The microbiome and metabolome of pre-term infant stool is personalized, and not driven by health outcomes including necrotizing enterocolitis and late-onset sepsis

Abstract

12Background: The assembly of the intestinal microbiota of extremely low birthweight (ELBW) infants has 13 an important impact on both immediate and long term health. ELBW infants are frequently given 14 antibiotics which are likely to perturb the assembly of the microbiota. Health complications are not 15 uncommon for ELBW infants; they face health crises including sepsis and necrotizing enterocolitis (NEC). 16Microbes are thought to be involved in the pathogenesis of NEC, but the mechanisms are unclear. New 17 understanding of the importance of human milk oligosaccharides and the establishment of a 18Bifidobacteria-dominated gut microbiota early in infancy suggest that all preterm infants have abnormal 19 microbial colonization. The initial assembly of intestinal microbial communities may have significant 20 impact on immune development and lifelong health. 21Results: We measured the bacterial composition and metabolite profile of 32 ELBW infants by 16S rRNA 22 gene sequencing and untargeted gas chromatography mass spectrometry of fecal samples. Infants 23 either remained healthy, developed late-onset sepsis, or developed necrotizing enterocolitis. The 24 . CC-BY-NC 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/125922 doi: bioRxiv preprint first posted online Apr. 10, 2017; 2 bacterial compositions were similar to what has been observed in other studies of preterm infants. Fecal 25 samples are dominated by aero-tolerant bacterial species, specifically Enterococcus, Enterobacteriaceae, 26 and Staphylococcus. Only three ELBW infants were colonized by Bifidobacteria. Fecal samples from 27 infants who developed NEC were not distinguishable from other infant samples based on bacterial 28 compositions (Permanova R 2 < 0.001, p = 0.99) or metabolite profiles (Permanova R 2 = 0.05, p= 0.24). 29Instead the bacterial composition (R 2 = 0.63, p < 0.001) and metabolite profile (R 2 = 0.43, p < 0.001) were 30 highly personalized for each infant. There were not significant correlations between the bacterial 31 composition and metabolite profiles of fecal samples (Mantel test r= 0.18, p < 0.001). 32Conclusions: Although antibiotics likely contribute to the instability of the ELBW infant intestinal 33 microbiota, personalized signatures of bacteria and metabolites are still clearly present. Neither the 34 bacterial composition or metabolite profile was unique in cases of disease. While bacteria certainly 35 contribute to the profile of metabolites present in feces, in these ELBW infants, significant correlations 36 between bacterial relative abundances as determined by 16S rRNA gene sequencing and untargeted GC-37 MS metabolite profiles were not detectable. 38 39 Background 40The intestinal microbiota of infants initially assembles by exposure to the mother's microbiota as well as 41 exposure to microbes in the environment [1]. In the first few days, the intestines are colonized by 42...

Associations between the gut microbiome and metabolome in early life

“…However, these studies have mostly focused on adult populations with specific metabolic disease etiologies such as inflammatory bowel disease. Only a limited number of studies [ 18 – 23 ] have simultaneously profiled the gut microbiome and metabolome from infant stool samples. These studies have preliminarily established that metabolomic profiles shift as taxonomic abundances change between subject case/control status [ 18 , 20 , 21 , 24 ].…”

“…Only a limited number of studies [ 18 – 23 ] have simultaneously profiled the gut microbiome and metabolome from infant stool samples. These studies have preliminarily established that metabolomic profiles shift as taxonomic abundances change between subject case/control status [ 18 , 20 , 21 , 24 ]. Specifically, Ayeni et al ( n = 48) [ 19 ] and Kisuse et al ( n = 35) [ 23 ] demonstrated that inter-sample distances calculated using metabolite abundances are correlated with those calculated from taxonomic profiles using Mantel tests across African and Asian cohorts.…”

Multi-omic Characterization of the Taxa-function Relationship in Infant Gut Microbiomes

“…However, these studies have mostly focused on adult populations with specific metabolic disease etiologies such as inflammatory bowel disease. Only a limited number of studies [18][19][20][21][22][23] have simultaneously profiled the gut microbiome and metabolome from infant stool samples. These studies have preliminarily established that metabolomic profiles shift as taxonomic abundances change between subject case/control status [18,20,21,24].…”

“…Only a limited number of studies [18][19][20][21][22][23] have simultaneously profiled the gut microbiome and metabolome from infant stool samples. These studies have preliminarily established that metabolomic profiles shift as taxonomic abundances change between subject case/control status [18,20,21,24]. Specifically, Ayeni et al (n = 48) [19] and Kisuse et al (n = 35) [23] demonstrated that inter-sample distances calculated using metabolite abundances are correlated with those calculated from taxonomic profiles using Mantel tests across African and Asian cohorts.…”

“…Specifically, Ayeni et al (n = 48) [19] and Kisuse et al (n = 35) [23] demonstrated that inter-sample distances calculated using metabolite abundances are correlated with those calculated from taxonomic profiles using Mantel tests across African and Asian cohorts. However, studies to date have either focused on preterm infants [18,20,21] or had small sample sizes (less than 50) [19,22,23]. We identified a major gap in defining microbiome-metabolome relatedness among infants from a population-based cohort capturing both early in infancy and near the first birthday, with regards to determining the strength of association and to identify key contributors to the overall concordance.…”