Inflammatory Proteomic Signatures of Cardiometabolic Traits and the Contribution of Visceral Adipose Tissue: A Cross-Sectional and Network Proteomic Analysis.

BACKGROUND: Visceral adipose tissue (VAT), fat located within the abdominal cavity, is strongly associated with systemic inflammation and poor cardiometabolic health. OBJECTIVES: This study investigates the cross-sectional relationships between multiple cardiometabolic traits, including VAT, and proteomic-based inflammatory signatures. METHODS: Body adiposity distribution quantified using dual-energy X-ray absorptiometry (DXA), cardiometabolic traits, and plasma proteomics inflammation panel (Olink Explore 384) were measured in a discovery cohort from the Vitamin D and Omega-3 Trial (VITAL; n = 525) and a replication cohort from the Cocoa Supplement and Multivitamin Outcomes Study (COSMOS; n = 371). We derived inflammatory proteomic markers of VAT, systolic blood pressure (SBP), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), fasting glucose, and insulin resistance (Homeostasis Model Assessment of Insulin Resistance; HOMA-IR). Inflammatory proteomic markers were identified via linear regression at a false discovery rate (FDR) < 0.05. RESULTS: VAT showed the strongest inflammatory proteomic profile, with 221 associated proteins (62% of those tested) and effect estimates of up to β = 0.59 (SE = 0.03). Proteins associated with the cardiometabolic traits showed directions of association largely consistent with those observed for VAT, except for HDL-C, which showed inverse associations. After adjusting the regression models for VAT, most protein associations with glucose and SBP (>97%), HOMA-IR (70%), TG (62%), and HDL-C (56%) were no longer statistically significant suggesting that VAT accounts for a substantial portion of the shared variance in these associations. To further characterize these shared molecular associations, we examined the Bayesian network architecture of 86 proteomic markers common to all cardiometabolic traits. Some of the VAT-attenuated protein signatures with high centrality were TGFB1, PDLIM7, COLEC12, and LAIR1. CONCLUSIONS: These hub-like proteins may reflect shared inflammatory pathways linking VAT with cardiometabolic traits and provide hypotheses for future mechanistic and therapeutic investigation.