Abstract It is not clear if the bacterial strains which comprise our microbiota are mostly long-term colonizers or transient residents. Studies have demonstrated decades long persistence of bacterial strains within the gut, but persistence at other body sites has yet to be determined. The vaginal microbiota (VMB) is often dominated by Lactobacillus , although it is also commonly comprised of a more diverse set of other facultative and obligate anaerobes. Longitudinal studies have demonstrated that these communities can be stable over several menstrual cycles or can fluctuate temporally in species composition. We sought to determine whether the bacterial strains which comprise the VMB were capable of persisting over longer time-periods. We performed shotgun metagenomics on paired samples from 10 participants collected 1 and 2 years apart. The resulting sequences were de novo assembled and binned into high-quality metagenome assembled genomes. Persistent strains were identified based on the sequence similarity between the genomes present at the two timepoints and were found in the VMB of six of the participants, three of which had multiple. The VMB of the remaining four participants was similar in species composition at the two timepoints but was comprised of different strains. For the persistent strains, we were able to identify the mutations which fixed in the populations over the observed time period, giving insight into the evolution of these bacteria. These results indicate that bacterial strains can persist in the vagina for extended periods of time, providing an opportunity for them to evolve in the host microenvironment. Importance The persistence of strains within the vaginal microbiota has not yet been characterized. Should these strains be capable of persisting for extended periods of time, they could evolve within their host in response to selective pressures exerted by the host or by other members of the community. Here, we present preliminary findings which demonstrate that bacterial strains can persist in the vagina for at least one year and evolve over time. In several cases, multiple strains persisted together in a community, indicating that co-evolution between bacterial strains could occur in the vagina. Our observations motivate future studies which collect samples from more participants, at more timepoints and over even longer periods of time. Understanding which strains persist, what factors drive their persistence, and what selective pressures they face will inform the development and delivery of rationally designed live biotherapeutics for the vagina.

Persistence andin vivoevolution of vaginal bacterial strains over a multi-year time period