Abstract:
Flagellin is a key component of the bacterial flagellum, a molecular structure essential for bacterial motility. In hosts, flagellin is recognized by receptors of the innate immune system: Toll-like receptor 5 (TLR5), and NAIP5/NLRC4 in animals, and Flagellin-Sensing receptor 2 (FLS2), Flagellin-Sensing receptor 3 (FLS3) in plants. To date, the study of flagellins has primarily focused on pathogens, largely neglecting flagellin diversity in the context of host-microbe interactions with commensal bacteria. My research group recently revealed that flagellin FlaB from Roseburia hominis, a gut commensal, is a weak TLR5 agonist, despite strong TLR5 binding, a phenotype we named 'silent recognition' by TLR5. This discovery hints at an unexplored diversity of flagellins and their mode of interaction with the host. Moreover, the distribution and evolution of flagellin diversity throughout natural microbial environments remains unclear. Thus, this work aims to identify patterns of diversity and evolution in flagellins associated with free-living and host-associated habitats.
I used publicly available flagellin sequences to profile flagellin communities in 785 metagenomic samples encompassing free-living and host-associated environments, performed enrichment analyses, and implemented codon models of evolution to detect pervasive and episodic positive selection. I found that free-living environments comprise a higher diversity of flagellin genes than host-associated environments, and that flagellin composition differentiates biomes, particularly in free-living and animal-associated samples. Moreover, I discovered different flagellin categories that were enriched in each biome, as well as evidence of pervasive negative selection and episodic positive selection occurring on the conserved domains of these enriched flagellins. These signatures of selection were consistently present around the regions interacting with host receptors. However, their location was not associated with the habitat.
Building upon the characterization of free-living and host-associated environments, I then aimed to characterize flagellins in human gut metagenomes to probe the relationship between their evolutionary history and their mode of interaction with TLR5. Using publicly available flagellin sequences, I profiled flagellin communities for 270 human gut metagenomes and performed homology searches to differentiate between predicted stimulatory and silent flagellins. My results indicated that the majority of flagellin in the human gut is encoded by members of the Lachnospiraceae family (Firmicutes), that 45% of the putative silent flagellins exhibited the predicted phenotype, and that the silent mode of interaction is widespread across unrelated flagellated bacteria.
This work expands our understanding of flagellin ecology in free-living and host-associated environments. Moreover, it provides insights into the role of natural selection in the diversification of flagellin across bacteria. Finally, it establishes a structured framework for formulating important hypotheses that will drive future explorations of flagellin-mediated host-microbe interactions.