P<0.0001) and increased Glu and GABA neuronal activities (P<0.0001, P<0.05), all of which were reversed by 2 Hz/100 Hz EA (P<0.0001, P<0.05).ConclusionBehavioral results show that 2 Hz/100 Hz EA is superior to single-frequency stimulation in improving mechanical hypersensitivity and associated negative affective behaviors in CION model mice, thus it is selected as the optimal intervention frequency for subsequent central response research. Under CION conditions, abnormal activation was observed in the Sp5C, VP, VPM, PaF, and DRN, accompanied by decreased activity of 5-HT neurons and increased activities of GABA and Glu neurons in the DRN, 2 Hz/100 Hz EA can reverse these phenomena. The tracing results indicated the existence of neuroanatomical connections of the LI 4-DRG-DH-VP and TG-Sp5C-VP pathways. These findings provide objective evidence for investigating the LI 4-VP-TG response mechanism underlying EA intervention in trigeminal neuralgia.