Metabolomics and Lipidomics Studies in Pediatric Type 1 Diabetes: Biomarker Discovery for the Early Diagnosis and Prognosis

Abstract

Aim: Type 1 diabetes (T1D) is an autoimmune disease with heterogeneous risk factors. Metabolic perturbations in the pathogenesis of the disease are remarkable to illuminate the interaction between genetic and environmental factors and how islet immunity and overt diabetes develop. This review aimed to integrate the metabolic changes of T1D to identify potential biomarkers for predicting disease progression based on recent metabolomics and lipidomics studies with parallel methodologies.

Methods: A total of 18 metabolomics and lipidomics studies of childhood T1D during the last 15 years were reviewed. The metabolic fingerprints consisting of 41 lipids and/or metabolite classes of subjects with islet autoantibodies, progressors of T1D, and T1D children were mapped in four-time dimensions based on a tentative effect-score rule.

Results: From birth, high-risk T1D subjects had decreased unsaturated triacylglycerols, unsaturated phosphatidylcholines (PCs), sphingomyelins (SMs), amino acids, and metabolites in the tricarboxylic acid (TCA) cycle. On the contrary, lysophosphatidylcholines (LPCs) and monosaccharides increased. And LPCs and branched-chain amino acids (BCAAs) were elevated before the appearance of islet autoantibodies but were lowered after seroconversion. Choline-related lipids (including PCs, SMs, and LPCs), BCAAs, and metabolites involved in the TCA cycle were identified as consensus biomarkers potentially predicting the development of islet autoimmunity and T1D. Decreased LPCs and amino acids indicated poor glycemic control of T1D, while elevated lysophosphatidylethanolamines and saturated PCs implied good glycemic control. Further pathway analysis revealed that biosynthesis of aminoacyl-tRNA, BCAAs, and alanine, aspartate, and glutamate metabolism were significantly enriched. Moreover, established cohort studies and predictive statistical models of pediatric T1D were also summarized.

Conclusion: The metabolic profile of high-risk T1D subjects and patients demonstrated significant changes compared with healthy controls. This integrated analysis provides a comprehensive overview of metabolic features and potential biomarkers in the pathogenesis and progression of T1D.

Figure 1

Flowchart of searching and filtering metabolomics and lipidomics studies on childhood T1D.

Figure 2

Heatmap of metabolite changes during the development of childhood T1D. Note: Four different time dimensions were included: T1—subjects' ages, from birth to 10 years old; T2—autoimmunity stages, within 1 year before seroconversion, at seroconversion, and within 1 year after seroconversion; T3—before or at diagnosis of T1D; T4—poor or good glycemic control status. Time points from “at birth” to “10 years” refer to the age of patients from birth to 10 years old. Time points from “−1 year” to “+1 year” refer to within 1 year before and after seroconversion. B-Ab⁺: before seroconversion. A-Ab⁺: after seroconversion. B-T1D: before the diagnosis of T1D. T1D: diagnosed as T1D. Poor: poor glycemic control. Good: good glycemic control. TCA acids: metabolites involved in the tricarboxylic acid cycle (or citric acid cycle, Krebs cycle). The label “a” to “f”: refers to inconsistent changes of metabolites due to the same number of increasing and decreasing changes summarized in the publications, with a number pair from 1 to 6.

Figure 3

Pathway analysis of differential metabolites during the pathogenesis and progression of childhood T1D.

Figure 4

Consensus biomarkers of the development of islet autoimmunity and childhood T1D and the effect of glycemic control on the prognosis.