Immunoglobulin G N-glycosylation predicts long-term risk of ischemic stroke: a nested case-control study.

BACKGROUND: Ischemic stroke (IS) ranks as the leading cause of disability and the second cause of mortality in adults worldwide. This investigation aimed to describe the fingerprint of immunoglobulin G (IgG) N-glycans and feature of circulating inflammatory cytokines that were associated with the progression of IS, as well as to screen potential biomarkers for screening individuals at higher IS risk. METHODS: A prospective cohort, Liaocheng City Stroke Risk Cohort (LCSRC), was launched in December 2014, in which 1,599 healthy participants were enrolled from Liaocheng People's Hospital (LPH). During nine years of follow-up, 67 participants who were diagnosed with IS were enrolled in the case group. Meanwhile, 201 individuals who did not develop IS were randomly selected as the controls. The hydrophilic interaction chromatography based on ultra-performance liquid chromatography (HILIC-UPLC) was utilized to detect the IgG N-glycans profiles in serum. A prediction model for the occurrence of IS was constructed using a logistic regression. RESULTS: The levels of GP1, GP2, GP3, GP4, GP6, GP7, GP11, GP13, GP19 and GP24 were considerably higher in the IS group compared with the controls, whereas GP5, GP8, GP14 and GP18 were lower among the IS cases. In addition, compared to the controls, the IS patients had substantially higher levels of matrix metalloproteinase-9 (MMP-9), tumor necrosis factor-α (TNF-α), and high-sensitivity C-reactive protein (hs-CRP). In terms of four major features of IgG N-glycosylation, a high level of fucosylation, monogalactosylation, and digalactosylation was significantly associated with a decreased risk for IS occurrence. Moreover, the glycosylation-based prediction model was of promising capability to predict IS risk with an area under the curve (AUC) of 0.756 (95%CI: 0.696 to 0.815). CONCLUSIONS: Our findings, on the basis of nine-year follow-up, demonstrated that the decreases in fucosylation and galactosylation of IgG are significantly associated with the development of IS, potentially mediated by upregulation of chronic inflammatory processes. IgG N-glycosylation fingerprint might serve as a potential biomarker for predicting long-term risk of IS.