363 - Impact of fermented food diet on the milk microbiome and cytokine composition

Publication Number: 3350.363

Katie M.. Strobel, University of Washington School of Medicine, Seattle, WA, United States; Anna-Ursula Happel, University of Cape Town, Cape Town, Western Cape, South Africa; Obakeng Luthando. Jona, University of Cape Town, Cape Town, Western Cape, South Africa; Sonia Malczyk, University of Cape Town, Cape Town, Western Cape, South Africa; Brian Kullin, University of Cape Town, Cape Town, Western Cape, South Africa; Marijke A. Fagan-Endres, University of Cape Town, Cape Town, Western Cape, South Africa; Heather B.. Jaspan, Seattle Children's, Seattle, WA, United States

Background: Lactating mothers' nutrition is essential to human milk composition, and in South Africa, significant economic disparities contribute to maternal malnutrition and infant stunting. Fermented foods like Mageu, an African maize porridge, have evidence of benefits for gut microbial health, including reduced inflammation and increased nutrient bioavailability. Our previous work found that consumption of Mageu altered the maternal gut microbiota and plasma ferritin levels.

Objective: We aimed to assess if maternal daily consumption of varying Mageu preparations versus no Mageu during lactation from 4 to 10 weeks post-partum impacted the mother's own milk (MOM) microbiota and cytokines.

Design/Methods: This was a pilot randomized controlled trial of 45 women and their infants in Khayelitsha, South Africa. Following enrollment at birth, women had a four-week washout with < 500 ml of fermented foods/week. During the intervention, 4 to 10 weeks post-partum, women consumed their randomized intervention (500 ml daily, live-culture Mageu [LCM], store-bought pasteurized Mageu [SBM], or no Mageu). MOM was collected at weeks 4 and 10. 16S rRNA gene sequencing of the V3-V4 region was performed and processed using DADA2 to determine taxonomy. Relative abundances were center-log-ratio (CLR) transformed. Linear mixed models were used to evaluate changes in abundance over time within the Mageu group. Luminex panel of 13 cytokines was performed to evaluate milk cytokine concentrations at week 10. Concentrations were natural log-transformed and compared using the Wilcoxon and Kruskal-Wallis tests.

Results: 39 participants had milk microbiota data, and 30 participants had cytokine data. Maternal clinical and dietary characteristics are described in Table 1. Alpha-diversity and Bray-Curtis beta-diversity measures did not significantly differ between Mageu groups (Figure 1A&B). Mothers who consumed LCM and SBM showed changes in CLR-transformed abundance of multiple taxa following the intervention (Table 2). The top 15 milk taxa included both gram-positive and gram-negative bacteria associated with infant growth (Figure 1C). There were no significant differences in cytokine concentrations among Mageu groups (p > 0.05 for all comparisons). However, mothers receiving any Mageu (LCM or SBM) had higher median interferon-gamma compared to no Mageu users at week 10 (2.1 (1.5) vs. 0.18 (2.3) ln(pg/mL), p=0.02).

Conclusion(s): Mageu consumption was associated with modest shifts in milk bacterial taxa and interferon-gamma. Future studies will evaluate the implications of these shifts in milk composition to the infant gut microbiota and growth.

Table 1
Maternal characteristics. Continuous variables are displayed as median (IQR) and nominal variables as % (N). Three 24-hour dietary recalls were performed at weeks 4, 7, and 10. The South Africa Food Database was used to calculate the mean energy and nutrient intake adjusted for intra-person variability. Continuous variables were compared using Wilcoxon rank sum tests, and nominal variables were compared with Fisher's exact tests.

Figure 1
A. Shannon alpha-diversity, B. Bray-Curtis distances principal coordinate analysis, and C. relative abundance of species across study weeks comparing Mageu groups. Green represents no Mageu, salmon represents store bought mageu (SBM), and lilac represents live culture Mageu (LCM).

Table 2
Results of linear mixed models evaluating CLR-transformed abundances of various taxa with random effect participant and the interaction term study week and study group. Analyses were not adjusted for multiple comparisons, given the exploratory nature of the study.