A history of symbiosis impacts the host evolutionary trajectory in experimentally evolved amoebas
Jahan, I.; Larsen, T. J.; Strassmann, J. E.; Queller, D. C.
Show abstract
Biological diversity driven by endosymbiosis arises from the intertwined evolution of microbes and their hosts. Each partner affects the fitness and therefore the evolution of the other. Here, we tested a further question: does the history of symbiosis affect evolution even after the partnership is dissolved? We analyzed phenotypic data from experimentally evolved strains of Dictyostelium discoideum hosts, each of which had had its symbiont removed, to study how their traits evolved. We found that host trait evolution was affected by the prior history of infection, specifically by which of three Paraburkolderia bacterial symbionts had been removed. Thus, symbionts affect not only current evolution but also generate path dependence that affects the subsequent evolutionary trajectories even after the symbionts are lost. Impact statementThe evolution of partner dependence in host-microbial symbioses has fundamentally shaped biological diversity and ecosystem function. To examine variation in symbiont dependence in the social amoeba, we compared how different strains of Dictyostelium discoideum respond evolutionarily after the loss of their bacterial symbionts. We analyzed phenotypic data from experimentally evolved strains and found that the absence of different symbiont species leads to distinct changes in the subsequent evolution of key traits like cell proliferation, slug migration, and spore production. This research expands our current understanding of microbial symbiosis by revealing that symbiont species may impact the evolution of their hosts even after the symbiont is gone. Data summaryWe used phenotypic traits data from our previous experimental-evolution dataset from the open-access repository Dryad (https://doi.org/10.5061/dryad.kkwh70s97). Scripts for the statistical analyses are available in a GitHub repository (https://github.com/jahanisrat/SymbiontLoss). The accompanying R project includes code to reproduce the graphs in the results section.
Matching journals
The top 9 journals account for 50% of the predicted probability mass.