The association between plasma IgG N-glycosylation and viral encephalitis in children: a hospital-based case-control study

BACKGROUND: Viral encephalitis (VE) is an acute inflammatory disease caused by viral infection. Children are at a significantly higher risk of developing VE than adults. Immunoglobulin G (IgG) N-glycosylation plays a key role in regulating the balance between anti-inflammatory and pro-inflammatory responses. This study aimed to investigate the profile of IgG N-glycosylation among children with VE. METHODS: This case-control study included 117 children with VE and 117 healthy controls. Plasma IgG N-glycans were detected using hydrophilic interaction liquid chromatography with the ultra-high-performance liquid chromatography (HILIC-UPLC). An enzyme-linked immunosorbent assay (ELISA) kit was applied to detect inflammatory cytokines. Canonical correlation analysis (CCA) was performed to investigate the correlation between IgG N-glycans and inflammatory cytokines. The Least Absolute Shrinkage and Selection Operator (LASSO) and machine learning algorithms were used to identify the significant glycans. RESULTS: Among 24 initially detected glycans, 14 were significantly higher and 3 lower in children with VE compared to the healthy controls (P < 0.05). Moreover, children with VE showed lower levels of fucosylation (P < 0.001) and agalactosylation (G0) (P < 0.001), whereas higher levels of bisecting N-acetylglucosamine (GlcNAc) (P = 0.026), sialylation (P < 0.001), and monogalactosylation (G1) (P < 0.001). Notably, the level of bisecting GlcNAc was significantly higher in VE children who had a shorter hospital stay compared to those with a longer hospital stay (P = 0.034). The levels of interferon-gamma (IFN-γ), interleukin-2 (IL-2), IL-6, high-sensitivity C-reactive protein (hs-CRP), and tumor necrosis factor-α (TNF-α) were significantly higher in children with VE than in the healthy controls (P < 0.001). IgG N-glycans composition was strongly correlated with inflammatory cytokines (r = 0.849). GP1, GP3, GP6, GP7, GP9, GP16, GP21, GP22, GP23, and GP24 were identified by Boruta algorithms to develop a glycan-based diagnostic model for VE. The areas under the receiver operating characteristic curve (AUCs) were 0.987 for the training set and 0.998 for the validation set. CONCLUSION: IgG N-glycosylation may contribute to the pathogenesis of VE. IgG N-glycans may act as potential biomarkers for monitoring VE in high-risk children.