Human milk oligosaccharides (HMOs) have specific dose-dependent effects on child health outcomes. The HMO profile differs across mothers and is largely dependent on gene expression of specific transferase enzymes in the lactocytes. This study investigated the trajectories of absolute HMO concentrations at three time points during lactation, using a more accurate, robust, and extensively validated method for HMO quantification. We analyzed human milk sampled at 6 weeks (n = 682), 6 months (n = 448), and 12 months (n = 73) of lactation in a birth cohort study conducted in south Germany, using label-free targeted liquid chromatography mass spectrometry (LC-MS2). We assessed trajectories of HMO concentrations over time and used linear mixed models to explore the effect of secretor status and milk group on these trajectories. Generalized linear model-based analysis was used to examine associations between HMOs measured at 6 weeks of lactation and maternal characteristics. Results: Overall, 74%, 18%, 7%, and 1% of human milk samples were attributed to milk groups I, II, III, and IV, respectively. Most HMO concentrations declined over lactation, but some increased. Cross-sectionally, HMOs presented high variations within milk groups and secretor groups. The trajectories of HMO concentrations during lactation were largely attributed to the milk group and secretor status. None of the other maternal characteristics were associated with the HMO concentrations. The observed changes in the HMO concentrations at different time points during lactation and variations of HMOs between milk groups warrant further investigation of their potential impact on child health outcomes. These results will aid in the evaluation and determination of adequate nutrient intakes, as well as further (or future) investigation of the dose-dependent impact of these biological components on infant and child health outcomes.
Background/Objectives: Although approximately 160 human milk oligosaccharides (HMOs) have been identified, current studies on HMO quantitation are limited to the 10–19 most abundant HMOs. We assessed the variations in the relative concentrations of 71 HMO structures over lactation in human milk samples by an advanced liquid chromatography–mass spectrometry approach. Methods: Samples were collected from 64 mothers at 6 weeks, 6 months, and 12 months of lactation in the Ulm SPATZ Health Study, a German birth cohort. In this longitudinal study, we fitted linear mixed-effect models to analyze changes in the log2-transformed and standardized HMO concentration over time. Based on the profile of 71 HMOs, we also fitted a group-based multi-trajectory (GBMT) model to cluster mothers secreting cluster type I milk, who account for the majority of lactating mothers. Results: We found that 52 HMOs had a decreasing trend (regression coefficients ranging from −1.41 to −0.17) and 9 had an increasing trend (regression coefficients ranging from 0.25 to 0.64) during lactation, and the findings were statistically significant after multiple testing corrections. Using human milk samples of 49 mothers with type I milk, we further identified two novel sub-clusters with distinct longitudinal trajectories of concentrations of 71 HMOs during lactation: Type I-a (N = 20) and I-b (N = 29). These sub-clusters were not associated with maternal non-genetic characteristics. Conclusions: Our findings extend existing knowledge about the structural diversity of HMOs and their variations over lactation. These may pave the way to investigate the potential nutritional benefits of various HMOs on infant health and early life development in the future.