Background

The objective of the study was to identify a breastmilk (BM) metabolite signature that discriminates normal weight (NW) and obese (OB) women during the first 6 months post-partum using untargeted metabolomics.

Methods

Maternal anthropometrics and body composition were assessed before week 10 of gestation. BM samples were collected at 2 weeks (2W), 2 months (2M), and 6 months (6M) postpartum from NW (BMI=18.5 – 24.9 kg/m2, N=77, 66 and 53, respectively) and OB (BMI>30 kg/m2, N = 82, 64 and 41, respectively) mothers. Gas chromatography time of flight-mass spectrometry was used to measure metabolites in BM. Partial least squared-discriminant analysis (PLS-DA), and PLS-regression (PLS-R) models were assessed to examine relationships between metabolites and maternal BMI (categorical and continuous) and fat mass (%).

Results

PLS-DA modeling predicted an average of 76% of held out samples across all sample periods, while PLS-R resulted in an average root mean error of prediction of 6.2 mg/kg2 and 12.7 kg for BMI and maternal fat mass, respectively.There were 21, 19, and 13 discriminant BM metabolites at 2W, 2M, and 6M post-partum, respectively. Monosaccharides and sugar alcohols were the most representative known BM metabolite classes identified, and their concentrations were all were elevated in OB women compared to NW (2W: mannose, ribose, lyxose, lyxitol; 2M: mannose, ribitol, glycerol, isothreonic acid, lyxitol; 6M: lyxitol and isothreonic acid).Other metabolites greater in BM from OW women at 2W: 1-monostearin, xylonolactone, shikimic acid, pseudo uridine, and dodecanol; 2M: N-acetyl-D-hexosamine, fumaric acid; 6M: succinic acid, uric acid, and tyrosine.

Conclusions

Carbohydrates, sugar alcohols and other metabolites discriminated BM of OB compared to BM of NW women. Future studies are needed to understand how these metabolites impact infant’s growth and development and evaluate what maternal factors (e.g. diet, adiposity) drives these differences.Funding: USDA ARS # 6026-51000-010-05S.