Transmissible plasmids spread genes encoding antibiotic resistance and other traits to new bacterial species. Here we report that laboratory populations of Escherichia coli with a newly acquired IncQ plasmid often evolve 'satellite plasmids' with deletions of accessory genes and genes required for plasmid replication. Satellite plasmids are molecular parasites: their presence reduces the copy number of the full-length plasmid on which they rely for their continued replication. Cells with satellite plasmids gain an immediate fitness advantage from reducing burdensome expression of accessory genes. Yet, they maintain copies of these genes and the complete plasmid, which potentially enables them to benefit from and transmit the traits they encode in the future. Evolution of satellite plasmids is transient. Cells that entirely lose accessory gene function or plasmid mobility dominate in the long run. Satellite plasmids also evolve in Snodgrassella alvi colonizing the honey bee gut, suggesting that this mechanism may broadly contribute to the importance of IncQ plasmids as agents of bacterial gene transfer in nature.

The evolution of antibiotic-resistant bacteria threatens to become the leading cause of worldwide mortality. This crisis has renewed interest in the practice of phage therapy. Yet, bacteria's capacity to evolve resistance may debilitate this therapy as well. To combat the evolution of phage resistance and improve treatment outcomes, many suggest leveraging phages' ability to counter resistance by evolving phages on target hosts before using them in therapy (phage training). We found that in vitro, λtrn, a phage trained for 28 d, suppressed bacteria ∼1,000-fold for three to eight times longer than its untrained ancestor. Prolonged suppression was due to a delay in the evolution of resistance caused by several factors. Mutations that confer resistance to λtrn are ∼100× less common, and while the target bacterium can evolve complete resistance to the untrained phage in a single step, multiple mutations are required to evolve complete resistance to λtrn. Mutations that confer resistance to λtrn are more costly than mutations for untrained phage resistance. Furthermore, when resistance does evolve, λtrn is better able to suppress these forms of resistance. One way that λtrn improved was through recombination with a gene in a defunct prophage in the host genome, which doubled phage fitness. This transfer of information from the host genome is an unexpected but highly efficient mode of training phage. Lastly, we found that many other independently trained λ phages were able to suppress bacterial populations, supporting the important role training could play during phage therapeutic development.