Differential Peripheral Blood Glycoprotein Profiles in Symptomatic and Asymptomatic COVID-19

Area of Science:

• Biochemistry
• Immunology
• Proteomics

Background:

• Glycosylation is a key post-translational modification affecting protein structure and function.
• Understanding differential protein glycosylation in infectious diseases is crucial for diagnosis and treatment.

Purpose of the Study:

• To analyze differential protein glycoisoform distributions in serum from COVID-19 patients and compare them with other conditions.
• To identify specific glycoprotein profiles predictive of symptomatic COVID-19 infection.

Main Methods:

• Liquid chromatography and mass spectrometry for glycopeptide quantification.
• Artificial intelligence for high-throughput data processing.
• Machine learning (LASSO regression, K-means clustering) for predictive modeling.

Main Results:

• Significant differences in 374 out of 597 glycopeptides between symptomatic and asymptomatic COVID-19 patients.
• Overlapping and distinct glycopeptide profiles were observed when comparing COVID-19 with sepsis and common cold infections.
• Machine learning models achieved high accuracy (up to 100%) in predicting symptomatic COVID-19 based on glycoprotein profiles.

Conclusions:

• Glycoprotein alterations in severe COVID-19 are partly a consequence of systemic inflammation but also include COVID-19-specific changes.
• Glycoproteomic biomarkers show potential for biomedical understanding and clinical management of severe acute infections.
• Specific glycoisoform changes may indicate COVID-19-specific consequences or predisposition to severe disease.