Search for a command to run...
Abstract The human skin microbiome interacts intimately with human health, yet the drivers of skin microbiome composition and diversity are not well-understood. The composition of the skin microbiome has been characterized as both highly variable and relatively stable, depending on the time scale under consideration, and it is not clear what role contact with environmental sources of microbes plays in this variability. We experimentally mimicked human skin contact with two common environmental sources of microorganisms — soils and plant leaves — and characterized the dynamics of microbial acquisition and persistence on skin on very short time scales. Repeatable changes in skin community composition following encounters with environmental sources were observed, and these trajectories largely depend on donor community biomass distributions. Changes in composition persisted for at least 24 hours and through a soap and water wash following exposures to relatively high biomass soil communities. In contrast, exposures to lower biomass leaf communities were undetectable after a 24 hour period. Absolute abundances of bacterial taxa in source communities predicted transmission probabilities and residence times, independent of phylogenetic considerations. Our results suggest that variability in the composition of the skin microbiome can be driven by transient encounters with common environmental sources, and that these relatively transient effects can persist when the source is of sufficient biomass. Importance Humans come into contact with environmental sources of microbes, such as soil or plants, constantly. Those microbial exposures have been linked to health through training and modulation of the immune system. While much is known about the human skin microbiome, the short term dynamics after a contact event, such as touching soil, have not been well characterized. In this study, we examine what happens after such a contact event, describing trends in microbial transmission to and persistence on the skin. Additionally, we use computational sampling model simulations to interrogate null expectations for these kinds of experiments. This work has broad implications for infection control strategies and therapeutic techniques that rely on modification of the microbiome, such as probiotics and faecal transplantation.