Stochastic colonization and host-to-host transmission shape gut bacterial variability

Abstract Understanding the kinetic processes that govern bacterial population dynamics within hosts is critical in developing effective strategies to control microbiota. However, inferring population dynamics is challenging due to large host-to-host bacterial population variability stemming from stochastic colonization events, as well as the inability to continuously monitor the bacterial population without disturbing the host. Using C. elegans fed E. coli under different diets, we show that early colonization acts as a stochastic bottleneck that drives substantial divergence in host-level bacterial loads, and that the spreading of bacteria from colonized worms to sterile ones regulates this variability by altering effective colonization pressure. These conclusions are reinforced using a simulation-based inference framework that quantifies stochastic within-host population dynamics from discrete snapshot data, enabling inference of effective colonization and growth rates across heterogeneous hosts with variable carrying capacities. Applying this framework, we further demonstrate that the bacterial predator B. bacteriovorus reduces average gut bacterial loads by two orders of magnitude, primarily by suppressing environmental recolonization and subsequent host-to-host transmission rather than eliminating established intra-host populations. Together, these results reveal that host-associated microbial population dynamics are strongly impacted by environmental colonization processes that modulate stochastic entry events. Competing Interest Statement The authors have declared no competing interest. Footnotes The abstract had the italicized words added properly. The Methods section has been revised to improve readability. Data availability Plate images are available on https://zenodo.org/records/20128602 (DOI: 10.5281/zenodo.20128602) and figure data is available https://github.com/PessoaP/Celegans_inference.