Related Experiment Video
Updated: Jul 9, 2026
Immunoglobulin G N-Glycan Analysis by Ultra-Performance Liquid Chromatography
Published on: January 18, 2020
Immunoglobulin G N-glycosylation predicts outcome in sepsis caused by pathogenic Gram-negative bacteria and
Huachen Wang1,2, Yan Zhang3, Anlu Ouyang3
1Institute of Infectious Diseases, The Second Hospital of Tianjin Medical University, Tianjin, China.
Background:
Sepsis remains a critical global health challenge with high mortality. Rapid distinction between Gram-positive and Gram-negative pathogens is critical for empiric antibiotic selection, yet reliable biomarkers for such pathogen stratification are lacking. Immunoglobulin G (IgG) N-glycosylation modulates inflammatory responses in various diseases, suggesting its potential role in sepsis pathogenesis and prognosis.
Methods:
In this nested case-control study, 180 septic patients (100 Gram-negative, 80 Gram-positive) and 100 healthy controls were enrolled. IgG N-glycosylation was analyzed using Hydrophilic interaction chromatography based on ultra-performance liquid chromatography (HILIC-UPLC). Inflammatory cytokines and clinical parameters were collected. Least absolute shrinkage and selection operator (LASSO) and logistic regression were used to identify glycan biomarkers and construct predictive models for pathogen type and 90-day mortality. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC).
Results:
In patients with Gram-negative sepsis, levels of 12 glycan peaks (A2B, M5, A2G1, FA2[3]G1, FA2[3]BG1, A2G2, A2BG2, FA2G2, A2G2S1, FA2G2S1, A2G2S2, A2BG2S2) were significantly decreased, while levels of FA1, A2, FA2, FA2B, FA2BG2, and FA2FG2S1 were increased (all P < 0.05, q < 0.05), compared to patients with Gram-positive sepsis. A model incorporating glycan peaks with routine clinical markers showed excellent discrimination (AUC = 0.931 in training, 0.917 in validation). Significant differences in fucosylation, sialylation, agalactosylation (G0), and digalactosylation (G2) levels were observed between septic survivors and septic non-survivors. For 90-day mortality, glycan FA2 was a strong independent predictor (AUC = 0.792), outperforming the SOFA score (AUC = 0.673). A combined model of FA2 and SOFA score improved predictive accuracy (AUC = 0.820).
Conclusion:
IgG N-glycosylation profiles serve as effective biomarkers for distinguishing pathogenic Gram-negative from Gram-positive sepsis and for predicting mortality. Integration of glycan peaks with clinical scores enhances risk stratification, highlighting their utility as a complementary tool in sepsis prognostication.
