Plasmids play an important role in bacterial evolution by transferring niche-adaptive functional genes between lineages, thus driving genomic diversification. Bacterial genomes commonly contain multiple, coexisting plasmid replicons, which could fuel adaptation by increasing the range of gene functions available to selection and allowing their recombination. However, plasmid coexistence is difficult to explain because the acquisition of plasmids typically incurs high fitness costs for the host cell. Here, we show that plasmid coexistence was stably maintained without positive selection for plasmid-borne gene functions and was associated with compensatory evolution to reduce fitness costs. In contrast, with positive selection, plasmid coexistence was unstable despite compensatory evolution. Positive selection discriminated between differential fitness benefits of functionally redundant plasmid replicons, retaining only the more beneficial plasmid. These data suggest that while the efficiency of negative selection against plasmid fitness costs declines over time due to compensatory evolution, positive selection to maximize plasmid-derived fitness benefits remains efficient. Our findings help to explain the forces structuring bacterial genomes: coexistence of multiple plasmids in a genome is likely to require either rare positive selection in nature or nonredundancy of accessory gene functions among the coexisting plasmids.IMPORTANCE Bacterial genomes often contain multiple coexisting plasmids that provide important functions like antibiotic resistance. Using lab experiments, we show that such plasmid coexistence within a genome is stable only in environments where the function they encode is useless but is unstable if the function is useful and beneficial for bacterial fitness. Where competing plasmids perform the same useful function, only the most beneficial plasmid is kept by the cell, a process that is similar to competitive exclusion in ecological communities. This process helps explain how bacterial genomes are structured: bacterial genomes expand in size by acquiring multiple plasmids when selection is relaxed but subsequently contract during periods of strong selection for the useful plasmid-encoded function.

The acquisition of plasmids is often accompanied by fitness costs such that compensatory evolution is required to allow plasmid survival, but it is unclear whether compensatory evolution can be extensive or rapid enough to maintain plasmids when they are very costly. The mercury-resistance plasmid pQBR55 drastically reduced the growth of its host, Pseudomonas fluorescens SBW25, immediately after acquisition, causing a small colony phenotype. However, within 48h of growth on agar plates we observed restoration of the ancestral large colony morphology, suggesting that compensatory mutations had occurred. Relative fitness of these evolved strains, in lab media and in soil microcosms, varied between replicates, indicating different mutational mechanisms. Using genome sequencing we identified that restoration was associated with chromosomal mutations in either a hypothetical DNA-binding protein PFLU4242, RNA polymerase or the GacA/S two-component system. Targeted deletions in PFLU4242, gacA or gacS recapitulated the ameliorated phenotype upon plasmid acquisition, indicating three distinct mutational pathways to compensation. Our data shows that plasmid compensatory evolution is fast enough to allow survival of a plasmid despite it imposing very high fitness costs upon its host, and indeed may regularly occur during the process of isolating and selecting individual plasmid-containing clones.

Conjugative plasmids play a vital role in bacterial adaptation through horizontal gene transfer. Explaining how plasmids persist in host populations however is difficult, given the high costs often associated with plasmid carriage. Compensatory evolution to ameliorate this cost can rescue plasmids from extinction. In a recently published study we showed that compensatory evolution repeatedly targeted the same bacterial regulatory system, GacA/GacS, in populations of plasmid-carrying bacteria evolving across a range of selective environments. Mutations in these genes arose rapidly and completely eliminated the cost of plasmid carriage. Here we extend our analysis using an individual based model to explore the dynamics of compensatory evolution in this system. We show that mutations which ameliorate the cost of plasmid carriage can prevent both the loss of plasmids from the population and the fixation of accessory traits on the bacterial chromosome. We discuss how dependent the outcome of compensatory evolution is on the strength and availability of such mutations and the rate at which beneficial accessory traits integrate on the host chromosome.

Plasmids drive genomic diversity in bacteria via horizontal gene transfer [1, 2]; nevertheless, explaining their survival in bacterial populations is challenging [3]. Theory predicts that irrespective of their net fitness effects, plasmids should be lost: when parasitic (costs outweigh benefits), plasmids should decline due to purifying selection [4-6], yet under mutualism (benefits outweigh costs), selection favors the capture of beneficial accessory genes by the chromosome and loss of the costly plasmid backbone [4]. While compensatory evolution can enhance plasmid stability within populations [7-15], the propensity for this to occur across the parasitism-mutualism continuum is unknown. We experimentally evolved Pseudomonas fluorescens and its mercury resistance mega-plasmid, pQBR103 [16], across an environment-mediated parasitism-mutualism continuum. Compensatory evolution stabilized plasmids by rapidly ameliorating the cost of plasmid carriage in all environments. Genomic analysis revealed that, in both parasitic and mutualistic treatments, evolution repeatedly targeted the gacA/gacS bacterial two-component global regulatory system while leaving the plasmid sequence intact. Deletion of either gacA or gacS was sufficient to completely ameliorate the cost of plasmid carriage. Mutation of gacA/gacS downregulated the expression of similar to 17% of chromosomal and plasmid genes and appears to have relieved the translational demand imposed by the plasmid. Chromosomal capture of mercury resistance accompanied by plasmid loss occurred throughout the experiment but very rarely invaded to high frequency, suggesting that rapid compensatory evolution can limit this process. Compensatory evolution can explain the widespread occurrence of plasmids and allows bacteria to retain horizontally acquired plasmids even in environments where their accessory genes are not immediately useful.