Abstract Body: Background. The immunoglobulin G (IgG) and plasma N-glycomes represent biomarkers of cardiovascular disease. However, few studies have examined relationships between glycomic profiles and an extensive list of cardiometabolic risk factors in healthy adults.
Hypothesis. Specific subsets of IgG and plasma glycans will associate with specific cardiometabolic traits.
Methods. Plasma and IgG glycans were analyzed using HILIC-UPLC and MALDI-TOF-MS (25 IgG N-glycans and 39 plasma N-glycans) in samples from 652 HERITAGE Family Study participants (57% female, 44% Black, 17-65 years old). Spearman correlations adjusted for age, sex, and race were performed between glycans and 68 cardiometabolic traits. Sparse multiple canonical correlation analysis (SMCCA) was used to identify a parsimonious set of glycans maximally related to cardiometabolic risk factors.
Results. Most (29/39) plasma glycans were associated with 10 or more traits (Fig 1A), while four IgG glycans were associated with 10 or more traits (Fig 1B). The SMCCA identified two canonical variates (CV) of groups of IgG and plasma glycans correlated with groups of phenotypes. CV1 was comprised of six IgG glycans, 12 plasma glycans, and 18 traits mostly related to body composition and indicative of worse cardiometabolic profiles (e.g., higher glycan abundance related to lower VO₂max and insulin sensitivity, higher visceral fat and BMI; Fig 2A). Conversely, CV2 was comprised of six IgG glycans, 13 plasma glycans, and 17 traits related to mostly body composition and lipids and indicative of better cardiometabolic profiles (e.g., higher glycan abundance related to lower triglycerides, total cholesterol, visceral fat, and BMI; Fig 2B). Seven plasma and four IgG glycans were found in both CVs, while five and six plasma glycans and two IgG glycans each were unique to CV1 and CV2, respectively. Although eight traits mainly related to body composition overlapped between CVs, traits related to insulin sensitivity and cardiorespiratory fitness were uniquely included in CV1 and lipoprotein-related traits in CV2.
Conclusions. We identified subsets of IgG and plasma glycans related to better or worse cardiometabolic profiles. Within the CVs, some associations between glycans and body composition traits overlapped. However, each CV is also related to a unique set of cardiometabolic traits. Future studies should investigate how changes in glycans relate to changes in cardiometabolic health.