Metabolomic signature reveals dysregulated lipoprotein profile in m.3243A>G carriers: a case-control study.

Abstract

Introduction: The pathogenic mitochondrial gene variant m.3243A>G disrupts oxidative phosphorylation and is associated with insulin resistance, both of which may be linked to unfavorable lipid metabolism. However, the metabolic alterations in m.3243A>G carriers, including what differentiates those with and without diabetes, remain incompletely understood.

Objectives: To investigate metabolomic profiles in fasting serum and urine samples from m.3243A>G carriers compared to healthy controls.

Methods: Metabolomic profiling of serum and urine samples using nuclear magnetic resonance-based metabolomics in m.3243A>G carriers (n = 28) was compared to healthy controls matched for age and sex. Additionally, profiles from m.3243A>G carriers with diabetes (n = 16) were compared with carriers without diabetes (n = 12) to identify potential metabolites associated with the presence of diabetes.

Results: Twenty-five metabolites in serum and 16 in urine were identified as metabolites separating m.3243A>G carriers from healthy controls. The m.3243A>G carriers presented with increased triglycerides across lipoprotein particles and altered very-low-density lipoprotein concentrations and composition. In addition, there were alterations in metabolites from a number of metabolic pathways, including glycolysis, the tricarboxylic acid cycle, glutathione, one-carbon, and nucleotide metabolism. A three metabolite-urine signature (uracil, hypoxanthine, and 1-methylnicotinamide) demonstrated discriminating potential between m.3243A>G carriers and controls in exploratory machine learning analyses (area under the curve values 0.94-0.99 and cross-validation prediction of 0.81-0.93). Among m.3243A>G carriers, branched-chain amino acids were higher in individuals with diabetes compared with carriers without diabetes.

Conclusion: Dysregulated lipoprotein metabolism represents a significant metabolic fingerprint of m.3243A>G carriers. Furthermore, higher levels of branched-chain amino acids may be associated with the presence of diabetes.