UCSF

In this dissertation, we demonstrate how clinical cohort studies of allergic disease can be paired with current sequencing and experimental technologies to identify microbial targets for disease prevention. The first year of life is a critical window during which the microbiome is shaped by, and shapes, the host immune and epithelial systems. This work is organized around the two organ systems most directly implicated in this process: the upper airway, where microbial interactions have a more direct influence on airway-specific disease, and the gut, where the microbiome shapes the baseline immune tolerance established in the first year of life and with it the overall capacity to develop allergic disease.In the upper airway, we show that genetic risk alleles long associated with asthma, GSDMB and ORMDL3, interact with the airway microbiome at 1 year of age to influence early-life wheeze risk. Infants carrying these risk alleles who are dominated by Moraxella, Streptococcus, or Haemophilus at 1 year have increased wheeze risk (Pint = 0.016 for all), while dominance by Corynebacterium, Dolosigranulum, Staphylococcus, or Bacillus reduces risk. In airway epithelial cells CRISPR-edited to be homozygous for the GSDMB risk variant rs7216389TT, we observed decreased expression of genes involved in antimicrobial response and neutrophil recruitment, along with increased microbial adherence, linking host genotype, epithelial function, and microbial colonization directly. In the gut, we characterized stool from a longitudinal, high-risk infant cohort across the first year of life. In a 6-month subset, we identified Bacteroides cellulosilyticus and Hungatella effluvii as asthma-protective factors, with putative roles in carbohydrate and nitrogenous compound metabolism respectively. Extending this analysis across the full first year, we assembled a 563-genome catalog from shotgun metagenomic sequencing and identified window-specific bacterial genomes whose presence or absence at defined developmental ages tracks atopic, eczema, and asthma outcomes. The strongest genome-level association with asthma was Akkermansia massiliensis, enriched at birth (q = 0.043) and elevated in prevalence through 6 months of age before normalizing to levels seen in infants who do not develop asthma, underscoring the importance of timing in microbial-associated allergic disease risk. This genome catalog provides the foundation for ongoing work identifying bacterial enzymatic targets, particularly within the linoleic acid pathway that produces 12,13-diHOME, and for integrating taxonomic, functional, and metabolomic data to define the axes of variation that best explain allergic disease risk. Together, these studies show that specific microbes, at specific times, in specific host genetic contexts, shape the trajectory toward allergic disease. The ultimate goal of this work is to design interventions that limit allergic disease development in future generations.