Files in this item



application/pdfDAVIS-DISSERTATION-2020.pdf (5MB)Restricted Access
(no description provided)PDF


Title:Microbial relationships between breastfeeding mothers and infants and response to maternal probiotic intervention
Author(s):Davis, Erin C.
Director of Research:Donovan, Sharon M
Doctoral Committee Chair(s):Swanson, Kelly S
Doctoral Committee Member(s):Holscher, Hannah D; Miller, Michael J
Department / Program:Nutritional Sciences
Discipline:Nutritional Sciences
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):human milk
Abstract:Infancy is a critical life stage for establishment of the gut microbiome. Perturbations to early microbial colonization have been associated with an increased risk for a number of chronic diseases in toddlerhood and childhood. Human milk (HM) contains a number of bioactive components that support microbial development. Microbes in HM provide a continuous source of commensal bacteria to the infant, and anti-inflammatory cytokines in HM promote tolerance to resident microbiota. However, HM varies among women; therefore, modulating composition may augment infant microbial development, and thus long-term health. The origin of the HM microbiota and its role in infant gut microbiota development remain unclear. In addition to the infant oral cavity and breast skin, the maternal gut is hypothesized to contribute to the HM microbiota. However, relationships between these sites and how the bacterial composition of HM may be influenced by feeding mode remain to be elucidated. Evidence that microbes are transported via an enteromammary pathway from the maternal gut to the mammary gland suggests that modulation of the maternal gut microbiota may in turn modulate that of HM. Perinatal probiotic supplementation has been shown to influence microbial and immune profiles of HM, but it is not known if probiotic-containing foods yield similar results. For these reasons, the overall objectives of this dissertation were to investigate microbial interrelationships between breastfeeding dyads and evaluate whether these communities can be modulated through maternal consumption of probiotic yogurt during lactation. Our first aim was to examine the relationships among fecal, oral, breast skin, and HM microbiota of exclusively breastfeeding (EBF) and mixed-feeding ([MF], human milk + infant formula) mother-infant pairs. Maternal and infant stool and saliva, breast skin, and HM were collected from EBF and MF dyads (N=33) at 6 weeks (wk) postpartum, and microbiota composition was analyzed in each site. The oral microbiota of MF infants was more diverse (p<0.05) and differed in overall microbial structure from that of EBF infants (p<0.01). At the genus-level, the oral microbiota of MF infants was comprised of a higher abundance of Veillonella but lower abundances of Streptococcus and Gemella than EBF counterparts (p<0.05). Bacterial communities of breast skin and HM were similar between feeding groups; however, Streptococcus abundance tended to be higher in HM of EBF mothers (p=0.10). Though feeding mode did not appear to have a large effect on breast skin and HM microbiota, Staphylococcus, Streptococcus, Rothia, Gemella, and Veillonella dominated in these communities and in infant saliva, likely reflecting their close contact during breastfeeding. Next, using SourceTracker2, we were able to predict microbial sources for just over 50% of the amplicon sequence variants (ASVs) in milk. Breast skin was estimated to contribute 45.9 ± 5.7 % (mean ± SEM) of the ASVs present in HM samples, supporting breast skin as a principal contributor to the HM microbiome. Infant and maternal saliva and maternal stool were also predicted to contribute but to a lower proportion. Notably, 46.7 ± 5.5% (mean ± SEM) of the HM microbiota was also estimated to arise from unknown sources not included in our analysis, which may be responsible for variation documented in the HM microbiota among women. Additionally, though a small percentage were estimated to come from sites sampled in our study, 98.8 ± 0.2% (mean ± SEM) of the ASVs in the infant fecal microbiota were predicted to arise from unknown sources. Thus, we hypothesize that a significant proportion of the infant gut microbiota may represent bacterial strains vertically transmitted during delivery and early lactation that persisted in the infant but not the mother. Lastly, after performing network analyses, we identified twenty-one significant co-occurrence relationships between bacteria in HM and on other maternal and infant body sites. The largest number of relationships were detected between bacterial genera in HM and infant stool. Actinomyces in HM was associated with infant fecal Escherichia-Shigella, Eggerthella, and Bacteroides, and Gemella and Corynebacterium in HM were also associated with Bacteroides in infant stool. These relationships suggest that the HM microbiota may have more of an influence on the proximal gastrointestinal tract, thereby indirectly influencing the colonic and fecal microbiota. Results from this study highlight several unique microbial associations between breastfeeding dyads and demonstrate that multiple maternal and infant microbial communities contribute to the HM microbiota, with breast skin being a primary source. Subsequently, we aimed to assess the impact of maternal probiotic yogurt consumption during lactation on composition and functional capacity of the maternal fecal microbiome (aim 2), microbial and immune composition of HM (aim 3), and the infant fecal microbiota (aim 4). Seventeen exclusively breastfeeding mothers and infants were enrolled at 6 wk postpartum, after which mothers were randomized to consume either probiotic yogurt or low-fat (1%) milk daily between 9 and 15 wk postpartum. Human milk and maternal and infant fecal samples, dietary, and demographic data were collected at 9, 12, 15, and 18 wk postpartum for microbiota and HM immune analyses. Results from 16S rRNA sequencing suggested no effect of either treatment on the maternal fecal microbiota. However, results from metagenomic analyses demonstrated that probiotic yogurt consumption increased maternal fecal abundance of Bifidobacterium animalis and Roseburia inulinivorans. The abundance of numerous microbial gene families and metabolic pathways were also enriched in the fecal microbiome of the yogurt group post intervention, including a pathway related to butyrate production. Taken together, these findings suggest probiotic yogurt is an effective strategy to modulate fecal microbiota composition. Furthermore, it may also lead to shifts in the functional capacity of the microbiome that hold the potential to benefit maternal health. Similar to 16S rRNA results of the maternal fecal microbiota, no changes in HM microbiota composition were observed over the course of the study in either treatment group. Though we attempted to also perform metagenomic analyses in HM, a large fraction of host cells and a low microbial biomass resulted in sequencing challenges and the inability to assess differences between treatment groups. However, we were still able to characterize findings from metagenomic analyses of 11 HM samples, demonstrating that the HM microbiota is even more diverse among women than previously appreciated. Functional analysis identified several microbial gene families and pathways related to nucleotide, amino acid, and cofactor metabolism offering information about potential interactions with the infant gut microbiota or other components in HM. Similar to 16S rRNA sequencing results, we observed no changes in concentrations of TGF-β1, TGF-β2, or IL-6 in either group between 9 and 15 wk postpartum. Finally, no changes were detected in the infant fecal microbiota over time in either the yogurt or milk groups. Though this study was well-controlled, the sample size was small, which significantly limited statistical power. Together, work presented in this dissertation highlights potential origins and function of the HM microbiome and unique ecological relationships between HM and microbial sites of breastfeeding dyads. Furthermore, it demonstrates probiotic yogurt consumption as a unique strategy to modulate the maternal gut microbiome, which could benefit maternal and possibly infant health. These data will help to guide future studies aimed at uncovering both mechanisms driving microbial community organization and potential targets for microbial modulation during the postpartum period.
Issue Date:2020-11-29
Rights Information:Copyright 2020 Erin C. Davis
Date Available in IDEALS:2021-03-05
Date Deposited:2020-12

This item appears in the following Collection(s)

Item Statistics