Ps-p13-6: defining the relationship between immune traits and lipids in twinsuk

Objective: The primary aim of this work was to evaluate associations of distinct immune cell subtypes and circulating lipids in a general population setting. Methods: The associations between 3,855 immune traits and lipids (HDL-C, LDL-C, TC, TG, non-HDL-C) were examined in 1,491 individuals from the TwinsUK cohort. Immune traits were defined by high-dimensional flow-cytometry which enabled in-depth assessment of all major lymphocyte and myeloid subsets in peripheral blood. Information on immune cell subset frequency (FREQ) and mean fluorescence intensity (MFI) was available. A linear mixed model regression analysis adjusted for family structure, gender, age, BMI and batch effect was used to examine the associations between immune traits and lipids. A sub-analysis in 441 individuals and 122 lipid metabolites quantified using nuclear magnetic resonance spectroscopy was undertaken to investigate the associations of 40 significant immune traits and lipoprotein particles. Secondary analyses explored the associations of lipids and immune traits with pulse wave velocity (PWV). A mediation model was constructed to test mediation effects of an immune marker on the total effect of lipids on PWV. Results: Significant associations were observed between 46 MFI and 6 FREQ immune traits and lipids (Bonferroni P < 2.45x10–5). These included associations between circulating lipids and different myeloid or lymphoid subsets including monocytes, NK cells and T cells. The top associations involved the expression of transmembrane activator and calcium modulator and cyclophilin ligand interactor (TACI) cytokine on classical (cM) and intermediate (iM) monocyte subsets. Decreased levels of non-HDL-C [cMP = 6.98x10–10, iMP = 3.72x10–7], LDL-C (cMP = 3.28x10–7, iMP = 1.75x10–5) and TG [cMP = 1.18x10–9, iMP = 2.97x10–6], and increased levels of HDL-C (cMP = 3.64x10–12, iMP = 1.44x10–10) were associated with higher TACI expression on both cM and iM subsets. Significant associations were identified between TACI expression and 13 lipid metabolites (Bonferroni P < 1.36x10–4). All linked metabolites referred to HDL related measures. Increased apolipoprotein A1 levels, total cholesterol in HDL and HDL2 and large and medium HDL related particles were associated with increased TACI expression on cM and iM subsets. The strongest association found between total cholesterol in HDL2-C and TACI expression levels on cM (P = 3.57x10–6). Mediation analysis showed that TACI expression on monocytes partially mediates the effect (∼10%) of TG on PWV. Conclusions: This study provides the first comprehensive picture of lipid-immune cell associations in the circulatory system. Interesting novel immune cell subset alterations and insights into the relationship between lipids, PWV and immunity have extended our knowledge. Further work is required to dissect the role of immune responses in cardiovascular disease.