Lipidomic Signatures of Insulin Resistance and Risk of Diabetes: A Longitudinal Study in American Indians
Abstract Body: Background: To identify lipidomic signatures of insulin resistance (IR) and their associations with the risk of diabetes in American Indians. Methods: We used untargeted liquid chromatography-mass spectrometry to quantify1,542 lipid species (518 known) in fasting plasma samples from 1,438 Strong Heart Family Study participants attending two exams (∼5.5 years apart between 2001-2003 and 2006-2009), with follow-up extending up to 16 years. IR was assessed using the homeostasis model assessment of IR (HOMA-IR). Participants in the top HOMA-IR tertile were classified as IR, and those in the bottom tertile as insulin-sensitive (IS). We first identified IR-associated lipid species using mixed-effects logistic regression, adjusting for age, sex, study center, education, BMI, smoking, alcohol use, physical activity, systolic blood pressure, lipid-lowering medication use, and visit time (baseline vs. follow-up). We then examined the independent associations of these lipids with incident diabetes. Finally, mediation analyses were performed to assess the extent to which these lipids mediated the relationship between IR and diabetes. Results: We identified 675 lipid species (269 known), predominantly glycerolipids, glycerophospholipids, and sphingomyelins, which were significantly associated with IR (FDR<0.05). Specifically, twelve lipids, mainly glycerolipids, were associated with an increased 5.5-year risk of diabetes (ORs per 1-SD increase: 1.20 to 2.12). Of these, eight lipids also showed an increased risk at 16-year follow-up (ORs per 1-SD increase: 1.33 to 1.82). Mediation analysis further indicated that four lipids mediated the positive association between IR and diabetes risk (mediated proportions range: 8.6% to 18.9%). Conclusion: We have identified robust lipidomic markers of IR that are independently associated with an increased risk of diabetes, suggesting a key mediating role in the IR-diabetes pathway.
Chen, Mingjing
(
Johns Hopkins University
, Baltimore , Maryland , United States )
Miao, Guanhong
(
University of South Florida
, Tampa , Florida , United States )
Wen, Xiaoxiao
(
University of South Florida
, Tampa , Florida , United States )
Zhang, Ying
(
University of Oklahoma Health Campus
, Oklahoma , Oklahoma , United States )
Cole, Shelley
(
Texas Biomedical Research Institute
, San Antonio , Texas , United States )
Zhao, Jinying
(
University of South Florida
, Tampa , Florida , United States )