The immune dysregulation landscape and dynamic regulation of competing endogenous RNAs in spinal cord injury.

Background: Spinal cord injury (SCI) triggers a complex secondary injury cascade that critically limits neural repair. Although microRNAs have been implicated in post-SCI inflammation and neural regeneration, the key competing endogenous RNA (ceRNA) network regulating immune dysregulation remains unclear. Methods: Peripheral blood miRNA data from six SCI patients and six healthy controls were combined with peripheral blood mRNA expression profiles from the GEO dataset (GSE151371) to identify differentially expressed miRNAs and mRNAs. Characteristic miRNAs were screened using LASSO and SVM-RFE machine learning algorithms, with core miRNAs subsequently validated via receiver operating characteristic (ROC) analysis. Their target mRNAs were predicted and validated, enabling the construction of a ceRNA regulatory network. Concurrent immune infiltration analysis was performed. WGCNA was applied to the core mRNAs to define immune-related co-expression modules and conduct functional enrichment. Furthermore, SCI samples were classified into two molecular subtypes with distinct immune microenvironments. Finally, RT-qPCR was used to quantify the expression of core miRNAs in validation cohorts, and functional experiments (CCK-8 and Transwell assays) were performed in C8-D1A astrocytes to assess the roles of selected miRNAs in cell proliferation and migration. Results: Six differentially expressed miRNAs were identified in SCI patients: four upregulated (hsa-miR-3680-5p, hsa-miR-410-3p, hsa-miR-6883-3p, hsa-miR-4508) and two downregulated (hsa-miR-3685, hsa-miR-6798-3p). ROC analysis verified the robust diagnostic value of these core miRNAs, with AUC > 0.85. RT-qPCR validation confirmed these expression patterns in the blood of SCI patients, consistent with bioinformatics predictions, p < 0.05. Their common targets (KLHL3, PRKCH, BACH2, MAP2K6, SIPA1L2) showed opposite expression patterns. WGCNA revealed immune-related modules enriched in T/B cell receptor and MAPK signaling pathways. Immune infiltration analysis indicated increased neutrophils and monocytes, with decreased lymphocytes. Notably, functional assays demonstrated that miR-3680-5p, miR-6798-3p, and miR-6883-3p did not directly affect astrocyte proliferation but significantly modulated astrocyte migration when co-cultured with macrophages, suggesting their involvement in astrocyte-macrophage crosstalk during SCI repair. Conclusion: We constructed an immune-related ceRNA network in SCI, identifying six core miRNAs that regulate immune pathways and astrocyte-macrophage interactions. These findings provide new insights into SCI pathogenesis and potential therapeutic targets.