In:
Circulation, Ovid Technologies (Wolters Kluwer Health), Vol. 130, No. suppl_2 ( 2014-11-25)
Kurzfassung:
BACKGROUND: Metabolic syndrome is associated with increased risk for cardiovascular disease and type 2 diabetes, but the molecular mechanisms underpinning its constituent risk factors are unclear. We sought to identify predictive markers of metabolic risk by testing for associations of lipids and metabolites with longitudinal changes in metabolic traits. METHODS: Discovery liquid chromatography-tandem mass spectrometry profiling of 154 lipids and gas chromatography-mass spectrometry profiling of 119 metabolites was conducted on plasma samples from 554 Framingham Heart Study participants with baseline and follow-up clinic examinations (5-7 years later) at which body mass index (BMI), triglycerides (TG), HDL cholesterol (HDL-C), and glucose were measured. Analytes were tested for association with longitudinal changes (Δ) in metabolic traits using general linear models, and multimarker panels were selected using forward selection. RESULTS: Our single marker analyses revealed distinct signatures of longitudinal changes in metabolic risk factors. Markers associated with ΔBMI were 1-palmitoyl lysophosphatidic acid (LPA 16:0; p=2.8x10 -4 ), its precursor lysophosphatidylcholine (LPC 16:0; p=6.7x10 -5 ), and four other LPC species. Subclasses of sphingomyelins (SMs) were associated with changes in TG, HDL-C, and glucose. Sphingadiene (d18:2) variants were associated with ΔHDL-C [SM(d18:2/24:1), p=2.8x10 -6 ; SM (d18:2/20:0), p=1.2x10 -4 ]. Canonical SMs were associated with ΔTG [SM (d18:1/16:0), p=1.8x10 -5 ; SM (d18:1/17:0), p=2.1x10 -5 ]. Two dihydrosphingosine (d18:0) variants were associated with Δglucose [SM 36:0, p=5.0x10 -5 ; SM (d18:0/24:0), p=3.2x10 -4 ]. Metabolite markers for ΔTG included quinic acid (p=9.1x10 -5 ) and sitosterol (p=1.0x10 -4 ). Top markers were selected in multimarker panels that explained a significant proportion of longitudinal change in each metabolic trait (2.5-15.3%) beyond baseline covariates. CONCLUSIONS: Using lipidomic and metabolomic profiling in parallel, we identified lipid-centric signatures of longitudinal changes in metabolic traits and demonstrated their predictive power. Our results suggest that specific derangements in lipid metabolic pathways may underlie metabolic risk.
Materialart:
Online-Ressource
ISSN:
0009-7322
,
1524-4539
DOI:
10.1161/circ.130.suppl_2.18728
Sprache:
Englisch
Verlag:
Ovid Technologies (Wolters Kluwer Health)
Publikationsdatum:
2014
ZDB Id:
1466401-X
