Electroacupuncture targets D-serine-related synaptic deficits in CUMS rats through modulating the astrocytic USP4/CB1R axis.

AIMS: This study aimed to identify key molecular changes in the hippocampal CA1 region underlying the antidepressant effects of electroacupuncture (EA) in a rat model of chronic unpredictable mild stress (CUMS), and to explore the potential involvement of astrocytic signaling pathways in synaptic regulation.

METHODS

To investigate the antidepressant mechanism of electroacupuncture (EA), a chronic unpredictable mild stress (CUMS) rat model was established. Seventy-six 7-week-old male Sprague-Dawley rats were randomly divided into a normal control group (NC, n = 16) and a CUMS modeling group (n = 60). After 3 weeks of CUMS, the modeling rats were further randomized into CUMS, paroxetine (Par), and EA groups (n = 16 each). EA was administered daily at GV20 ("Baihui") and GV29 ("Yintang") acupoints for two weeks. Sucrose preference test (SPT) and open field test (OFT) were performed at baseline, after modeling, and on days 7 and 14 of intervention. Hippocampal CA1 tissues were collected for Nissl staining and transmission electron microscopy (TEM) to assess neuronal and synaptic structural changes. To explore the underlying molecular mechanisms, an iTRAQ-based proteomic analysis was performed on CA1 tissues from the NC (n = 2), CUMS (n = 3), and EA groups (n = 3). Differentially expressed proteins were screened, and USP4 and CB1R were identified as candidate molecules. Subsequently, the expression, physical interaction, ubiquitination status, and downstream signaling (PLC/IP3/D-serine) of USP4 and CB1R were validated using western blot, co-immunoprecipitation (Co-IP), ubiquitination assays, immunofluorescence and real-time quantitative PCR.

RESULTS

Compared with the CUMS model group, both EA and paroxetine treatments significantly reversed CUMS‑induced depressive‑like behaviors. Histological and ultrastructural assessments showed that EA alleviated neuronal damage, restored synaptic architecture (clearer synaptic interfaces, increased presynaptic vesicles, and recovered postsynaptic density thickness/electron density), and normalized NMDAR protein levels. To explore the underlying mechanisms, iTRAQ‑based proteomic screening identified USP4 and CB1R as candidate molecules. Western blot and immunofluorescence confirmed that EA upregulated CB1R and GFAP expression in CA1 astrocytes while downregulating USP4. Co‑immunoprecipitation revealed that CUMS disrupted the physical interaction between USP4 and CB1R, which was restored by EA. Ubiquitination assays further showed that EA suppressed both total (FK2) and K48‑linked (P4D1) ubiquitination of CB1R induced by CUMS. Consequently, EA activated the astrocytic PLC/IP₃/D‑serine signaling axis, with key components significantly elevated in the EA group compared with the CUMS group.

CONCLUSION

This study demonstrates that the astrocytic USP4/CB1R regulatory node in the hippocampal CA1 region is a critical pathological hub underlying CUMS-induced synaptic dysfunction. Mechanistically, CB1R stability is regulated by USP4-mediated deubiquitination, and the CB1R-dependent PLC/IP₃/D-serine signaling cascade functions as a glial checkpoint for synaptic homeostasis. EA exerts its antidepressant effects by restoring this checkpoint, thereby rescuing synaptic structure and function.