Impact of early-life perturbations on pediatric microbiome maturation

ABSTRACT During the first 3-years of life (YOL) the infant gut microbiome (GM) rapidly diversifies both in structure and function, concomitant with dietary and environmental transitions. Critically, the GM response to specific external stimuli is patient-specific, complicating individualized risk predictions. Healthy GM maturation includes accruing

multiple strains of the same species, which frequently differ in key functions. These functional differences, ac- centuated by horizontal gene transfer (HGT) and de novo mutations, could resolve conflicting associations of the same species with both health and disease. The rationale behind our proposal is that strain- and species-

level variation in bacterial functions drives heterogenous GM responses to early-life (EL) dietary and antibiotic perturbations, which explains, in part, individualized developmental trajectories. This proposal pursues two highly complementary Aims: 1) Define strain-resolved functional maturation of the pediatric gut microbiome and

2) Investigate the acute effects of EL antibiotic (ELA) perturbation on strain dynamics, HGT, and micro- biome maturation in preterm neonates and microbiota-humanized mice. Aim 1 will test the hypothesis that EL environmental exposures shape genomic diversification of gut species, causing lasting changes in GM com-

munity structure and microbial functions. We will leverage our unique set of 2,436 stools collected over the first 9 YOL from infants variably exposed to dietary and environmental stimuli. By combining culture-enriched meta- genomics, metatranscriptomics, and metabolomics, we will determine taxa-function relationships at the sub-spe-

cies level and power statistical models that predict the impact of EL exposures on strain diversification, microbe- function associations, and transcriptional activity. Aim 2 will test the hypothesis that ELAs acutely alter strain dynamics and stimulate HGT and that the GM response to ELA can be predicted from baseline composition and

function. Here, we will interrogate 160 stools flanking variable ELA exposure in 80 preterm neonates in the first 4 months of life, combining culture-enriched metagenomics with selective culture and isolate sequencing to char- acterize the preterm `plasmidome' and profile post-ELA strain dynamics and HGT. To identify microbiome-intrin-

sic responses to ELA, we will utilize an innovative transgenerational mouse model where germ-free dams receive human, preterm, microbiota that is vertically transferred to their pups, which are treated with parenteral antibiot- ics. We will use the resulting data to predict individual GM responses to specific antibiotics based on composition,

resistance gene content, and bacterial functions. Our proposal is innovative because our interdisciplinary re- search team will characterize strain-level bacterial functions to understand the heterogeneity of GM responses to EL perturbations on two pre-existing sets of human specimens; it is significant because it will identify features

that predict species-resolved GM-specific responses to EL selection. Our work will advance pediatric microbi- ome research by comprehensively characterizing strain-resolved functional maturation and GM disruption to understand individual variation leading towards a future of personalized, microbiome medicine.