Postdoctoral Researcher University of Minnesota St Paul, Minnesota, United States
Background: Maternal metabolic status is linked to alterations in human milk composition that may affect infant health. The human milk metabolome (set of low molecular weight metabolites) can be modified by both maternal and infant factors, but how it may differ by maternal metabolic health is unclear. Assessing differences in the human milk metabolome may reveal how maternal metabolic status influences infant health postnatally among breastfeeding dyads and how breastfeeding may benefit infants born to mothers with obesity or gestational diabetes (GDM). Objective: To determine associations between maternal gestational diabetes mellitus (GDM) and perinatal weight status and the human milk metabolome in a prospective cohort of exclusively breastfeeding mothers at 1 month postpartum. Design/Methods: Data from participants in the Mothers and Infants Linked for Healthy Growth (MILk) study and the Maternal Milk, Metabolism, and the Microbiome study (4M Study) (n=350) were included in these analyses. Untargeted liquid chromatography-gas chromatography-mass spectrometry was used to quantify relative metabolite abundances in milk. Multiple linear regression analyses tested associations between GDM, pre-pregnancy BMI category (normal weight vs. overweight/obese), gestational weight gain (below/within guidelines vs above guidelines), or postpartum weight loss and milk metabolite abundances. Models were adjusted for maternal factors such as age, education, and parity and infant age and sex. Metabolite Set Enrichment Analysis identified alterations in pathways associated with maternal metabolic status. All p-values were adjusted for false discovery rate (FDR) using the Benjamini–Hochberg procedure (FDR < 0.05 considered significant). Results: In adjusted models, maternal GDM status was significantly associated with decreased abundances of stearoylcarnitine, glycine, and phenylacetic acid and increased abundances of 9-decenoylcarnitine, pregnanolone sulfate, and 2-hydroxybutyric acid. Maternal perinatal weight variables were not significantly associated with metabolite abundances after FDR correction. Several pathways with metabolites implicated in GDM were enriched by nominal p-value (pyruvate metabolism, aminoacyl-tRNA biosynthesis, etc.), but none remained significant after FDR correction.
Conclusion(s): In a large US-based contemporary cohort of exclusively breastfeeding participants, human milk metabolite abundances differed by maternal GDM status. Future work will aim to understand how these differences may be important to infant health outcomes, such as growth and neurodevelopment.
