Abstract
Background: The World Health Organization (WHO), United Nations Children’s Fund (UNICEF), and American Academy of Pediatrics (AAP) recommend exclusive breastfeeding for the first 6 months of life, followed by continued breastfeeding alongside complementary foods for up to 2 years or longer, as desired. These guidelines are grounded in extensive research demonstrating the numerous short- and long-term benefits of breastfeeding for both lactating parents and their infants. However, despite the growing body of literature on human milk (HM), significant gaps remain in understanding how a mother’s diet influences the composition of HM and its potential impact on infant health. For example, food-borne allergens in HM may cause allergic responses in HM-fed infants, but the appearance and abundance of these allergens is highly variable both within and between lactating individuals. This variability complicates recommendations and can add psychological distress for individuals nursing infants with food sensitivity. Furthermore, for mothers wishing to feed their infants HM, management of infant food reactivity is currently limited to maternal dietary elimination. Bovine milk (BM) and soy are commonly identified triggers for food-sensitive infants and thus are frequently eliminated from the diet by lactating parents. More research is needed investigating the impacts of bovine milk and soy elimination and re-introduction on maternal stress, infant behavior and gastrointestinal (GI) symptoms. Finally, the bacterial composition of HM varies among women, but factors associated with this variation are poorly understood. Few studies have evaluated the relationship between the HM microbiome (HMM) and maternal diet. Objective: The research included in this dissertation aimed to (1) identify bovine- and soy-derived peptides in HM after maternal elimination and reintroduction of BM and soy beverage (SB), (2) evaluate the impact of maternal BM and SB elimination and consumption on the HMM, and (3) evaluate the impact of maternal BM and SB elimination and re-introduction on maternal perceived stress, maternal/infant gastrointestinal (GI) symptoms, and infant behavior. Methods: In this randomized, cross-over, dietary intervention trial, 38 lactating participants underwent 2 study phases, each including a 5-day diet elimination, 3-day diet intervention, and 2-day washout. Participants eliminated BM-containing products during one phase and soy-containing products during the other. Each diet intervention required daily consumption of increasing amounts of BM or SB (175, 295, and 415 mL). Peptidomic analysis was performed on a subset of 75 HM samples (produced by 24 participants) collected after dietary elimination, and 2 and 4 h after BM/SB consumption (415 mL). Peptides were isolated via ethanol precipitation and C18 solid-phase extraction, analyzed by LC-MS/MS, and identified with Proteome Discoverer. DNA was extracted from HM collected at baseline, after the 5-d diet elimination, and 4 h after BM/SB consumption on d3 of each intervention period. The HMM was characterized via full-length sequencing of the bacterial 16S rRNA gene. Maternal stress was assessed using the Perceived Stress Scale (PSS) at baseline, d 5 of the elimination periods, and d 3 of each dietary intervention. Maternal and infant GI symptoms and infant behavior were measured using daily surveys collected during both intervention and washout periods. Results: We identified 121 bovine-derived peptides (associated with 6 proteins) in HM collected during the BM phase. From most to least abundant, these proteins were β-lactoglobulin, κ-casein, αs1-casein, β-casein, α-lactalbumin protein variant D, and glycosylation-dependent cell adhesion molecule 1. Differences in relative abundances were identified for 14 peptides when comparing before, and 2 and 4 h after BM consumption. We identified 8 peptides of possible soy origin in HM collected during the SB phase, but they were not matched to parent proteins with adequate confidence. When examining the bacterial composition of HM at the genus level, Shannon diversity was greater during BM elimination than during consumption of BM or soy elimination. Simpson diversity was greater during BM consumption and soy elimination than during BM elimination. Bacterial richness was greater in HM produced during BM elimination and soy elimination than during BM consumption. Bacterial evenness was greater during BM elimination and at baseline compared to soy elimination. Similarly, species-level Shannon diversity was greater during BM elimination than BM consumption, and richness was greater during BM elimination and soy consumption than BM consumption. Relative amounts of individual taxa did not differ across time points, but Arthrobacter agilis (0.52 ± 2.16%; n=2) was only present in HM produced during SB consumption, and Brevundimonas vesicularis (3.07 ± 10.68%; n=4), Actinomyces johnsonii (0.01 ± 0.04%; n=2), and Kocuria marina (0.15 ± 0.73%; n=2) were only present during BM elimination. No differences were identified in maternal stress when compared across time points. There was a trend toward more infant bowel movements (p = 0.07) and a greater probability of maternal GI symptoms (p = 0.10) during washout d 2 after maternal consumption of BM compared to SB consumption. After controlling for multiple comparisons, a trend indicated greater time sleeping (p = 0.06) and breastfeeding (p = 0.08) during washout d 2 after SB consumption compared to BM consumption. Conclusions: The research examined in this dissertation provides additional evidence that the maternal diet can influence certain components of HM and may also affect maternal and infant GI symptoms and infant behavior. The identification of BM-derived peptides, although in low relative abundances, in all samples collected after dietary elimination (and BM consumption) may indicate that 5 days of elimination is not adequate for the full elimination of BM-derived peptides in HM or may suggest that low levels of these peptides are persistently present, perhaps from other environmental sources. Soy-derived peptides may not be present in HM following consumption of SB – at least in this population. More research is needed to evaluate factors impacting peptide appearance and determine if some non-human peptides are consistently present in HM despite dietary elimination. Our data indicate that bacterial richness, evenness, and alpha diversity of the HMM are greater after short-term BM elimination and may be affected differently based on the specific food group being eliminated and re-introduced. The variation in HMM alpha diversity appears to be due to specific low-prevalence taxa that were only present during short-term BM elimination and SB consumption. Daily consumption of SB or BM may have an impact on maternal GI symptoms, infant bowel movements, and infant behavior, specifically regarding time spent breastfeeding and sleeping in a 24 h period, but the appearance of these effects may be delayed. Future studies examining these relationships should also explore the impacts of longer periods of dietary elimination as well as the elimination and consumption of other foods or food groups commonly restricted by lactating women.