Conversations between systemic immunity and the gut microbiome in early life

Therapies targeting the microbiome hold great promise to improve the wellbeing of children. Their success will depend on understanding host-microbiome conversations. The gut microbiome differs between geographically diverse populations. Does this contribute to distinct immune phenotypes? Does the same perturbance impact these diverse populations in the same way? How does the composition of the fungal microbiome compare to bacterial? What tools are needed to listen to these conversations where it matters the most, in vulnerable populations living in low-resource settings? These questions can be ethically asked in humans. To find answers to these questions, I assessed the relationship between systemic immune responses and the gut microbiome and compared bacterial to fungal colonization in geographically diverse child cohorts, hypothesizing that population-specific microbiomes correlated to distinct systemic innate immune phenotypes. To test this, we profiled the bacterial microbiomes of 2-year-old children from Belgium, Canada, Ecuador, and South Africa and measured their cytokine responses to innate stimuli. Certain immune differences between the cohorts correlated with the abundance of select bacterial taxa. Splenocytes of germ-free mice inoculated with human stools responded to stimulation in a manner consistent with the corresponding human donors, indicating that microbiomes can direct systemic innate immunity. I hypothesized that immune responses of HIV-exposed uninfected (HEU) and HIV-unexposed uninfected (HUU) children within each site differed in population-specific ways and that these differences correlated to distinct microbomes. While HEU were distinguished from HUU children by immune responses and microbiome composition in population-specific ways, differences in the microbiome did not correlate with altered immune phenotypes. I compared Bacterial vs. fungal composition over the first 5 years of life in a rural African population and found that both changed in kingdom-specific ways, suggesting they were shaped by separate selective forces. To overcome the hurdle of investigating newborns, I designed and implemented an experimental protocol in a low-resource setting that allows extraction of ‘big data’ out of the very small samples. Taken together, these findings suggest that therapies targeting the microbiome must consider population differences, and that placing newborns in low-resource settings at the forefront of our research is not only warranted, but feasible.