Characterization of plasma lipidomics in adolescent subjects with increased risk for type 1 diabetes in the DiPiS cohort

Introduction: Type 1 diabetes (T1D) is caused by the destruction of pancreatic islet beta cells resulting in total loss of insulin production. Recent studies have suggested that the destruction may be interrelated to plasma lipids.

Objectives: Specific lipids have previously been shown to be decreased in children who develop T1D before four years of age. Disturbances of plasma lipids prior to clinical diagnosis of diabetes, if true, may provide a novel way to improve prediction, and monitor disease progression.

Methods: A lipidomic approach was utilized to analyze plasma from 67 healthy adolescent subjects (10-15 years of age) with or without islet autoantibodies but all with increased genetic risk for T1D. The study subjects were enrolled at birth in the Diabetes Prediction in Skåne (DiPiS) study and after 10-15 years of follow-up we performed the present cross-sectional analysis. HLA-DRB345, -DRB1, -DQA1, -DQB1, -DPA1 and -DPB1 genotypes were determined using next generation sequencing. Lipidomic profiles were determined using ultra-high-performance liquid chromatography quadrupole time-of-flight mass spectrometry. Lipidomics data were analyzed according to genotype.

Results: Variation in levels of several specific phospholipid species were related to level of autoimmunity but not development of T1D. Five glycosylated ceramides were increased in insulin autoantibody (IAA) positive adolescent subjects compared to adolescent subjects without this autoantibody. Additionally, HLA genotypes seemed to influence levels of long chain triacylglycerol (TG).

Conclusion: Lipidomic profiling of adolescent subjects in high risk of T1D may improve sub-phenotyping in this high risk population.

Keywords: Autoantibodies; Autoimmunity; Lipidomics; Metabolomics; Prediction; Type 1 diabetes.