Untargeted metabolomic analysis of human milk from healthy mothers reveals drivers of metabolite variability

Abstract

Abstract Understanding the human milk metabolome can help inform infant nutrition and health. Untargeted metabolomics was used to study breast milk from 31 healthy participants to assess the shared metabolites in milk from participants with various backgrounds and understand how different demographic, health, and environmental factors impact the milk metabolome. Breast milk samples were analyzed by four separate UPLC-MS/MS methods. Metabolite Set Enrichment Analysis was used to study the most and least variable metabolites. The associations between participant factors and the metabolome were assessed with redundancy analyses. Among all 31 participants and between each untargeted UPLC-MS/MS method, 731 metabolites were detected, of which 389 were shared among all participants. Of the shared metabolites, lactose was the least and lactobionate the most variable metabolite. In the biological super pathway analysis, xenobiotics were the most variable metabolites. Infant age, maternal age, number of live births, and pre-pregnancy BMI were associated with the milk metabolome. In conclusion, the most variable metabolites originate from environmental exposures while the well-conserved core metabolites are linked to cell metabolism or are crucial for infant nutrition and osmoregulation. Understanding the variability of the breast milk metabolome can help identify components that are crucial for infant nutrition, growth, and development.